An Investigation of Environmental Factors that Affect the Behavior and Welfare of
Domestic Cats (Felis sylvestris catus)
DISSERTATION
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of
Philosophy in the Graduate School of The Ohio State University
By
Judith Lynn Stella, B.Sc.
Graduate Program in Veterinary Preventive Medicine
The Ohio State University
2013
Dissertation Committee:
Charles Buffington, Advisor
Candace Croney, Thesis Director
Linda Lord
Stephanie Lewis
ii
Copyrighted by
Judith Lynn Stella
2013
ii
Abstract
Domestic cats are the most commonly kept companion animal in the U.S. with large
populations of owned (86 million), free-roaming (70 million), research (13,000) and
shelter (2-3 million) cats. Therefore, large numbers of cats are confined to cages each
year, so offering the most appropriate housing environment could lead to improvements
to their welfare and possibly in the outcomes of biomedical research, shelter adoptions
and veterinary care for cats.
Most research on the welfare of confined cats has been aimed at modifications to
their cages (micro-environments) but a few have aimed to understand the effects of
aspects of the room (macro-environment), such as the effects of noise, lights, odors, and
predictability of the husbandry routine. One of societies concerns for species that are
confined is space allotment and recently this concern has been raised in relation to
confined cats. Typical cat housing in the U.S. is a small, single cage of less than 0.56
square meters (6 square feet) of floor space leading to an inability to express normal
species-typical behaviors. Although several groups have established recommendations
for cage size little research has been conducted in this area. Finally, an area of
research that remains relatively unexplored in the domestic cat is their cognitive abilities,
specifically memory, in regard to their responses to the cage environment. It is likely
that cats that are frequently confined and handled may respond negatively or positively
to this confinement and handling as a function of their memory of past experiences.
Study one aimed to evaluate the behavior of cats housed in enriched or
unenriched macro and micro environments. Study two aimed to evaluate the behavior
and welfare of cats housed in enriched or unenriched macro and micro environments
with 1.1 square meters (11.8 square feet) of floor space, greater than the typical space
allowance provided to caged cats in the U.S. Results indicate that cats respond
adversely to factors in the macro environment that they may have perceived as
threatening, and these factors were at least as relevant to them as were factors in the
micro environment. Additionally, provision of 1.1 square meters of floor space did not
iiv
change cat behavior, at least in the first 48 hours of confinement. It seems that the
quality of the environment may be more important than cage size. Study three aimed to
indirectly assess the cat’s long-term memory of a confinement housing experience
through comparisons of initial responses to a controlled confinement experience to
responses one year later. The results suggest that cats may form memories and that
those previously housed in an enriched macro environment habituated more quickly
upon re-exposure to that environment.
Future research is needed to study how other aspects of the environment, such as
factors pertaining to temperature, lighting and odors, and human-animal interactions
might affect cat behavior and welfare, the minimum space requirements for cats and
how length of confinement affects this parameter and direct assessment of the cognitive
abilities of cats, particularly their ability to form long-term memories of salient
experiences.
iii
v
Dedication
To all the cats of the OSU FIC research colony
RIP my friends
iv
v
Acknowledgements
I will be forever grateful to Dr. Tony Buffington for giving me an opportunity when
he hired me to be a research assistant and for encouraging me to enroll in graduate
school. I am indebted to Dr. Candace Croney, my thesis advisor, for all of her guidance
through the writing of this manuscript. Your contribution was invaluable. I would also
like to thank Dr. Linda Lord and Dr. Stephanie Lewis for serving on my committee.
I would like to express my thanks to Traci Shreyer for her assistance with
handling cats (and owners) on study days. Finally, I would like to express my gratitude
to all the faculty, staff and students who volunteered their cats for this project.
v
Vita
1993……………………B.S. Animal Science, Pennsylvania State University
1996……………………A.S. Zoo Animal Technology, Santa Fe CC
2009-2013 ……………Graduate Research Associate, Department of Veterinary
Preventive Medicine, The Ohio State University
Publications
Stella JL, Croney, CC, Buffington, CAT. Effects of Stressors on the Physiology and
Behavior of Domestic Cats. Journal of Applied Animal Behavior Science 2013; 143:157163.
Hague, DW, Stella, JL, Buffington, CAT. Effects of Feline Interstitial Cystitis on Acoustic
Startle Responses in Domestic Cats. American Journal of Veterinary Research 2013;
74:144-7
Stella JL, Lord LK, Buffington CAT. Sickness behaviors in response to unusual external
events in healthy cats and cats with feline interstitial cystitis. Journal of the American
Veterinary Medical Association 2011;238:67-73.
Rubio-Diaz D, Pozza ME, Dimitrakov JD, Gilleran JP, Giusti MM, Stella JL, RodriguezSaona LE, Buffington CAT. A candidate serum biomarker for bladder pain
syndrome/interstitial cystitis. Analyst 2009; 134:1133-7.
Pozza MA, Stella JL, Chappius-Gagnon AC, Wagner SO, Buffington CAT. PinchInduced Behavioral Inhibition (“Clipnosis”) in Domestic Cats. Journal of Feline Medicine
and Surgery. February 2008.
Stella JL and Buffington CAT. Individual and environmental effects on health and
welfare. In Bateson, P. and Turner, DC (eds) The Domestic Cat: The Biology of its
Behavior. (3rd edition) Cambridge:Cambridge University Press. Expected publication
2013.
viv
Fields of Study
Major Field: Veterinary Preventive Medicine
Specialization: Applied Ethology and Animal Welfare Science
vii
v
Table of Contents
Abstract………………………………………………………………………………………..…ii
Dedication…………………………………………………………………………………….....iv
Acknowledgements……………………………………………………………………………..v
Vita……………………………………………………………………………………………....vi
Publications……………………………………………………………………………………..vi
Fields of Study………………………………………………………………………………….vii
Table of Contents………………………………………………………………………………viii
List of Tables…………………………………………………………………………..………...x
List of Figures……………………………………………………………………………………xii
Chapter 1: A History of the Domestic Cat and Implications for its Welfare……………….1
Chapter 2: Improving Cat Welfare……………………………………………………………12
Chapter 3: Environmental Factors that Affect the Behavior and Welfare of Domestic Cats
(Felis sylvestris catus) Housed in Cages…………………………………………………….26
Chapter 4: The Behavior and Welfare of Domestic Cats (Felis sylvestris catus) Allotted
more Cage Space than the Norm for U.S. Housing Facilities…………………………….62
Chapter 5: Long Term Memory in the Domestic Cat (Felis sylvetris catus): Does Housing
Environment Affect Behavior during Re-Exposure………………………………………….91
Chapter 6: Conclusions and Applications………………………………………………….117
References…………………………………………………………………………………….123
Footnotes………………………………………………………………………………………135
Appendix A: Demographics………………………………………………………………….136
v
viii
Appendix B: Macro Environment Schedule………………………………………………..140
Appendix C: Cat and Client History Form………………………………………………….142
Appendix D: Approach Test………………………………………………………………….148
Appendix E: Cage Condition Study One Results………………………………………….150
Appendix F: Scan Sample Data Study One……………………………………………….156
Appendix G: Cage Condition Results Study Two…………………………………………159
Appendix H: Scan Sample Data Study Two……………………………………………….161
Appendix I: Cat and Client History Form Study Three……………………………………165
Appendix J: Cage Condition Study Three………………………………………………….169
Appendix K: Scan Sample Data Study Three……………………………………………..172
ix
v
List of Tables
Table 3.1 Cage factors ……………..…………………………………………………………31
Table 3.2 Ethogram………………………………………………………………………...32-33
Table 3.3 Results of generalized linear mixed model study one………………………….36
Table 3.4 Scan sampling of behavior study one-all treatments…………………………...41
Table 3.5 Scan sampling behavior study one- macro environment.……..……………… 42
Table 3.6 Scan sampling behavior study one- micro environment……………………….42
Table 3.7 Focal sample frequency data study one…………………………………………45
Table 3.8 Focal behavior duration study one………………………………………………..47
Table 3.9 Approach Test Study One…………………………………………………………50
Table 4.1 Results of generalized linear mixed model study two…………………………..70
Table 4.2 Scan sampling behavior study two- all treatments……………………………...74
Table 4.3 Scan sampling behavior study two- macro environment……………………….75
Table 4.4 Scan sampling behavior study two-micro environment………………………...75
Table 4.5 Focal sample frequency data- study two…………………………………………78
Table 4.6 Focal behavior duration- study two……………………………………………….80
Table 5.1 Results of generalized linear mixed model- study three………………………..98
Table 5.2 Scan sampling behavior study three- year 1 and 2……………………………101
Table 5.3 Scan sampling behavior study three- year 1…………………………………...102
Table 5.4 Scan sampling behavior study three-year 2……………………………………102
Table 5.5 Scan sampling behavior study three-M+ year 1 vs. 2…………………………102
Table 5.6 Scan sampling behavior study three- M- year 1 vs. 2…………………………102
Table 5.7 Focal sample frequency data-study three………………………………………108
Table 5.8 Focal behavior durations- study three…………………………………………..109
vx
Table A.1 Cat and housing information study one………………………………………...138
Table A.2 Cat and housing information study two…………………………………………138
Table A.3 Cat and housing information study three……………………………………….139
Table B.1 Daily schedule…………………………………………………………………….141
Table E.1 Simple effect comparisons, food intake study one……………………………151
Table E.2 Estimates, food intake study one……………………………………………….152
Table E.3 Simple effect comparisons macro environment, food intake study one…….152
Table E.4 No urine or urine out of box- study one………………………………………...153
Table E.5 No BM or BM out of box- study one…………………………………………….153
Table E.6 Simple effect comparisons, BM study one……………………………………..154
Table E.7 Simple effect comparisons macro environment, BM study one……………..155
Table E.8 Additional sickness behaviors- study one……………………………………...155
Table F.1 Position in cage- study one………………………………………………………157
Table F.2 Vocalizations- study one…………………………………………………………158
Table G.1 No urine or urine out of box- study two………………………………………...160
Table G.2 No BM or BM out of box- study two…………………………………………….160
Table G.3 Additional sickness behaviors- study two……………………………………...160
Table H.1 Position in cage- study two………………………………………………………163
Table H.2 Vocalizations- study two…………………………………………………………164
Table J.1 Simple effect comparisons, food intake study three…………………………..170
Table J.2 No urine or urine out of box- study three……………………………………….171
Table J.3 No BM or BM out of box- study three………………………………………….. 171
Table J.4 Additional sickness behaviors- study three…………………………………….171
Table K.1 Position in cage year 1- study three…………………………………………….173
Table K.2 Position in cage year 2- study three……………………………………………174
Table K.3 Vocalizations- study three……………………………………………………….175
vxi
List of Figures
Figure 3.1 Study room…………………………………………………….…………………..29
Figure 3.2 Enriched and unenriched cage………………………………………………….30
Figure 3.3 Cage setup unenriched cage…………………………………………………....30
Figure 3.4 Food intake- study one…………………………………………………………...39
Figure 3.5 Urinations - study one…………………………………………………………….39
Figure 3.6 Defecations- study one…………………………………………………………...40
Figure 3.7 Normal cage use- study one……………………………………………………..40
Figure 3.8 Change in cage use- study one………………………………………………….40
Figure 3.9 Sickness behaviors- study one…………………..………………………………41
Figure 3.10 Scan sampling behavior- study one……………………………………………42
Figure 3.11 Hide box use- study one……………………………….………………………..43
Figure 3.12 Perching- study one…………………...…………………………………………43
Figure 3.13 Vocalizations- study one………………………………………………………...44
Figure 3.14 Focal behavior frequency data- study one…………………………………….46
Figure 3.15 Focal behavior, alert duration- study one……………………………………...48
Figure 3.16 Focal behavior, freezing duration- study one………………………………….48
Figure 3.17 Focal behavior, hiding or attempting to hide duration- study one…………..48
Figure 3.18 Latency to interact, all groups- study one……………………………………..50
Figure 3.19 Latency to interact macro and micro environments- study one……………..50
Figure 3.20 Duration of interaction all groups- study one………………………………….51
Figure 3.21 Duration of interaction macro and micro environments- study one…………51
Figure 3.22 Mean approach score all groups- study one………………………………….51
xii
Figure 3.23 Mean approach score macro and micro environments- study one…………52
Figure 4.1 Food intake- study two…………………………………………………………….72
Figure 4.2 Urinations- study two………………………………………………………………72
Figure 4.3 Defecations- study two……………………………………………………………73
Figure 4.4 Normal cage use- study two……………………………………………………...73
Figure 4.5 Change in normal cage use- study two………………………………………….73
Figure 4.6 Sickness behaviors- study two…………………………………………………...76
Figure 4.7 Scan sample behaviors- study two………………………………………………79
Figure 4.8 Hide box use- study two………………………………………………………….76
Figure 4.9 Perching- study two……………………………………………………………….77
Figure 4.10 Vocalizations- study two…………………………………………………………77
Figure 4.11 Change in vocalizations- study two…………………………………………….77
Figure 4.12 Focal behavior alert duration- study two……………………………………….81
Figure 4.13 Focal behavior freezing duration - study two………………………………….81
Figure 4.14 Focal behavior grooming duration - study two………………………………..81
Figure 4.15 Focal behavior resting duration - study two…………………………………...82
Figure 4.16 Focal behavior hiding or attempting to hide duration - study two…………...82
Figure 4.17 Latency to interact all groups- study two………………………………………83
Figure 4.18 Latency to interact macro and micro environments- study two……………..83
Figure 4.19 Duration of interaction interact all groups- study two………………………..83
Figure 4.20 Duration of interaction macro and micro environments- study two…………84
Figure 4.21 Mean approach score all groups- study one………………………………….84
Figure 4.22 Mean approach score macro and micro environments- study one…………84
Figure 5.1 Food intake, urinations and defecations- study three………………………….99
Figure 5.2 Cage use- study three………………………………………………….………..100
Figure 5.3 Sickness behaviors- study three……………………………………………..…101
xiii
Figure 5.4 Scan sample behavior- study three…………………………………………….103
Figure 5.5 Hide box use- study three……………………………………………………….104
Figure 5.6 Change hide box use- study three……………………………………………..104
Figure 5.7 Perching- study three…………………………………………………………….105
Figure 5.8 Change perching behavior- study three……………………………………….105
Figure 5.9 Vocalizations- study three……………………………………………………….106
Figure 5.10 Change in vocalizations- study three…………………………………………106
Figure 5.11 Frequency focal behaviors- study three……………………………………...107
Figure 5.12 Duration of focal behaviors- study three……………………………………..109
Figure 5.13 Latency to interact step- study three………………………………………….110
Figure 5.14 Duration of interaction- study three…………………………………………...110
Figure 5.15 Mean approach score between years- study three…………………………111
Figure 5.16 Mean approach score within years- study three…………………………….111
Figure F.1 Position in cage- study one……………………………………………………..157
Figure H.1 Position in cage- study two……………………………………………………..162
xiv
Chapter 1
A History of the Domestic Cat and Implications for its Welfare
Domestication of the cat
The modern domestic cat (Felis silvestris catus) is the product of at least 12
thousand years of natural selection in a world increasingly dominated by humanity
(Driscoll et al., 2009). The domestic cat, as a subspecies of the wildcat (Felis silvestris),
evolved from, and still is, a solitary hunter of small prey and the prey of larger carnivores.
Although cats are increasingly kept as pets, often confined to the indoors in many parts
of the world, they have been described as “exploited captives,” (Clutton-Brock, 1999) not
yet truly domesticated.
The process of domestication involves both culture and biology; it has been defined
as the exploitation of a species by humans, most often for economic reasons. The
process starts when animals are incorporated into human social groups or communities
and become objects of ownership, inheritance and purchase, and ends when the
population’s breeding, organization of territory and food supply comes under permanent
human control - isolating it from the wild progenitor species (Clutton-Brock, 1992).
Therefore, a domestic animal is one whose mate choice is influenced by humans and
whose docility and tolerance of humans is genetically determined. Although controlled
breeding is critical to the domestication process, an animal bred in captivity is not
necessarily domesticated, and tame animals, such as Asian elephants, who appear to
be domesticated but do not have breeding controlled by humans, are not true
domesticates.
Wildcats are improbable candidates for domestication; they have specialized
diets (obligate carnivores), a solitary social system, and defend exclusive territories
making them more attached to places than people (Driscoll et al., 2007; Driscoll et al.,
2009). Additionally, their utility to humans is debatable; even as mousers, terriers and
ferrets out perform cats (Rogers, 1998). There is some evidence that early civilizations
1
sought out wildcats to tame as pets, but the more likely explanation for domestication is
that the cats exploited the anthropogenic environment and were tolerated by people,
which eventually led to divergence from the wild form. The story of cat domestication
likely proceeded in this manner: approximately 12,000 years ago agriculture originated in
the Fertile Crescent of the Near East. This became a new ecological niche for species
that could acclimate to living near humans. Rodents likely were the first commensal
species, and became a reliable food source for native wildcats. Wildcats which adapted
to this “urban” environment then became a human commensal themselves. This
explanation is supported by phylogenetic and phylogeographic evidence that the
speciation of the domestic cat from the wildcat occurred simultaneously along side
human civilizations (Driscoll et al., 2007). Therefore, unlike other domesticates, the cat
was likely produced through natural selection.
Understanding domestication is essential to the study of animal welfare. An
animal’s behavioral organization has been shaped by evolution to use information
obtained from the environment to react to an event or to interact with an environmental
feature to form rules of response for similar events or stimuli. The extent to which these
“decision rules” of the ancestral species become altered by domestication may influence
the negative subjective experiences (suffering) of an animal especially when there is a
mismatch between an animal’s current environment and the environment in which its
decision rules evolved (Inglis, 2000; Cameron-Beaumont et al., 2002). In this sense, we
may consider domestic cats as similar to zoo animals, with the proximity of conspecifics
and other animals, combined with limited resources and opportunities to express
species-typical behavior potentially influencing cats’ perceptions of control.
Cat welfare problems in modern society
In the 1980’s the cat became the most commonly kept companion animal in the
United States, with one third of U.S. households having a pet cat (Lockwood, 2005).
The most recent statistics from the American Society for the Prevention of Cruelty to
Animals (ASPCA) and the Humane Society of the United States (HSUS), estimate that
there are approximately 86 million owned cats and between 50-70 million feral and/or
free-roaming cats in the United States (ASPCA, 2012; HSUS, 2012). Two to three
million cats enter shelters each year, of which 70-75% are euthanized. A recent study
2
estimated that 10 to 25% of the companion animal population of the United States is
destroyed each year, making euthanasia the number one cause of death of companion
animals (Lepper et al., 2002; ASPCA, 2012). Yet despite their popularity, the cat still
has not reached the status of the dog as a companion animal. Attitudes towards them
are often ambivalent. A study by Kellert and Berry (1980) found that 17.4% of
Americans expressed some dislike of cats versus 2.6% who disliked dogs (Serpell,
2000; Lockwood, 2005). One researcher commenting on the differences in Americans’
attitudes towards cats and dogs said, “People who hate cats tend to be proud of the fact,
and brag about it as if it proved something honest and straightforward in their natures.
Nobody brags about hating dogs. To hate dogs would be mean-spirited and peculiarly
unpatriotic; dogs are a very American concept, fraternal, hearty and unpretentious, while
cats are inscrutable like the wily Oriental and elitist like the European esthete”
(Lockwood, 2005). The popularity of anti-cat books like A Hundred and One Uses for a
Dead Cat, I Hate Cats and The Cat Hater’s Handbook, which makes light of cat abuse,
attests to western cultural ambivalence about cats. How human perceptions and
attitudes towards the domestic cat may affect the quality of human-cat relationships as
well as cat welfare is an important consideration in modern society.
Cruelty and hoarding
The first anti-cruelty law to protect cats was the 1835 revision of the 1822 animal
welfare legislation of the United Kingdom. Historically, cases of animal cruelty have
been viewed as property crimes, but society increasingly views animal cruelty as a moral
or violent crime (Lockwood, 2005). In a recent analysis of reports from HSUS covering
the years 2001-2004, Lockwood (2005) reported that of 931 cases of neglect, 89.6%
involved dogs and 10.4% involved cats, and although significantly more cat than dog
cases involved malice (often a requirement for a criminal charge of animal cruelty), fewer
charges were filed. These differences may be a reflection of the societal view of cats as
self-sufficient and less likely to suffer if left abandoned and uncared for, and the
normalizing of hostility towards cats that may allow for underreporting and fewer
prosecutions of cruelty towards cats than toward dogs (Lockwood, 2005).
Further investigation by Lockwood (2005) based on prisoner interviews, found
that cats are a favorite animal victim of abuse because of their long tails, fur that burns,
3
easily broken bones and they are small enough to be carried and dropped from heights
(Lockwood, 2005). According to the authors, cat abuse has been correlated with future
violence, psychopathology and criminology. Others have also suggested cat abuse was
associated with sexually motivated murders of women as in the case of serial murderer
Keith Jesperson, who claimed to be responsible for more than 100 murders, many of
them prostitutes (Lockwood, 2005).
Another form of abuse is animal hoarding which, according to the Animal Legal
Defense Fund (http://www.aldf.org), is the number one crisis affecting companion
animals in the United States. It is estimated that up to 250,000 animals are involved in
hoarding cases each year. Cats are victimized most often, likely because they are
readily available in almost any community and they are easier to hide than dogs (Animal
Legal Defense Fund, 2012; Tufts, 2012). Additionally, the average number of animals
killed is nearly twice as high in cat hoarding cases as in dog hoarding cases (Lockwood,
2005). Hoarding is a welfare issue because of the obvious detrimental effects caused by
nutritional deficiency, lack of veterinary care and unsanitary conditions present, but just
as importantly because of the social stress caused by living at an unnaturally high
population density, unnatural social groups, and the lack of socialization to humans that
ultimately results in the death of many seized animals.
Feral and free-roaming cats
Another cat welfare issue is that of feral and free-roaming populations, estimated
in the United States to be between 50 (HSUS, 2012) and 70 (ASPCA, 2012) million.
Cats have a reputation for resilience, surviving traumas or accidents that might kill a
human or dog, and because of this and their independent nature it has been suggested
that it is easier for people to throw them out or abandon them (Lockwood, 2005). It is
difficult to accurately count the number of feral cats because: (1) there is disagreement
about the definition of a feral cat, (2) little is known about their average lifespan, (3)
distribution, (4) rates of migration in and out of the pet population, and (5) reproductive
rates in differing climates, seasons, and environments (e.g., urban, suburban, rural,
agricultural area, forest, etc.).
As mentioned earlier, among domesticated species cats maintain the unique
characteristic of owned populations interbreeding freely with feral populations rather than
4
having mating strictly controlled by humans. The most common explanation for the
maintenance of feral populations has historically been too few generations for complete
domestication. Alternatively, cats have until recently been maintained as both a pet and
for rodent control, obviating the need for selection pressure to produce two distinct
populations. Recently a third explanation has been proposed; the cat as an obligate
carnivore, a nutritional characteristic unique among domestic mammals, may have
maintained through selection, both hunting and scavenging ability to meet its nutritional
needs. Until recently humans did not have the knowledge or ability to meet the
nutritional requirements of the cat, so they therefore maintained the ability to feed
themselves (Bradshaw et al., 1999). This behavioral and ecological flexibility has
allowed the domestic cat to co-exist with humans in relationships ranging from a fully
dependent pet, to a commensal, to complete independence with movement between
categories within a few generations. The problem of feral cats is undeniably large and
often emotional; the two biggest concerns related to presence of feral cats are the risk of
zoonoses and predation of wildlife, particularly songbirds.
The incidence of zoonotic disease transmission from cats to humans is relatively
unknown because of the diseases of interest, only plague and rabies are nationally
reportable to the Centers for Disease Control and Prevention (CDC). Plague is a
concern in the southwest United States where it is endemic in wild rodents that are
preyed upon by cats. The clinical appearance of plague in cats’ mimics feline leukemia
virus (FeLV), putting veterinary staff at increased risk (Patronek, 1998).
In the United States, rabies is diagnosed in cats more than any other domestic
animal, with most infected cats residing in states where an epizootic of raccoon rabies is
present (CDC, 2012). Yet worldwide, 99% of human rabies is acquired from dogs
(WHO, 2012). Since 1980, no case of rabies transmission from a cat to human has
been reported in the United States. It is unknown how Trap-Neuter-Return (TNR)
programs influence rabies prevalence. Vaccination prior to release could decrease the
risk of rabies by creating an additional boundary between humans, owned pets and
wildlife, or it could increase the potential transmission by encouraging cats to
congregate, the vaccinated as well as unvaccinated (Patronek, 1998).
The other zoonotic disease of potential impact to humans is toxoplasmosis. High
sero-prevalence in free-roaming cats has been documented, probably due to high
contact rates with rodents. Alfonso et al (2007) reported positive antibodies to T. gondii
5
in 51% of the cats tested. Higher infection rates were found in males than in females,
and depended on the infection rate and type of prey the cats were ingesting. There is
little data on incidence of transmission to humans except in a few isolated cases of
outbreaks that can be traced to specific places.
To put the risk of zoonotic disease transmission from cats into perspective, 74
cases of brucellosis, 23,568 cases of Lyme disease, 1,340 cases of malaria, 44,863
cases of salmonellosis, 464 cases of neuroinvasive West Nile virus, and 11,817 cases of
varicella (chickenpox) were reported to the Centers for Disease Control in 2011. Only
two cases each of plague and rabies were reported during this time, and none were
acquired from a cat (CDC, 2012).
When considering the impact of cats on the prey populations, popular dogma
holds that the domestic cat is detrimental to wildlife because it is not an indigenous
species and wildlife has not evolved strategies to avoid predation. Although true of
many island eco-systems, fauna of the northern continents of Europe, Asia and North
America as well as Africa have evolved with the relatives of the domestic cat (i.e., Felis
sylvestris, Lynx lynx, and others) and are well adapted to the ecology of this predatorprey relationship. Although it has been repeatedly reported that cats are responsible for
one billion bird deaths per year there is no strong research to support this assertion and
in fact, the population level effects of cat predation on bird populations in continental
environments are poorly documented and have not been demonstrated (Patronek, 1998;
Fitzgerald, 2000). Studies tend to focus on predation rates rather than the impact of this
predation (Goldstein, 2011). Cats are opportunistic hunters, so scaling up of individual
predation estimates from small samples to national population-wide estimates is
problematic. The abundance of prey species across habitats, climates, and seasons as
well as the availability of provisioned food and scavenging opportunities means cat
populations will vary in their dietary profiles both temporally and spatially. Ultimately, the
problem, if in fact one exists, may more likely be caused by the high numbers of feral
and free-roaming cats, leading to hyper-predation.
The debate about predation by cats is an interesting one in light of the history of
domestication of the species. The presumed reason cats were domesticated in the first
place was because of their status as a commensal species tolerated by humans in
return for protection of grain stores. In modern times, they are often vilified for the exact
same behavior that in the past favored their relationship with humans.
6
Cats in confinement
In 1866 the American Society for the Prevention of Cruelty to Animals (ASPCA),
the oldest humane organization in the United States, was founded in New York City
through the efforts of Henry Burgh to end the cruelty inflicted on cart horses (ASPCA,
2012). In the period after World War II companion animals were living longer because of
advances in veterinary medicine and the advent of pet food production. This led to
increases in fecundity and unwanted animals and an increased awareness of the
problem ultimately resulting in the modern animal sheltering “business”. This has arisen
in an unorganized, haphazard manner and presently there still is no single agency
available to oversee the number of animals entering or exiting shelters. Consequently,
there are no accurate statistics making planning, developing programs and/or evaluating
the current system virtually impossible (Zawistowski et al., 1998).
Recently, using records from the ASPCA, Zawistowski et al. (1998) looked for
trends in the number of companion animals relinquished to shelters from 1895-1994. It
showed cats consistently out numbered dogs both in intake and euthanasia. Further
inspection of the data also revealed a precipitous drop in the number of animals entering
shelters in New York City after World War II as well as the number of animals’
euthanized per 10,000 people particularly after implementation of a mandatory neuter
policy of all adoptable animals in 1972. Even so, 74% of dogs and 78.8% of cats that
enter a shelter in the United States are euthanized (Zawistowski et al., 1998).
Relinquishment
In 1994, The National Council on Pet Population Study and Policy (NCPPSP) was
established “to gather and analyze reliable data that further characterize the number,
origin, and disposition of pets (dogs and cats) in the United States…” (NCPPSP, 2012).
Through a series of studies by this group and others, the reasons for relinquishment and
return of cats as well as some characteristics of the cat, owner and household that are
risk factors have been identified. The most common reasons for relinquishment were
abandonment/stray, owner circumstances (move, illness, divorce, financial), unwanted
kittens, and allergies (Patronek, 1996; Casey, 2009). A study by Salmon (2000) which
analyzed data from 1,286 feline surrenders in which owners were asked to provide
reasons for relinquishment, found behavioral reasons to be the second most common
7
cause for cat relinquishment. At least one behavioral reason was given for 28% of cat
relinquishments with the most common ones including house soiling (43.2%), problems
between pets (18.9%), aggression toward people (14.6%), unfriendly (5.4%), fearful
(3.8%) and destructive behavior (12.4%). A significant association of the presence of
another pet in the household and the addition of a dog in the past year was found, so
that it appears that cats living in a single animal household have reduced risk of
relinquishment for behavioral reasons (Patronek, 1996; Casey, 2009).
A further study by New et al. (2000) aimed to define the characteristics of
relinquished animals and their owners in comparison to those of pet-owning households.
A knowledge check found that significantly fewer people relinquishing cats knew that
cats pounce, scratch, or bite as a form of play, and that the number of cats in the home
affects cat behavior. The relinquishers also exhibited significant knowledge deficits
regarding the female estrous cycle. Undesirable behaviors that were significant risks for
relinquishment included house soiling, destructive behavior, and being overly active.
Additional studies have found that cat factors strongly associated with a risk of
relinquishment include younger age, mixed breed and being sexually intact (with cost
being stated as a main reason for a lack of sterilization). Post-acquisition risk factors
that increased the likelihood of relinquishment included confinement to the basement or
garage most of the day, not having access to the outdoors, being primarily cared for by
an adult female, and having behavioral problems (Patronek, 1996). Owner related
factors associated with an increased risk of relinquishment included owner attachment
and expectations, not having owned another cat as an adult, and having expectations of
a particular role for the cat to fill (Patronek, 1996).
To better understand characteristics associated with poor outcomes, Kass et al.
(2001) gathered data on the kinds of pets people relinquish for euthanasia compared to
those relinquished for adoption. Of the 282 cats presented for euthanasia in this study,
50 (18%) were specifically for behavioral reasons. House soiling was the most important
single behavioral reason for euthanasia (38%) but other reasons included aggression
toward people (28%) or animals (8%), biting people or other animals (18%), destructive
behavior (14%), and fear of or unfriendliness toward owner (12%). The owners
confirmed that 83% of these cats had no illness, so behavior alone was reason for the
euthanasia request (Kass et al., 2001).
8
Finally, Casey (2009) reported that 20% of adopters (dogs and cats) return the
animal in the first year, 4% within two weeks of adoption and 8% within six months.
Additionally, the most common reason for no longer having a pet one year post-adoption
was death, which was higher in cats (36%) than dogs (15%). A behavioral problem was
the reason given for 20% of the cats returned. Female cats were more likely to be
returned than males (55% and 45% respectively) and adult cats of four months to seven
years were more likely to be returned (52%) than kittens (38%) and cats older than
seven years (10%). It has been reported that pet behaviors are interpreted differently by
retainers and returners (“she’ll outgrow it” vs. “so we got rid of it”). This highlights the
need to educate adopters to modify their expectations of the adoption and adjustment
period, perhaps encouraging or increasing the number of adopters that seek help when
problems arise shortly after adoption.
Cats confined in homes
Interestingly, the welfare of cats in homes is not usually addressed to the extent it
is within a research colony or a shelter, although similar problems are likely to arise in all
environments that confine cats. Only 3% of the cat population in the United States is
purebred with the remaining 97% being self-bred which may imply that most “problem”
behaviors are not abnormal but rather natural behaviors that need to be redirected to
appropriate substrates (Jongman, 2007). For example, a study of cat behavior in the
home by Morgan and Houpt (1989) found the most common behavior problems to be
scratching furniture (60%), eating houseplants (42%), conspecific aggression (36%),
food stealing (25%), hissing/aggression to people (17%), house-soiling (16%), excessive
vocalizations (16%), fabric chewing (7%), and “shy” (4%). Heidenberger (1997) reported
the most frequent behavior problems cited by cat owners to include anxiety (16.7%),
scratching furniture (15.2%), feeding problems (10.9%), aggression (10.5%),
inappropriate urination and spraying (8.2%), and defecation in the house (5.1%). Many
of these behaviors have been observed in response to inappropriate environments
under laboratory conditions as well (Stella et al., 2011). While the proportion of cats
showing inter-cat aggression and house-soiling in the home seems to be the same as
the proportion relinquished for these behaviors, cats appear to be returned to shelters for
behavior reasons more often than they are relinquished for them (38% return vs. 16%
relinquished) (Patronek, 1996; Casey, 2009).
9
In a study that used owner self-reports to further explore the living conditions of
indoor housed cats and the problems associated with their housing and behavior,
Heidelberger (1997) discovered that 24% of cats did not have their own food bowl, 51%
had to share the litter pan with other cats, on average cats rested 7.5 hours at night and
7.6 hours during the day, and only 19% of cats were reported to play - averaging three
play bouts per day. Cats in groups of two or three had more problems than did single
cats. Outdoor access was negatively associated with problem behavior, so cats that
could go outside had virtually no problems. Anxiety was the most frequent behavior
complaint reported (16.7%); another 37.9% of cats exhibited clearly anxious behavior on
occasion. The most frequently mentioned stimulus was a visit by strangers, and the
most common reaction of the cat was to run away, hide, crouch and lower its tail.
Neutered females, cats adopted between the ages of 5-12 months and cats weighing
greater than 4 kg were more likely to be anxious. Cats acquired from a shelter, as a
stray, or from a friend were all more likely to show anxiety more often than those born in
the owner’s home. The number of cats in the home and the amount of available space
per cat also were relevant, with multiple cat households and limited space per cat both
leading to homes more likely to have a cat described as anxious. Additionally, studies
have found that people who interact often and regularly with their cat throughout the day
report fewer problem behaviors, indicating that the quality of the human-animal
relationship may be a factor as well (Heidenberger, 1997).
Lastly, a study of 336 cats referred to a behavior clinic in Spain found that
problem behaviors presented for treatment were similar to those cited as problems in the
home or as reasons for relinquishment, but with the following distribution: aggression
(47%), inappropriate elimination (39%), compulsive behavior (3.5%), excessive
vocalization (2.5%), fear and phobias (2.5%), and “other”, which was comprised of
anorexia, scratching furniture, and over-activity (5.4%) (Amat et al., 2009). One possible
reason for the different distribution may be that some behaviors, particularly aggression
and inappropriate elimination, although common are not tolerated by owners. In
agreement with other studies, inter-cat aggression was found to be a common problem.
This may be a species typical behavioral response for a solitary species that finds it is
unable to disperse and lacks a well developed intra-specific communication system, as
proposed by Leyhausen (1973).
10
From these studies it is clear that the owner is the most important animal welfare
factor in pet-keeping because, especially in indoor-housed cats, he/she determines all
the living conditions. Apparent behavioral problems of “pets” may be normal behaviors
that are unwanted by the owner. Alternatively, the exhibition of “problem” behaviors
could be a response to a poor environment, one in which the cat cannot cope. Turner
(2000) has summarized it like this: “Many behavioral problems result from a lack of
consideration of the needs of the cat, poor or changing housing conditions, unrealistic
expectations of the owner or inadequate interactions between the owner and the cat”.
Ultimately, inadequate housing and handling diminishes welfare for the pet.
Two consistent findings in all these studies are that having outdoor access, even
if limited, decreases the risk of behavior problems and relinquishment. Interestingly,
neutered females are typically overrepresented for risk of behavioral problems, including
anxiety, and for return or relinquishment. Further research is needed to determine why
this is the case.
11
Chapter 2
Improving Cat Welfare
Welfare concepts relative to domestic cats
Animal welfare has been defined as “… how an animal is coping with the conditions
in which it lives. An animal is in a good state of welfare if (as indicated by scientific
evidence) it is healthy, comfortable, well nourished, safe, able to express innate
behavior, and it is not suffering from unpleasant states such as pain, fear, and distress.
Good animal welfare requires disease prevention and appropriate veterinary treatment,
shelter, management and nutrition, humane handling and humane slaughter or killing.”
(OIE, 2012). Welfare is a characteristic of the individual, existing on a continuum that
varies from good to poor and will change across the days, seasons, reproductive states,
and life stages of an animal. Assessment of cat welfare (as is the case for all animals)
involves evaluation of three categories; physical health and biological functioning,
affective states, and natural living and expression of species typical behaviors (Broom,
1988). All areas can and should be investigated to gain a broad perspective and to
ensure optimal welfare. Thus, the welfare of each individual should be monitored daily
and appropriate modifications made as needed. Using the criteria outlined above, an
animal can be placed into one of three categories; well-being, fair-being, or ill-being.
Well-being implies the animal is coping well with the environment and only minor coping
responses are required. Fair-being means the animal is experiencing stress and coping
with difficulty. This state requires the animal to engage moderate coping responses. An
animal classified in a state of ill-being is stressed, failing to cope and needs help. This
requires major coping responses and the animal is in a vulnerable state (Curtis, 1985).
No individual needs to have absolute freedom from aversive stimuli to have good
welfare, but the individual needs to be able to cope (Broom, 1991; Morgan and
Tromborg, 2007). In the lives of captive animals, the perception or actual lack of ability
to control their surroundings is perhaps the greatest stressor they experience. Most
animals in captivity have little or no control over who their social partners are, how much
12
space they can put between themselves and others, the type, amount or availability of
food, nor the quality or quantity of environmental stimuli including lights, noise, odors
and temperature (Morgan and Tromborg, 2007). Predictability or the lack there-of, is
another aspect of the captive environment that may be stressful. Studies have shown
that, when allowed to do so, animals will choose predictability over unpredictability,
especially in regard to aversive events (Weiss, 1971; 1972; Morgan and Tromborg,
2007). Predictability refers both to temporal aspects as well as familiarity of caretakers
and the environment. A consistent, predictable daily routine is essential, particularly
when an animal is confined. Daily cleaning and feeding procedures conducted at the
same time of day and performed by a familiar person allow cats to predict potentially
aversive events. A variety of animals from macaques to rats show physiological and
behavioral responses to cage-cleaning associated with a stress response (Morgan and
Tromborg, 2007). For example, in a study of laboratory housed cats, changing the time
of daily husbandry and feeding resulted in increased sickness behaviors, also indicative
of a stress response (Stella et al., 2011; 2013).
Stress and disease in cats
Chronic exposure to unpredictable and uncontrollable environments may have
adverse health consequences for cats. External environmental events that activate the
hypothalamic stress response system may be termed stressors (McEwen, 2008). A
stressor is defined as a stimulus that disrupts homeostasis. Stressors can be classified
as (1) physical with either a positive or negative psychological aspect such as cold, heat
or noise; (2) psychological such as a learned response to a previously experienced
adverse condition; (3) social which includes disturbed interactions among individuals
such as territorial disputes; and (4) challenges to cardiovascular and metabolic
homeostasis such as exercise, hemorrhage or heat. Stressors can be further
categorized as either acute (a single, intermittent, time-limited exposure) or as chronic (a
prolonged and intermittent or continuous exposure). In order to reestablish
homeostasis, the adaptive response of the individual, that includes both physiological
and behavioral components, will be activated. The goals of this coping response include
reduction of the harmful environmental conditions and enhancement of the prospects of
recovery, tolerance or adjustment to the negative events, maintenance of emotional
equilibrium, and preservation of social relationships (Pacák and Palkovits, 2001).
13
Inadequate perception of control and predictability also can activate the stress
response system in animals due to interference with attempts to cope with their
environments. Lack of control can lead to symptoms of chronic stress including
anorexia, weight loss, inhibition of exploratory behaviors, learned helplessness,
stereotypies, and aberrant immune responses (Bassett and Buchanan-Smith, 2007).
However, absolute control is not necessary, and in fact predictable but dynamic routines
may lead to optimal welfare (Broom, 1991; Broom, 1996). Adaptation occurs genetically
at the population level over generations and phenotypically in the individual either by
physiological acclimatization or by learned behavioral adjustment. An animal may never
be perfectly acclimated to its environment and may develop behavioral strategies to
reach optimality and maintain homeostasis. A threatened homeostasis is the result of
the difference between the actual and expected state, and the responses of the animal
to minimize this difference can be interpreted as its attempt to cope (Broom, 1991; van
der Harst, 2009).
Evidence suggests that many of the differences between domestic and wild
populations result from quantitative changes in the thresholds for performing a behavior
rather than in qualitative changes in the behavior itself (Tarjei, 1989). When animals, at
the level of the species, are able to maintain homeostasis through employment of
behavioral and/or physiological responses, they are said to have acclimated.
Environmental stimuli can be so intense, prolonged, noxious, or novel that they exceed
the tolerable range of the animal (Broom and Johnson, 2000). Under such
circumstances, the individual’s attempts to cope with the environment are inadequate,
which may lead to injury, poor health, suffering, and reduced life expectancy. For cats,
such stressors include loud or unfamiliar noises, sudden movements, novel and
unfamiliar places and objects, and the approach of strangers (humans, cats, or other
animals) into their personal space. As in other species, breed and individual differences
in temperament or stress susceptibility (Feaver et al., 1986; Adamec, 1991), as well as
individual variation in experience (Gottlieb and Halpern, 2002), also influence responses
to the environment (Boissy, 1995; Adamec et al., 1998).
Overall, therefore, welfare can be regarded as influenced by both positive and
negative experiences and will be good or poor depending on factors including the
availability of resources and the length of exposure to and intensity of the stressor, that
allow the individual to cope with its environment. Impaired welfare can be considered as
14
a chronic imbalance between positive and negative experiences resulting in chronic
stress and failure to cope. It is now assumed that, similar to humans, chronic stress may
induce mental suffering in animals with or without associated physical health problems.
Environmental enrichment and its role in animal welfare
Environmental enrichment, as defined by David Shepherson “is a concept which
describes how the environments of captive animals can be changed for the benefit of the
inhabitants. Behavioral opportunities that may arise or increase as a result of
environmental enrichment can be appropriately described as behavioral enrichment”
(Young, 2003). An alternate definition from Valerie Hale is “a process for improving zoo
animal environments and care within the context of their inhabitants’ behavioral biology
and natural history. It is a dynamic process in which changes to structures and
husbandry practices are made with the goal of increasing behavioral choices to animals
and drawing out their species appropriate behaviors and abilities, thus enhancing animal
welfare” (Young, 2003). The goals of environmental enrichment are to (1) increase
behavioral diversity, (2) reduce the frequencies of abnormal behavior, (3) increase the
range or number of normal behavior patterns, (4) increase positive utilization of the
environment, and (5) increase the ability to cope with challenges in a more ‘normal’ way
(Young, 2003).
There are many aspects of the captive environment that may affect the welfare of
an animal. Individuals must acclimate to their physical environment, and captive
environments do not always match the physical environment to which the species was
evolutionarily acclimated. Various aspects of the environment are of particular
importance to captive cats including the macro and micro environments, human-animal
interactions, and the social environment.
The macro environment is the housing space (room, building, or barn) and its
surroundings, and contains factors that include the thermoregulatory environment,
lighting, odors, and sounds (Morgan and Tromborg, 2007). These environmental factors
will affect animals differently depending on whether they are housed in cages or areas
where they are unable to control or move away from a factor they find aversive. The
thermoregulatory environment exerts a major influence on animal welfare. Captive
animals are often unable to express temperature regulating behaviors because of a lack
of resources available to do so. It has been suggested that after many generations of
15
being raised in climate controlled facilities some species lose the ability to thermoregulate properly. Often, the thermo-neutral zone of the species is not considered in
animal housing; instead the emphasis is often placed on the comfort of the caretakers.
Cats prefer ambient temperatures that are much warmer than many species. The
thermoneutral zone for domestic cats is 30-38oC (NRC, 2006), yet most cat housing
areas are not this warm, and most homes and laboratory housing for cats are
maintained closer to 22 ±2o C (NRC, 1996).
Another macro environmental factor is lighting. The intensity (lux), type provided
(e.g., fluorescent, natural, incandescent, or LED), as well as the light:dark cycle all can
affect animals. These factors are of particular importance for reproduction. Aspects
such as flicker frequencies of fluorescent bulbs also should be considered in housing
facilities, because these can affect animal welfare (Morgan and Tromborg, 2007).
Because almost all mammals depend more on olfactory cues (macro-osmotic) than
humans do, aversive odors can be an additional source of chronic stress for confined
animals. For cats, potentially objectionable odors include dogs (a natural predator),
unfamiliar conspecifics, alcohol (from hand rubs), cigarettes, cleaning chemicals
(including laundry detergent), some perfumes, and citrus scents.
A final macro environmental factor to consider is sound frequency range and
intensity. The auditory frequency range of cats (and most species) exceeds that of
humans, making assessment of the welfare implications of high frequency noise difficult.
Sound pressure levels (intensity) in nature range from 20-37 dB in savannah habitats to
27-40 dB in the rain forest (Morgan and Tromborg, 2007), whereas they regularly
exceed 100 dB in shelters and laboratories during routine husbandry (Coppola et al.,
2006). Furthermore, sound pressure levels of 73 dB from the banging of metal cages
have been found to activate the stress response system of rats, leading to a 100-200%
increase in plasma corticosterone levels (Baldwin et al., 2007). Therefore, maintaining
sound pressure levels around 60 dB (quiet conversational level) may be beneficial to
animal welfare.
Another group of factors to consider when assessing the captive environment
include those found in the micro environment, which is the individual cage, pen or stall
housing the animal. Relevant factors of the micro environment include enclosure size,
food (type and presentation), elimination facilities, and outlets for the expression of
species typical behaviors, which include hiding and perching opportunities in cats. The
16
type and presentation as well as the availability of these features of the environment can
be either a source of stress or enrichment.
An important micro environmental factor that may affect animal welfare is the
quantity and quality of space provided to the animal. In the wild, home range size may
vary on a seasonal or annual basis, and is affected by food and water distribution and
availability, frequency and intensity of social interactions, and quantity and quality of
shelter. Captive animal spaces are generally reduced in both quantity and quality in
comparison to options available to their wild or free-roaming counterparts. Therefore,
the captive environment should be behaviorally relevant, with the quantity and quality of
space provided allowing for the development and normal expression of species-typical
behavioral patterns.
Another important aspect of the micro environment is the availability, type and
presentation of food. In captivity, food is most often provided in the form of a formulated,
uniform and consistent diet, placed in a single location so that time and energy related to
foraging behavior is greatly reduced. Cats are opportunistic hunters of small prey that
typically eat small meals throughout the day, so provision of one or two meals of
commercial dry cat food may lead to under- or over-feeding from boredom when
confined.
The preferred shelter of a species is another important determinant of welfare
that may be understood by studying the shelter seeking behaviors of the wild
counterparts. In captivity, the type of shelter offered may permit partial isolation from
conspecifics and humans, which may be of critical importance to some individuals. For
cats, access to a hide box serves such a purpose. Additionally, it has been shown that
cats seem to prefer to monitor their surroundings from elevated vantage points, and
seem to welcome provision of climbing frames, hammocks, platforms, raised walkways,
shelves or window seats (Rochlitz, 2000). Cats are both a predator and prey species, so
climbing and hiding appear to be important species typical behaviors. Other important
feline specific behaviors include scratching and marking, which maintains claw health
and leaves both visual and pheromonal territorial marks. Thus, appealing, appropriate
objects need to be provided to confined cats as outlets to express these behaviors.
Another important aspect of the captive environment is human-animal
interactions. In the wild, the fitness of some species is affected either directly by
hunting, fishing, or trapping and/or indirectly by habitat destruction or addition of toxins to
17
the environment by humans. In captivity, acclimation to human presence is an important
fitness-determining factor since humans select for tameness and cull less tamed or
behaviorally acceptable individuals or don’t allow them to reproduce.
A human-animal relationship can be said to exist if a number of repeated
interactions between the animal and human occur, eventually allowing each to make
predictions about the others’ behavior. Both positive and negative human-animal
relationships are important in the context of animal welfare. In human-animal
relationships, the human mostly determines the number and nature of interactions and
hence the relationship; the animals more often react to the humans’ actions rather than
initiate them. When put into the context of environmental enrichment, human contact
can meet the criteria that traditional environmental enrichment aims to meet; the
identification and provision of the appropriate stimuli necessary for the physiological and
psychological well-being of captive animals (Shepherdson, 1998).
Research in livestock has found that interactions between stockpeople and the
animals in their care can influence both the productivity and welfare of livestock and the
quality of life of the stockpeople (Coleman et al., 1998; Hemsworth, 2007). A number of
studies have investigated the relationships between the attitudes and behaviors of
stockpeople and the productivity and welfare of livestock; the most important factors
determining stockperson behavior towards the animals in their care appear to be
personality and attitude (Coleman et al., 1998).
Despite generations of selective breeding, one of the potentially most frightening
events that many animals are likely to experience is exposure to humans. The
predominant reaction is fear, and it has been proposed that this is because animals
often perceive encounters with humans as predatory (Waiblinger et al., 2006). Fearful
responses of animals lead to more negative attitudes of caretakers therefore increasing
the likelihood of poor interactions recurring. However, an animal’s fear of people can be
reduced even if it had previously been treated poorly, leading to improved behavior, if
the human interactions are positive, such as gentle handling or feeding of preferred food
items (Claxton, 2011). As with all other aspects of confinement, control and
predictability of caretaker behavior are of great importance to how the animal perceives
humans. Hemsworth (1987) showed that intermittent unpredictable negative handling of
pigs (in a ratio of one negative for every six positive exposures) led to pigs becoming as
18
fearful and chronically stressed when approached by humans as pigs that were
consistently handled negatively (Hemsworth et al., 1987).
Likewise, in effective cat management, a familiar person appears to be essential.
Wild felids are considered sensitive to the captive environment, which can result in large
numbers of abnormal behaviors and decreased reproductive success. Studies have
shown that this can be ameliorated by improved keeper-cat relationships. Mellen (1991)
found a positive correlation between the quality of keeper interactions, and increased
reproductive success in small captive felids. Wielebnowski et al. (2002) found a
negative correlation between fecal cortisol concentrations and the amount of time the
primary keeper spent with clouded leopards, and a positive correlation between fecal
cortisol and the number of keepers. The interpretation of these results was that a higher
number of keepers prevented the animals from forming and maintaining a predictable
relationship with any keeper, thus increasing the stress of captivity. Additionally, a study
of domestic chickens showed that they easily learned to discriminate between two
people, one that offered a food reward and one that did not (Davis and Taylor, 2001). If
animals are able to recognize and discriminate among the humans they regularly come
in contact with, then these same persons can become predictors of salient events (food,
pain) in the animals’ lives. As a result, consistent, positive human-animal interactions
may facilitate improved welfare.
Finally, the social environment is of great importance to confined animals. One
of the greatest differences between wild and captive environments is the reduction in
choices available in captivity. The selection of social partners, for mating or other
reasons is largely determined by the individual in the wild, and escape or avoidance of
aggressive conspecifics is limited only by natural barriers. This is drastically altered in
captivity. Increased population density, unnatural social groups (both sex and age
groups are more uniform then would occur in nature), and frequent regrouping all can
potentially cause social stress (Price, 1984). The social behavior of domestic cats
exhibits great plasticity. It appears to be influenced by ontogeny so that kittens
socialized to other cats, humans, dogs, etc. during the sensitive period of socialization
are likely to acclimate to life in social groups more readily than are kittens raised by their
mothers alone (Mendl, 2000). This social plasticity appears to be distributed across the
family Felidae as illustrated by a study of 16 species of small felidae from five lineages
which found that the expression of affiliative behavior toward humans was widely
19
distributed across the small felidae, not concentrated in the domestic cat lineage
(Cameron-Beaumont et al., 2002).
Ultimately, the environmental enrichment needs of the cat will be similar whether
it is confined to a home or a cage in a shelter, research facility, veterinary hospital, or
boarding facility. Aspects of the abiotic environment that can be perceived as potential
threats or aversive stimuli whether parts of the macro or micro environment, humananimal interactions, the social environment or the predictability and control of the
environment all interact to influence a cat’s well-being.
Welfare of cats in cages
Each year millions of cats are housed in cages in veterinary hospitals, shelters,
and research laboratories, so understanding aspects of the cage environment that
facilitate or prohibit the ability of cats’ to cope will potentially impact the welfare of large
numbers of individuals. Novelty, confinement and the inability to express species typical
behaviors may result in cats experiencing distress (Broom and Johnson, 2000), and their
responses to it may include decreased appetite, withdrawal from social groupings,
increases in salivary, blood and fecal cortisol, increases in urinary cortisol:creatinine
ratio, decreases in grooming and an increase in the frequency and intensity of attempts
to hide (Carlstead et al., 1993; Stella et al, 2011; 2013). Medical interventions (e.g.,
vaccinations, treatment for parasites and neutering) while potentially beneficial to the
cat’s physical health, can introduce additional stressors and thus impact the
psychological health of the cat.
Cats evolved in environments where hiding was an adaptive response to threat
of predation so it is likely that pet cats also display this behavior in threatening
environments, such as veterinary hospitals and shelters. Because thwarting attempts to
hide can contribute disproportionately to any overall measure of stress (Overall and
Dyer, 2005), one of the initial environmental enrichments often suggested to help cats to
cope with confinement has been provision of hiding and perching opportunities. For
example, McCune (1994) and Rochlitz (2000) have shown that the ability to hide may be
essential to cats when exposed to stressors. Hiding behavior, which is correlated with
enhanced ACTH response and increased urinary cortisol concentrations, has been
identified as an indicator of stress (Carlstead et al., 1993). These studies suggest that
not allowing cats the opportunity to hide may adversely affect their welfare.
20
In shelters, the idea that cats that hide decrease their adoptability often overrides
this welfare concern. One study (Kry and Casey, 2007) aimed to determine if adding a
hide box improved the cats’ ability to cope with the stressful environment, allowing the
cat to become more comfortable, extroverted and interactive with unfamiliar people.
They reported that cats provided Hide, Perch & Go boxes approached more often and
retreated less than did control cats (those with no box), and were more often seen
sleeping restfully than controls. In addition, vigilance behavior has been associated with
anxiety-related behavior problems in house cats, and the control cats in this study
showed this alert type behavior. Cats in the enriched group were observed in or on their
hide box 77% of the time, whereas control cats attempted to hide 36% of the time.
There was no difference in time to adoption between the groups, disproving the rationale
for not providing cats with this enrichment. Importantly, cats appeared to be coping,
indicated by lower Cat Stress Scores, by day three, whereas control cats exhibited
behaviors indicative of a change to chronic stress by the end of the two week study
period. The Cat Stress Score is a tool that is often used to assess stress in cats that
describes seven possible stress levels based on body posture and behaviors (see
Kessler and Turner, 1997 for details). Despite its frequent use, it has been proposed
that what is really being measured is fear as evidenced by the three highest scores
being labeled fearful, very fearful and terrorized. Additionally, it assumes that we have a
reliable and accurate way to measure stress in cats, which, at this time, we do not (see
McMillian, Franklin, Letter to the Editor, JAVMA, April 15, 2012). Therefore, this caveat
should be considered when interpreting the results of studies that have employed this
method.
A similar study (Gourkow and Fraser, 2006) examined the responses of cats in
four different treatments; single housing with usual care, single housing with enrichment,
communal housing with usual care, and communal housing with enrichment. Results
indicated that Cat Stress Scores were similar in all groups on day one, but thereafter
cats in single housing with usual care had higher Cat Stress Scores than all other
groups. They also had the lowest adoption rates, the longest length of time waiting for
adoption, and exhibited more fearful behavior than did cats in the other groups. In this
study both housing and handling were manipulated, so either could have produced the
effect seen.
21
In addition, Ottway and Hawkins (2003) aimed to test the hypothesis that cats in
long-term shelter care housed in groups of unfamiliar conspecifics have diminished
welfare (higher Cat Stress Scores) due to the unstable and inappropriate social
grouping. A comparison of 12 adult cats unfamiliar with each other, communally housed
in a large run and cats that were either singly housed or pair housed with a familiar
conspecific (former housemate) was conducted. The results indicated that the Cat
Stress Score was higher in cats housed communally than in cats housed in single units
or with previously familiar conspecifics. Communally housed cats spent more time
hiding than single housed cats (26% versus 4%). Play behavior was only observed in
1% of the observation periods, and exclusively in single housed cats or in cats housed
with familiar conspecifics. The authors concluded that cats housed communally
experienced moderately higher levels of stress than cats housed in discrete units and
they had more difficulty coping, probably due to the instability of the group with unstable
groups being more stressful than group living itself. Similarly, de Monte and Pape
(1997) concluded that for adult cats, single housing may not be considered a “totally
unfortunate housing situation”, especially if the cats have daily positive interactions with
humans.
In studies of small exotic felids, the domestic cat has been used as a model to
identify and address welfare problems. These wild species are notoriously intractable
and easily stressed in captivity resulting in reproductive failure, a major obstacle in the
maintenance of endangered species. Carlstead et al. (1993) imposed a 21-day
psychological stressor on singly housed cats that included unpredictable caretaking and
mildly aversive handling, a chronic psychological stressor for confined cats. Stressed
cats exhibited decreased activity levels and increased attempts to hide compared to
controls. They also had an increase in adrenocortical output (increased urine cortisol
concentrations), enhanced adrenal sensitivity to ACTH and reduced p pituitary sensitivity
to luteinizing hormone-releasing hormone. The researchers concluded that the
environment had resulted in activation of the stress response in the cats, and that hiding
was an important behavior for modulating HPA axis activation caused by an
unpredictable environment (Carlstead et al., 1993).
A recent study by McCobb et al. (2005) evaluated stress levels among cats in
usual and enriched housing via behavioral assessment (Cat Stress Score) and
monitoring of urine cortisol:creatinine ratios in four different shelters. Results indicated
22
that stress levels among the cats were highest in the morning and decreased throughout
the day. A slight negative correlation between the number of days spent in the shelter
and the Cat Stress Score was found with the Cat Stress Score decreasing with
increasing time spent in the shelter. In agreement, the mean morning Cat Stress Score
of the cats in the holding areas was higher than that of the cats in the adoption area. No
difference was found between the Cat Stress Score of owner surrender and strays.
Additionally, 24% of the cats had signs of systemic disease including upper respiratory
signs, vomiting and diarrhea. No significant relationship was found between the noise
level at the shelter and Cat Stress Scores, but cats that were housed where they could
see, hear and/or smell dogs did have higher urine cortisol:creatinine ratios. This study
found indicators of distress among the cats including almost 25% having signs of
systemic illness and more then 25% of the urine samples collected had trace amounts of
hematuria. The authors concluded the biggest factor affecting the cats’ stress level in
the different types of shelters appeared to be the extent of their exposure to dogs. Cats
in areas with more exposure to dogs had higher Cat Stress Scores than did cats in other
high noise areas, which appeared to have an additive effect in that it increased stress
levels more in cats who were obviously ill than for cats that had no signs of disease.
Cats housed in enriched environments had lower stress levels than those housed in
traditional shelters (McCobb et al., 2005).
Moreover, Rochlitz et al. (1998) assessed the quarantine experience of seven
cats over six months. The cats in this study needed two to five weeks to acclimate to the
quarantine situation. The authors concluded that hiding was an important behavior
expressed by cats confronted with an aversive situation, such as a novel environment.
The postures and facial expressions of stress-induced sleep (feigned sleep) are different
from those of restful sleep, and this defensive sleep is more often seen in cats that can’t
hide. The withdrawal of friendly human contact was particularly stressful for cats used to
receiving a lot of attention and may be important in shelter environments as well; more
so for owner surrender cats than for strays (Rochlitz et al., 1998). Dybdell et al. (2007)
subsequently investigated this in a study designed to assess the social history of the
cats admitted to the shelter. The Cat Stress Score was used to assess stress, and the
observers were blinded to which group (owner surrender or stray) the cat belonged.
Cats were scored for the first three days of housing while in the holding area. No effect
of gender or neuter status was found. However, cats surrendered by their owners had
23
higher Cat Stress Scores than did stray cats. Overall, cats that were deemed suitable
for adoption had lower Cat Stress Scores than did cats that were deemed unsuitable and
subsequently euthanized. Moreover, cats in the owner surrender group became ill
significantly sooner than cats in the stray group did. In agreement with the Rochlitz’s
(1998) findings, this study showed that all cats experienced a stress response
associated with entry to the shelter, but the owner surrender cats had an additional
psychosocial stressor of forced social separation from their primary caretaker and home
environment. Alternatively, owner surrender cats may come from an unfavorable
environment that led to behavior problems and relinquishment, and may already be
more distressed than strays at the time of admit.
Finally, Kessler and Turner (1997) assessed cat acclimation to boarding over two
weeks and compared the boarding cats’ Cat Stress Scores to those of control cats living
in a shelter. They evaluated single, pair and group housing situations. The results
indicated that two thirds of the cats acclimated, one third found boarding distressful, and
4% never acclimated so boarding was deemed inappropriate for that group. The daily
Cat Stress Scores of the singly housed cats declined significantly from day one to day
five, and overall stress levels continued to decrease during the two weeks of boarding.
However, in agreement with the findings of earlier studies they never reached the level
of the control cats. This is an important finding since cats in shelters may not have time
to acclimate before being re-homed. In fact, most failed adoptions and returns take
place within two weeks of adoption. The period of greatest risk for cats appears to fall
within the time they are acclimating to the new environment, indicating that current
protocols may not be sufficient to allow cats to fully adjust to the new environment and
thus impact cat welfare.
Confinement of cats, in cages or homes, may lead to poor welfare through
inadequate environments that do not meet the needs of cats, including factors related to
the macro and micro environment, the human-cat interactions and the social
environment. Poor welfare may be reflected in poor physical health, illness and disease
or behavioral problems such as house soiling, fearful and aggressive behaviors. These
factors may lead to a breakdown in the human-animal bond and ultimately to
abandonment, relinquishment to a shelter, or euthanasia, and thus, require further
investigation. Given this, the objectives of this study were (1) to examine the effects of
the macro and micro environments on cat behavior, (2) to evaluate the behavior and
24
welfare of cats housed in cages with increased space allowances, and (3) to determine if
the domestic cat has long term memory for a confinement housing experience and if the
quality of the environment affects the behavior of the cat upon re-exposure to the
environment.
25
Chapter 3
Environmental Factors that Affect the Behavior and Welfare of
Domestic Cats (Felis sylvestris catus) Housed in Cages
1. Introduction
Domestic cats are the most commonly kept companion animal in the United
States, estimated by the American Society for the Prevention of Cruelty to Animals
(ASPCA) to number 86 million; approximately one third of households in the United
States keep a pet cat (ASPCA, 2012). Presumably, many of these cats will be
hospitalized by a veterinarian or be housed in a boarding facility during their lifetime.
Additionally, an estimated 70 million unowned, feral or free-roaming cats also populate
the United States, many of which will be attended to by one of the 1,600 Trap Neuter
and Return (TNR) programs throughout the country (ASPCA, 2012; HSUS, 2012).
Further, an additional 13,000 cats are used in biomedical research (USDA, 2012) and 23 million cats enter American shelters annually (ASPCA, 2012). Therefore, large
numbers of cats are confined to cages each year, so offering the most appropriate
housing environment could lead to improvements to their welfare and possibly in the
outcomes of biomedical research, shelter adoptions and veterinary care for cats.
Most research on the welfare of confined cats has been aimed at modifications to
their cages, particularly the addition of hiding and perching opportunities (Rochlitz, 2000;
Gourkow and Fraser, 2006; Kry and Casey, 2007). A few studies have aimed to
understand the effects of other aspects of the environment, such as the effects of noise,
lights, odors, caretaker interactions, and predictability of the husbandry routine
(Carlstead et al., 1993b; Stella et al., 2011). Consequently, for the purposes of this
study, environmental factors that could be perceived as stressful by confined cats were
placed into four categories. These include the macro and micro environments,
predictability and control of the environment and human-animal interactions. The macro
environment can be thought of as the room the cage is in. Room factors that could
26
affect cats include the ambient temperature, lighting both the type (fluorescent, natural,
incandescent, LED) as well as the light:dark (L:D) cycle, odors including cleaning
chemicals, unfamiliar conspecifics, predators (dogs), perfume, laundry detergents,
cigarette smoke, alcohol hand rubs, and intensity and type of noise (barking dogs,
growling/hissing cats, loud music, loud talking). The cat’s micro environment pertains to
aspects of the cage. These include the flooring substrate (slatted or solid, stainless steel
or some other material), hiding and perching opportunities, food presentation (for
example, are wet and dry in the same bowl), and features of the elimination area
including the type and size of litter pan as well as the depth, type and texture of the litter.
In addition to room and cage factors, several other aspects of the cat’s environment
potentially impact well-being. For example, studies by Weiss (Weiss, 1971; 1972) and
others have shown that events are more aversive to animals when they are
unpredictable and/or uncontrollable, so the degree of predictability and control the
animal has over the environment also needs to be considered in confinement housing
facilities. Finally, studies of livestock (Hemsworth et al., 1987; 1989; 1996) and exotic
cats in zoos (Wielebnowski et al., 2002) have shown that the quality of the humananimal interaction also affects the welfare of confined animals.
The aim of this study was to evaluate the behavior of cats housed in enriched or
unenriched macro and micro environments. The hypothesis was that cats housed in the
enriched environments would be less distressed (as evidenced by their behavior) than
cats in the unenriched environments.
2. Materials & Methods
2.1 Subjects: Adult cats between 0.9-13 years of age (mean age 4.1 years)
were recruited from The Ohio State University faculty, staff and students of the College
of Veterinary Medicine. Seventy six neutered cats, 41 male and 35 female, completed
the study. All cats were healthy and current on viral rhinotracheitis, calicivirus,
panleukopenia, and rabies vaccines at the time of participation. Informed consent was
collected from all owners who volunteered a cat for the study. Cats were randomly
assigned to one of four treatment groups that consisted of combinations of an enriched
(M+) or unenriched (M-) macro (room) environment and an enriched (m+) or unenriched
(m-) micro (cage) environment (Appendix A, Table A.1). Participants were admitted to
27
the study between 17:00-20:00 hours on day 0 and placed in an individual stainless steel
cage. The cage door was covered by a towel that had been sprayed with Feliwaya, a
commercially available synthetic cat pheromone, and the cat was left to acclimate
overnight. Behavioral observations were collected for eight hours a day from 08:0016:00 on days 1 and 2, after which cats were released to their owners’ between 17:0020:00 hours on day 2. The Animal Care and Use Committee of The Ohio State
University and the Clinical Research Advisory Committee of the Veterinary Medical
Center approved all experimental procedures used in this study.
2.2 Macro environment: Cats were housed in the Ohio State University
Veterinary Medical Center (OSUVMC) vivarium. A 14:10 light: dark schedule was
maintained to mimic length of day in Ohio at that time of year (May/June). A mean ± SD
room temperature of 22 ± 1.6°C (72 ± 4°F) was maintained throughout the vivarium.
The room housing the cats had dimensions of 4.57 meters by 4.88 meters and had
cages along three walls (Figure 3.1) with a small attached ante room used as a storage
and office area. Twenty cats were randomized to each replicate with enriched and
unenriched cages balanced between upper and lower cages. The enriched room
(treatments M+m+ & M+m-) was a managed environment with minimal disturbances
from people, barking dogs or other unpredictable noises and events. The husbandry
routine was temporally consistent, with cats being provided care at the same time and in
the same order each day to provide some predictability about the environment. Cats in
the unenriched room (treatments M-m+ & M-m-) environment were confronted with
multiple, random, unpredictable disturbances each day that included loud conversations,
talk radio, recordings of barking dogs, and noise from cage doors and equipment being
moved or dropped (Appendix B, table B.1). Routine husbandry occurred at different
times of day, and cats were cared for in a random order each day. All cats were cared
for by a single researcher to ensure consistent handling. Cages were spot cleaned so
as to minimize disruption to the cat and to ensure consistency. The daily husbandry and
test procedure schedule is outlined in Appendix B.
28
Figure 3.1 Study room
2.3 Micro environment: Cats were housed in individual stainless steel cages
measuring 70 × 78 × 75 cm (0.55 square meters). The front half of the slatted cage floor
was covered by a matb. The enriched cage (treatments M+m+ & M-m+) contained a two
tiered cardboard Hide, Perch & Goc box that had a lower hiding area (53 x 30 x 22 cm)
with two access openings and an open upper sitting area (53 x 30 x 9 cm), placed in the
left rear corner of the cage. Bedding (84 x 74 cm cage pad folded into quarters) was
provided in both the ‘hide’ and ‘perch’ areas. A plastic litter pan (32 x 22 x 8 cm) was
placed in the right rear cage corner filled with Sani Chipsd litter at a depth of 3 cm. One
cup of dry commercial cat foode, 1.5 oz of commercial canned cat foodf, and water was
provided in separate 0.6L (20 oz) stainless steel bowls (figure 3.2). Unenriched cages
(treatments M+m- & M-m-) contained bedding (84 x 74 cm cage pad folded into
quarters) and a plastic litter pan (32 x 22 x 8 cm) with Sani Chipsd litter at a depth of 1.5
cm; no Hide, Perch & Goc box was provided. Food (one cup of dry commercial cat foode
and 1.5 oz of commercial canned cat foodf) was provided in the same 0.6L (20 oz)
stainless steel bowl. Water was provided in a separate bowl (figure 3.2), and cage items
were rearranged daily (figure 3.3).
29
Figure 3.2 Enriched (right) and unenriched (left) cage.
Rear of cage
Food/water Bedding
Rear of cage
Bedding
Litter pan
Litter Pan
Food/water
Front of cage
Front of cage
Day 0
Day 1
Rear of cage
Litter pan
Bedding
food
water
Front of cage
Day 2
Figure 3.3 Cage setup unenriched cage
2.4 Data Collection: All participating owners were asked to complete a prestudy questionnaire for each cat consisting of information about demographics, health
history and home environment (Appendix C). Prior to routine husbandry each day, one
researcher stood in front of each cage for 30-60 seconds recording food intake,
urination, defecation, cage use, and additional sickness behaviors (vomiting, diarrhea,
eliminating out of the litter pan) for each cat (Table 3.1). During husbandry anything that
could not be seen from outside the cage was recorded (e.g., cached food, vomit, or
eliminations out of the litter pan).
30
Behavior
Food Intake
Cage Condition
Outcomes
>½
<½
No Use
Description
Ate more than 50% of the offered food.
Ate less than 50% of the offered food.
Nothing has been touched.
Used normally
The cage is slightly messy, some litter out of
the pan, some food on the floor of the cage
Everything has been moved or destroyed;
bowls and/or litter pan are dumped over,
bedding is moved, etc.
Urine is deposited in the litter pan.
Disrupted
Urination
Defecation
In litter pan
Out of litter pan
Urine is deposited out of the litter pan.
No Urine
In litter pan
Cat did not urinate.
Feces are deposited in the litter pan.
Out of litter pan
Other Sickness
Behaviors
Feces are deposited out of the litter pan.
No BM
Lower GI
symptoms
Upper GI
symptoms
Cat did not defecate.
Feces are soft or diarrhea.
Vomiting (forceful expulsion of the contents of
its stomach through its mouth) of food, hair,
bile, or foreign materials.
Table 3.1 Cage Factors
Behavioral observations were collected between 08:00-16:00 hours using two
sampling techniques. A scan sample was collected every two hours that included the
cat’s position in the cage, the type of behavior(s) it was exhibiting and vocalizations
based on an ethogram for cats in cages developed based on observations of cats in
cages in shelters, veterinary hospitals and research laboratories (Table 3.2). The
observer stood quietly in the middle of the housing room and recorded these
parameters. Duration of observation time for each scan sample was approximately
three minutes.
On the alternate hours, a five minute continuous focal sample of the observed
behaviors using the same ethogram for cats in cages (Table 3.2) was video recorded for
later coding. Two cats were recorded simultaneously for five minutes with a total of ten
replicates per observation hour. Video cameras were placed on tri-pods, one recording
a cat housed in an upper cage and one recording a cat housed in a lower cage, while
the researcher left the housing room to minimize observer effects on the cats’ behavior.
31
After the last scan sample collection on day two, all cat cage doors were covered
and a three- step stranger approach test was conducted for all cats starting
approximately 30 minutes later (Appendix D). Cats were tested in a randomized order.
Each cat cage was uncovered for the test immediately prior to commencement of step
one. One male unfamiliar to the cats served as the stranger throughout the study. Data
were recorded live as well as video recorded for further analysis. During step one, the
stranger stood quietly one meter from the cage for 30 seconds. Next, the stranger took
a step closer, placed his hand on the cage door and stood quietly for 30 seconds. In the
last step, the stranger opened the cage door and stood quietly with his hand extended
toward the cat in the cage for 30 seconds. The cage door was re-covered immediately
after completion of step three. Latency to interact, duration of interaction, and a
sociability score (1-5) were recorded for each cat at each step.
Category
Affiliative
Maintenance
Table 3.4. Ethogram behaviors
Behavior
Approach: The cat comes toward the observer.
Rub: The cat rubs its body along the ground or object (observer). This
behavior can be subdivided according to the part of the body used:
Head- cheek, forehead, ears, lips, chin; Body- neck, flank; Tail
Eye contact: The cat looks directly at the observer
Tail Up: A cat raises its tail to a vertical position.
Play: Object- The cat manipulates an object with its paws in an
apparently playful manner. The cat may pat, throw, pounce or wrestle
with the object. Self- The cat plays with its own body, usually the tail.
Social- The cat directs play at another cat or the observer.
Other: Roll- The cat turns over on the ground. Knead- The cat presses
and stretches its paws on a surface, alternating feet.
Eat: The cat consumes food.
Drink: The cat laps water or other liquid.
Grooming: The cat grooms itself by licking its body or by licking its paw
and passing the paw over its head.
Rest: The cat remains generally inactive with eyes closed but
occasionally opens them to scan the area; ears flicking regularly.
Stretch: The cat extends itself or its limbs to full length.
Yawn: To open the mouth wide with a deep inhalation, usually
involuntarily from drowsiness, fatigue, or boredom.
Scratch: Object-The cat repeatedly scrapes its extended claws against
a rough surface. Self- The cat scratches itself as if it has an itch.
Climb: The cat ascends an object.
Table 3.2. Ethogram
continued
32
Table 3.2 continued
Agonisitic
Avoidant
Vigilant
Vocalizations
Position in Cage
Lunge: The cat makes a sudden movement forward
Crouch: The cat is positioned with its ventrum and legs in contact with
the ground, the paws are folded.
Stare: The cat gazes fixedly at another cat or human and is not easily
distracted. It is often directed at another’s eyes.
Lip lick: The cat licks its lips briefly. This is a non-appetitive behavior.
Freeze: The cat doesn’t move from its position.
Other: Attempt to escape- The cat tries to get out of the cage; may
scratch at the door, wall, floor, with or without vocalizing. Flattened
body- This is an extreme version of crouch. The cat is often behind
something, has flattened ears, and is hiding its head and/or averting its
gaze.
Turned away: The cat bends its neck and its head away from the
observer without turning its body.
Attempt to hide: Part of or the entire cat is behind or under something
in the cage.
Avert gaze: The cat avoids looking at another cat or the observer for an
extended period, but it may keep it in its peripheral vision.
Startle: The cat starts or jumps involuntarily, as by surprise or alarm.
Dilated pupils: The pupils are fully dilated.
Erect ears: The ears are pointed upward.
Alert: The cat remains generally inactive with eyes fully open and flicks
ears occasionally as it scans its surroundings.
Tense: The cat is rigid; muscles are tensed ready to react.
Increased respiratory rate: The cat is breathing fast or at an increased
rate with or without being active.
None: The cat makes no sound.
Meow: A cat makes a distinct sound, usually when it is trying to obtain
something from another cat but it is often directed toward human
caretakers.
Growl: A cat makes a low-pitched rumbling noise.
Hiss: A cat makes a drawn-out SSSS sound, which is unvoiced.
Front: Cat is in the front half of the cage.
Rear: The cat is in the rear half of the cage.
In litter pan: The cat is positioned in the litter pan but is not urinating or
defecating.
Hide: The cat is positioned in a hiding box (lower part of Hide Perch and
Go).
Perch: The cat is positioned on an elevated structure (upper part of
Hide Perch and Go).
2.5 Statistical Analysis
Cage condition: Food intake, urination, and defecation outcomes were analyzed
by generalized linear mixed models. The models included fixed effects of treatment,
day, the interaction of treatment and day, and subject as a random effect (modeled as a
random intercept). Simple effect comparisons of treatment by day using least squares
33
means was conducted on these parameters when indicated. Analyses were performed
in SAS 9.2g using the GLIMMIX procedure (Stroup, 2011). A second analysis was
performed to assess the effect of the macro environment. Treatment (M+) grouped
M+m+ and M+m- and treatment (M-) grouped M-m+ and M-m-. In this analysis M+
contained 36 cats and M– contained 40 cats. Finally, a third analysis was done to
assess the effect of the micro environment. Treatment (m+) grouped M+m+ and M-m+
and treatment (m–) grouped M+m- and M-m-. In this analysis, m+ and m- each
contained 38 cats.
Comparison of the cats’ cage use in all four treatment groups was analyzed
using Pearson’s chi square tests on the frequency of the three outcomes on day one and
day two using STATA 11i. This analysis was repeated to assess the effect of macro and
micro environments grouping treatments as described above. Finally, a Wilcoxon sign
rank test of each treatment group comparing the average number of cats that exhibited
normal cage use on day one to those that did so on day two was performed using
GraphPad Prism 5h to assess the change in the number of cats exhibiting normal cage
use.
For comparison with earlier work (Stella et al., 2011; 2013), decreased appetite,
no eliminations for 24 hours, eliminating out of the litter pan, upper gastrointestinal (UGI)
and lower gastrointestinal (LGI) signs were dichotomized as either present or absent and
summed as total sickness behavior. Wilcoxon sign rank test of each treatment group
comparing day one to day two was performed using GraphPad Prism 5h.
Scan sample behavior data: Scan sampled behaviors were grouped for
analysis based on their association with affect or subjective emotional states.
Maintenance and affiliative behaviors comprised one group, reflecting positive affect,
whereas agonistic, vigilance and avoidant behaviors comprised the second group,
reflecting negative affect or distress. Data were dichotomized as either present or
absent and Pearson’s chi square tests were performed on the frequency data at each
sampling point using STATA 11i. Statistical significance was set at the level of p=0.006
to account for multiple comparisons (p=0.05/9 sample points or comparisons; p=0.006).
This analysis was repeated to assess the effect of macro and micro environments
grouping treatments as described above.
Analysis of the scan samples of position in cage was complicated by the
presence of structural zeros in the data set, meaning the cats in treatments M+m- and
34
M-m- only had three choices for position in cage due to absence of hiding and perching
opportunities. Therefore, only hiding and perching behavior in the enriched cage groups
was further analyzed. Pearson’s chi square tests on the number of cats in the hide box
at each scan sample point were performed comparing treatment M+m+ to treatment Mm+ using STATA 11i. Additionally, an average score was calculated for each cat on day
one and day two by calculating the average number of scan samples the cat was
observed in the hide box. Wilcoxon sign rank tests of treatment M+m+ and treatment Mm+ comparing day one scores to day two scores and comparing the two treatments on
day one and on day two were conducted. This analysis was repeated with the perching
data.
Analysis of vocalizations was complicated due to small frequency counts in each
cell. A summary of the number and percentage of cats vocalizing at each time point is
presented. Pearson’s chi square tests were performed on the vocalization data at each
sampling point using STATA 11i. Statistical significance was at the level of p=0.006 to
correct for multiple comparisons. Additionally, Pearson’s chi square tests of the number
of cats meowing or hissing/growling at each scan sample point were performed
comparing the macro environments (M+ to M-). An average score was calculated for
each cat on day one and day two by calculating the average number of samples the cat
was observed meowing and hissing/growling. Wilcoxon sign rank tests of M+
(treatments M+m+ & M=m-) and M- (treatments M-m+ & M-m-) comparing day one
scores to day two scores were conducted, as were comparisons of the two treatments
on day one and on day two.
Focal sample behavior data: Data were collected on the frequency and/or
duration of the behaviors of the ethogram. Data was summarized and behaviors that
were not exhibited in at least 5% of the samples were removed from further analysis.
The remaining frequency behaviors were approach, rub, tail-up, alert, eat/drink, groom,
rest, stretch, yawn, lip lick, turn away, and startle. These behaviors were dichotomized
as either present or absent and analyzed with a Pearson’s chi square test at each
sampling point using STATA 11i. Statistical significance was set at the level of p=0.006
to account for multiple comparisons. The remaining duration behaviors were tail-up,
alert, groom, rest, crouch, freeze, turn away, hide or attempt to hide, tense, and
increased respiratory rate. The behaviors recorded as tense and increased respiratory
rate were removed from further analysis due to the inability to quantitatively assess
35
these parameters. The remaining behaviors were analyzed using two way repeated
measures ANOVA in GraphPad Prism 5h.
Stranger approach data: Latency to interact, duration of interaction and the
mean approach score at each of the three steps of the approach test were analyzed
separately by one way ANOVA and unpaired t-tests with GraphPad Prism 5h.
3. Results
3.1 Cage condition: Results of the generalized linear mixed model are
presented in table 3.3.
Food Intake
All treatments
Macro Environment
Micro Environment
Effect
Num df
Dem df
F-value
p-value
Tx
3
72
2.65
0.055
Day
1
72
10.78
0.002
Tx*Day
3
72
1.55
0.2
Tx
1
74
8.11
0.006
Day
1
74
10.73
0.002
Tx*Day
1
74
0.77
0.4
Tx
1
74
0.02
0.9
Day
1
74
11.03
0.0014
Tx*Day
1
74
3.91
0.052
Tx
3
72
0.96
0.7
Day
1
72
35.23
<0.0001
Tx*Day
3
72
0.11
0.8
Tx
1
74
0.10
0.8
Day
1
74
35.62
<0.0001
Tx*Day
1
74
0.17
0.8
Tx
1
74
1.74
0.4
Day
1
74
35.77
<0.0001
Tx*Day
1
74
0.08
0.5
Tx
3
72
0.06
0.97
Day
1
72
16.40
0.0002
Tx*Day
3
72
1.60
0.2
Tx
1
74
0.12
0.8
Day
1
74
15.12
0.0002
Tx*Day
1
74
5.25
0.02
Tx
1
74
0.40
0.5
Day
1
74
17.90
<0.0001
Tx*Day
1
74
0.72
0.3
Urination
All treatments
Macro Environment
Micro Environment
Bowel Movement
All treatments
Macro Environment
Micro Environment
Table 3.3 Results of generalized linear mixed model
36
3.1.1 Food Intake: The percentage of cats in all treatment groups that ate more
than 50% of the offered food is depicted in Figure 3.4. On day one, 20% or fewer of the
cats in each treatment group ate more than half of the offered food. On day two, there
was an increase in the number of cats in all treatment groups that ate more than 50% of
the offered food. The effect of treatment approached significance (p= 0.055) and the
effect of day was statistically significant (p= 0.002). Further analysis revealed that on
day two, a comparison of treatments M+m- to M-m+ was statistically significant
(p=0.006) as was the comparison of treatment M+m- to M-m- (p=0.043) (Appendix E,
Table E.1). Additionally, the comparison of treatments M+m- and M-m+ on day one
versus day two was marginally significant (p=0.052), while the comparison between
treatments M+m- and M-m- at each day found no significant effects (p=0.51) (Appendix
E, Table E.2).
When considering the macro environment, a significant effect of both treatment
(p= 0.006) and day (p= 0.002) was found (figure 3.4). Further, a comparison of M+ and
M- was statistically significant (p=0.005) on day two (Appendix E, Table E.3). No other
significant results were identified.
When considering the micro environment, no significant effect of treatment (p=
0.89) was found, but the effect of day was significant (p= 0.001) (figure 3.4). Further
analysis did not identify any other significant results.
3.1.2 Urination: The number of cats in each treatment that did not urinate
during the study period and those that urinated out of the litter pan is shown in Appendix
E, table E.4. The number of cats that urinated out of the litter pan was too few to
analyze. Cats that did not urinate were compared to those that eliminated in the litter
pan. The percentage of cats that urinated in the litter pan is depicted in Figure 3.5. The
effect of treatment was not significant (p=0.7) but the effect of day was significant
(p<0.0001) meaning that more cats in all treatment groups urinated in the litter pan on
day two than on day one. Further comparisons did not identify any other significant
results.
Analysis of the macro environment showed the effect of treatment was not
significant (p=0.8) but the effect of day was (p<0.0001) (figure 3.5). Further
comparisons did not identify any other significant results.
37
Analysis of the micro environment showed that the effect of treatment was not
significant (p=0.4) but that day was (p<0.0001) (figure 3.5). Further comparisons did not
identify any other significant results.
3.1.3 Bowel Movement (BM): The number of cats in each treatment that did not
defecate during the study period and those that defecated out of the litter pan is shown
in Appendix E, table E.5. The number of cats that had a BM out of the litter pan was too
few to analyze. Those that did not have a BM were compared to those that eliminated in
the litter pan. The percentage of cats that had a BM in the litter pan is depicted Figure
3.6. The effect of treatment was not significant (p=0.97), but the effect of day was
(p=0.0002) meaning more cats in all treatment groups had a BM in the litter pan on day
two than on day one. Additionally, a comparison of treatments M+m+ and M-m+ was
significant on day two (p=0.04) (Appendix E, Table E.6) but a comparison of treatments
M+M+ and M-m+ at day one compared to day two revealed no significant effects.
Analysis of the macro environment showed that the effect of treatment was not
significant (p=0.8) but that day was (p=0.0002). Additionally, the interaction between
treatment and day (p=0.02) was significant (figure 3.6). Additional comparisons of
treatment M+ and M- identified a significant difference on day two (p=0.01) (Appendix E,
Table E.7). Further, the comparison of M+ and M- on day one compared to day two
was significant (p=0.02) (Appendix E, Table E.7).
Analysis of the micro environment also showed no significant effect of treatment
(p=0.5) but the effect of day was significant (p<0.0001) (figure 3.6). Further
comparisons did not identify any other significant results.
3.1.4 Cage Use: Analysis of how cats used the cage (none, normal, disrupted)
identified no statistically significant results for comparisons of all treatments, macro or
micro environments. Depicted in figure 3.7 is how the cats used their cage presented as
the percentage of cats exhibiting each outcome in all treatment groups as well as in the
macro and micro environments on day one and day two. The change in the number of
cats that used the cage normally from day one to day two in each treatment group was
statistically significant for all treatment groups, and the macro and micro environments
as well (figure 3.8).
3.1.5 Sickness Behavior: Only eight instances of additional sickness behavior
(vomiting and diarrhea) were recorded, which provided too few observations to analyze.
A summary is presented in Appendix E, table E.8. For comparison with earlier work,
38
decreased appetite, no eliminations for 24 hours, eliminating out of the litter pan, UGI
and LGI signs were combined. Statistically significant decreases in sickness behavior
from day one to day two were identified for treatment M+m+ (p=0.02) and treatment
M+m- (p=0.002) but not for cats in treatment M-m+ or M-m- (figure 3.9a and 3.9b).
Food Intake
Treatment p=0.055 Day p=0.002
100
Food Intake macro environment
Treatment p=0.006 Day p=0.002
M+m+ (n=17)
100
M+m- (n=19)
80
Food Intake micro environment
Treatment p=0.89 Day p=0.001
100
M+ (n=36)
M- (n=40)
M-m+ (n=21)
m+ (n=38)
m- (n=38)
80
80
40
20
% of cats
60
% of cats
% of cats
M-m- (n=19)
60
40
20
0
Day 2
40
20
0
Day 1
60
0
Day 1
Day 2
Day 1
Day 2
Figure 3.4 Food intake- Percentage of cats that ate more than half of the offered food;
from left to right all groups, macro and micro environments.
Urinated in the litter pan macro environment
Treatment p=0.8 Day p<0.0001
Urinated in the litter pan
Treatment p=0.7 Day p<0.0001
100
M+m+ (n=17)
80
% of cats
60
m+ (n=38)
m- (n=38)
80
M-m+ (n=21)
M-m- (n=19)
% of cats
M+ (n=36)
M- (n=40)
M+m- (n=19)
80
Urinated in the litter pan micro environment
Treatment p=0.4 Day p<0.0001
100
60
% of cats
100
40
40
60
40
20
20
20
0
Day 1
0
Day 1
Day 2
0
Day 1
Day 2
Day 2
Figure 3.5 Urinations- Percentage of cats that urinated in the litter pan; from left to right
all treatments, macro and micro environment
39
BM in the litter pan macro environment
Treatment p=0.8 Day p=0.0002
BM in the litter pan
Treatment p=0.97 Day p=0.0002
100
100
80
M+m- (n=19)
m+ (n=38)
m- (n=38)
80
% of cats
60
% of cats
M-m+ (n=21)
M-m- (n=19)
% of cats
100
M+ (n=36)
M- (n=40)
M+m+ (n=17)
80
BM in the litter pan micro environment
Treatment p=0.5 Day p<0.0001
60
40
60
40
40
20
20
20
0
0
Day 1
0
Day 1
Day 2
Day 2
Day 1
Day 2
Figure 3.6 Defecations- Percentage of cats who defecated in the litter pan; from left to
right all treatments, macro and micro environment
Cage use
Cage use
80
% of cats
% of cats
80
100
60
40
Cage use
None
Normal
Disupted
80
60
40
20
20
2
2
M-_D1
M-_D1
M-_D2
m+_D1
Treatment Group and Day
-_
D
2
m-_D1
m+_D2
m-_D2
Treatment Group and Day
M
-m
D
D
-_
+m
M
M
M
-m
+_
1
+_
+m
-m
M
M
D
2
1
-_
D
1
D
D
-m
+_
D
1
-_
+_
+m
M
+m
40
0
M+_D1
M
60
20
0
0
None
Normal
Disupted
100
% of cats
None
Normal
Disupted
100
Treatment Group and Day
Figure 3.7 Normal cage use- as a percentage of the cats that displayed normal use; from
left to right all treatments, macro and micro environment
Normal Cage Use day 1 vs day 2
1.5
p=0.02
p=0.0006
p=0.04
Normal Cage Use day 1 vs day 2
Macro Environment
p=0.02
Normal Cage Use day 1 vs day 2
Micro Environment
1.5
p=0.001
0.0
0.5
0.0
p<0.0001
1.0
0.5
0.0
2
-m
M
-m
M
Treatment Group and Day
1.0
-_
D
1
-_
D
2
M
-m
+_
D
1
M
-m
+_
D
2
1
D
M
+m
-_
D
-_
+m
M
+_
D
2
+m
M
+m
+_
D
1
-0.5
p=0.002
average # cats
average # cats
p<0.0001
0.5
M
average # cats
1.5
1.0
-0.5
-0.5
M+_D1
M+_D2
M-_D1
Treatment Group and Day
M-_D2
m+_D1
m+_D2
m-_D1
Treatment Group and Day
Figure 3.8 Change in cage use- the number of cats that displayed normal cage use day
1 to day 2; mean +/- SD; from left to right all treatments, macro and micro environment
40
m-_D2
Percent of cats that exhibited
at least one sickness behavior
Average number of sickness behaviors
p=0.8
p=0.08
100
3
percentage of cats
average SB/day
p=0.002
p=0.02
4
2
1
-_
2
2
80
60
40
-m
-_
1
1
M
-m
M
1
+_
+_
M
-m
2
-_
M
+m
M
-m
1
2
-_
M
+m
+_
+m
M
M
+m
+_
1
0
M-m+ (n=21)
M-m- (n=19)
M+m- (n=19)
M+m+ (n=17)
Treatment Group and Day
2
Day
Figure 3.9a. Average number +/- SD of
sickness behaviors.
Figure 3.9b. Percent of cats exhibiting at least
one sickness behavior
Figure 3.9 Sickness behaviors- Change in number of sickness behaviors displayed day
1 to day 2; mean +/- SD (right) and percent of cats exhibiting at least one sickness
behavior (left)
3.2 Scan sample data analysis
3.2.1 Behavior: Analysis of the number of cats exhibiting affiliative and
maintenance behaviors at each time point revealed that time points 4 (p<0.001), 5
(p<0.001), and 6 (p=0.001) were statistically significant (Figure 3.10; Table 3.4).
Analysis of the macro environment revealed that time points 4 (p<0.001), 5 (p<0.001), 6
(p<0.001) and 8 (p=0.006) were statistically significant (Figure 3.10; Table 3.5). Finally,
analysis of the micro environment revealed that no time points were statistically
significant at the p=0.006 level (Figure 3.10; Table 3.6).
Treatment
M+m+
M+mM-m+
M-mp-value
Time 1
4
6
1
2
0.105
Time 2
4
8
6
5
0.564
Day 1
Time 3
8
14
8
5
0.025
Time 4
9
16
3
3
<0.001*
Time 5
10
17
1
3
<0.001*
Time 6
12
16
7
8
0.001*
Day 2
Time 7
Time 8
12
13
18
18
10
11
12
12
0.022
0.023
Time 9
15
18
18
13
0.269
Table 3.4 Scan sampling behavior- Number of cats in each treatment group at each time
point exhibiting affiliative or maintenance behaviors
41
N
17
19
21
19
Treatmen
t
M+
Time
1
10
Time
2
12
Day 1
Time Time 4
3
22
25
M-
3
11
13
p-value
0.01
9
0.47
0
0.01
2
28
Day 2
Time Time
7
8
30
31
Time
9
31
4
15
22
23
23
<0.001
*
<0.001
*
0.01
2
0.006
*
0.13
6
Time 5
Time 6
27
6
<0.001
*
N
3
6
4
0
Table 3.5 Scan sampling behavior- Number of cats in the macro environments at each
time point exhibiting affiliative or maintenance behavior.
Treatment
Time 1
Time 2
Day 1
Time 3
m+
mp-value
5
8
0.361
10
13
0.451
16
19
0.490
Time 4
Time 5
12
19
0.102
11
20
0.027
Time 6
Day 2
Time 7 Time 8
Time 9
N
19
24
0.294
22
30
0.067
33
31
0.708
38
38
24
30
0.129
Table 3.6 Scan sampling behavior- Number of cats in the micro environments at each
time point exhibiting affiliative or maintenance behaviors
100
Day 2
Day 1
100
100
*
*
Affiliative and Maitenenace Behaviors
micro environment
Affiliative and Maitenenace Behaviors
macro environment
Affiliative and Maintenance Behaviors
M+m- (n=19)
*
Day 2
Day 1
M+m+ (n=17)
80
*
80
M-m+ (n=21)
*
Day 2
Day 1
M+ (n=36)
*
m+ (n=38)
m- (n=38)
80
*
M- (n=40)
40
% of cats
% of cats
% of cats
M-m- (n=19)
60
60
60
40
40
20
20
20
0
1
2
3
4
5
Sample
6
7
8
9
0
1
0
1
2
3
4
5
6
7
8
9
2
3
4
5
6
7
8
Sample
Sample
Figure 3.10 Scan sampling behavior- Percentage of cats exhibiting affiliative or
maintenance behaviors, scan sampling; from left to right all treatment groups, macro and
micro environments
3.2.2 Position in cage: A summary table and graphs of position in the cage are
presented in Appendix F (table F.1, figure F.1). Comparison of the number of cats in the
hide box at each scan sample point was performed comparing treatments M+m+ and Mm+ with no significant results identified at any time point (figure 3.11). Further analysis
42
9
of treatment M+m+ and treatment M-m+ comparing day one scores to day two scores
showed a statistically significant decrease in score from day one to day two in treatment
M+m+ only (p=0.01) (figure 3.11).
A comparison of the number of cats perching at each scan sample point was
performed comparing treatments M+m+ and M-m+. Results showed that time point 6
was statistically significant (p=0.003) (figure 3.12). Further analysis of treatment M+m+
and treatment M-m+ comparing day one scores to day two scores showed a statistically
significant increase in score from day one to day two in treatment two only (p=0.04)
(figure 3.12).
Hide
Percentage of Cats in Hide Box
1.5
average hide score
Day 2
Day 1
100
% of cats
80
60
M-m+ (n=21)
40
M+m+ (n=17)
20
P=0.08
P=0.01
1.0
0.5
0.0
0
1
2
3
4
5
6
7
8
9
-0.5
Time Point
M+m+_1 M+m+_2
M-m+_1
M-m+_ 2
Treatment Group and Day
Figure 3.11 Hide box use- Percentage of cats in the hide box at each time point (left)
and change in average hide score day 1 to day 2, mean +/-SD (right)
Percentage of Cats Perching
Perch
Day 2
Day 1
1.5
average perch score
100
% of cats
80
60
*
M-m+ (n=21)
M+m+ (n=17)
40
20
0
1
2
3
4
5
6
7
8
P=0.3
P=0.04
1.0
0.5
0.0
9
-0.5
Time Point
M+m+_1 M+m+_2
M-m+_1
M-m+_ 2
Treatment Group and Day
Figure 3.12 Perching- Percentage of cats perching at each time point (left) and change
in average perch score day 1 to day 2, mean +/-SD (right)
43
3.2.3 Vocalization: A summary of the number and percentage of cats vocalizing
at each time point is presented in Appendix F, table F.2. The percentage of cats that
exhibited meowing or hissing/growling are depicted in figures 3.13 (combined data focal
and scan samples). Meowing at time point 16 was significant (p=0.006). No time point
was significant for hissing/growling. Analysis comparing M+ and M- environments on
day one and on day two identified a statistically significant decrease in hissing/growling
in the cats housed in M+ (p=0.05) from day one to day two.
Meowing
Day 2
Day 1
50
Hissing/growling
40
*
% of cats
% of cats
Day 2
Day 1
50
40
30
20
M+m- (n=19)
M-m- (n=19)
10
M-m+ (n=21)
1
2
3
4
5
6
7
8
20
M+m- (n=19)
M-m- (n=19)
M+m+ (n=17)
M-m+ (n=21)
10
M+m+ (n=17)
0
30
0
9 10 11 12 13 14 15 16 17
1
Time Point
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17
Time Point
Figure 3.13 Vocalizations- Percentage of cats who exhibited meowing (left) and
hissing/growling (right), scan sampling
3.4 Focal sample data analysis:
3.4.1. Frequency: Results of the analysis of the frequency behaviors approach,
rub, tail-up, alert, eat/drink, groom, rest, stretch, yawn, lip lick, turn away, and startle are
presented in table 3.7. No significant differences for eat/drink, groom, stretch, turn
away, yawn and startle were identified at any time point. The differences in frequency of
approach were statistically significant at time point 5 (p=0.005), of alert behavior at time
point 6 (p=0.005), of rub at time point 6 (p=0.005), of tail-up behavior at time point 6
(p<0.0001) and 8 (p=0.0001), of resting behavior at time point 3 (p=0.002), and of lip
licking behavior (non-appetitive) at time point 2 (p=0.003) (figure 3.14).
Further analysis to assess the effects of the macro and micro environments did
not identify any significant differences at any time point for approach or rub. Statistical
significance in the macro environment was found for tail-up behavior at time point 6
(p=0.006), resting behavior at time point 3 (p<0.0001) and lip licking behavior at time
point 2 (p=0.001). Assessment of the micro environment found no significant differences
at any time point for the above behaviors.
44
Behavior
Approach
Macro
Micro
Rub
Macro
Micro
Alert
Macro
Micro
Tail-up
Macro
Micro
Eat/drink
Macro
Groom
Macro
Micro
Stretch
Macro
Yawn
Macro
Micro
Rest
Macro
Micro
Lip Lick
Macro
Micro
Turnaway
Macro
Micro
Startle
Macro
Micro
1
0.9
0.8
1.0
0.6
0.05
0.09
0.2
0.3
0.06
0.6
0.2
1.0
0.5
0.1
0.4
0.3
0.2
0.6
0.3
0.6
0.8
0.6
0.2
0.4
0.6
0.8
0.6
0.4
0.3
0.2
0.7
0.5
0.4
0.09
2
0.5
0.2
0.9
0.03
0.02
0.3
0.04
0.2
0.07
0.1
0.2
0.3
0.3
0.5
0.6
0.4
0.8
0.3
0.3
0.9
0.8
0.7
0.4
0.003
0.001
0.06
0.2
0.7
0.06
0.3
0.8
0.5
3
0.4
0.3
0.3
0.02
0.1
0.02
0.01
0.5
0.003
0.3
0.5
0.6
0.3
0.9
0.2
0.1
0.4
0.08
0.03
0.5
0.8
0.2
0.002
<0.0001
0.8
0.2
0.7
0.06
0.4
0.8
0.6
0.7
0.3
0.3
Time point
4
5
0.4
0.005
0.3
0.1
0.3
0.05
0.03
0.03
0.06
0.007
0.08
0.2
0.1
0.07
0.3
0.5
0.05
0.01
0.4
0.02
0.3
0.05
0.3
0.03
0.03
0.3
0.01
0.3
0.8
0.8
0.9
0.3
0.9
0.9
0.5
0.8
0.01
0.5
0.03
0.3
0.05
0.5
0.08
0.6
0.6
0.2
0.3
0.5
0.7
0.5
0.06
0.06
0.5
0.009
0.007
0.4
0.9
0.3
0.8
0.1
0.7
0.6
0.4
0.7
0.7
0.6
0.3
0.08
6
0.09
0.03
0.3
0.005
0.1
0.008
0.005
0.1
0.01
<0.0001
0.006
0.01
0.9
0.5
0.3
0.06
0.4
0.8
0.5
0.6
0.9
0.2
0.05
0.3
0.04
0.06
0.06
0.07
0.5
0.4
0.4
0.3
Table 3.7 Focal sample frequency data
45
7
0.06
0.1
0.3
0.6
0.9
0.7
0.2
0.8
0.03
0.09
0.3
0.06
0.1
0.04
0.04
0.007
0.3
0.2
0.3
0.009
0.001
0.9
0.9
0.8
0.6
0.8
0.9
0.9
0.4
0.4
0.7
0.7
0.6
8
0.2
0.05
0.6
0.08
0.05
0.09
0.02
0.7
0.002
0.001
0.01
0.02
0.05
0.3
0.3
0.2
0.6
0.8
0.1
0.5
0.7
0.7
0.5
0.3
0.04
0.3
0.01
0.7
0.4
0.3
0.8
0.7
0.6
Approach
40
Object Rub
*
Day 1
Day 2
60
Day 1
Day 2
50
20
% of cats
% of cats
30
M+m- (n=19)
10
M+m+ (n=17)
M-m+ (n=21)
0
3
2
1
4
5
6
8
7
M-m- (n=19)
*
40
30
M+m- (n=19)
20
M+m+ (n=17)
M-m- (n=19)
M-m+ (n=21)
10
Time Point
0
1
2
3
4
5
6
7
8
Time Point
Alert
Day 1
100
*
Day 2
M-m- (n=19)
M+m- (n=19)
80
60
Day 2
Day 1
50
60
M-m+ (n=21)
M+m+ (n=17)
40
% of cats
% of cats
Tail-up
20
0
1
2
3
4
5
6
7
*
40
*
30
M+m- (n=19)
20
M+m+ (n=17)
M-m- (n=19)
M-m+ (n=21)
10
8
Time Point
0
1
2
3
4
5
6
7
8
Time Point
Lip Lick
Rest
80
M-m+ (n=21)
*
M+m+ (n=17)
M-m- (n=19)
40
% of cats
50
% of cats
Day 2
Day 1
60
M+m- (n=19)
30
*
60
M+m- (n=19)
40
M-m- (n=19)
20
20
10
0
M-m+ (n=21)
M+m+ (n=17)
1
0
1
2
3
4
5
6
7
Day 2
Day 1
100
2
3
4
5
6
7
8
Time Point
8
Time Point
Figure 3.14 Focal behavior frequency data- Percentage of cats that exhibited approach,
object rub, alert, tail-up, resting and lip licking behaviors in each treatment group, focal
sampling. * denotes statistical significance
3.4.2. Duration: Results of the analysis of the duration behaviors alert, freeze
and hide or attempt to hide are presented table 3.8. Analysis of alert behavior showed
46
that the effects of both time and treatment were significant (p<0.0001) (figure 3.15).
Further analysis of alert behavior in the macro environment showed that the effect of
time was significant (p<0.0001) (figure 3.15) while the effect of both time and treatment
were significant in the micro environment (p<0.0001) with the cats in the m- condition
exhibiting longer durations of alert behavior than those in m+ (figure 3.15).
Analysis of freezing behavior showed that the effect of both time (p<0.0001) and
treatment (p=0.005) were significant (figure 3.16). Further analysis of stay behavior in
the macro environment showed that the effect of time (p<0.0001) and treatment
(p=0.001) were significant (figure 3.16b) as they were in the micro environment (time
p<0.0001, treatment p=0.02) (figure 3.16).
Analysis of hiding or attempting to hide showed that the effect of both time
(p=0.005) and treatment (p<0.0001) were significant (figure 3.17). Further analysis
showed that only the effect of time (p=0.0003) was significant in m+ environment and no
significant differences were found in the m- environment (figure 3.17).
No statistically significant effects of time or treatment were found for duration of
tail-up, grooming, resting, turning away, and crouching behaviors.
All
Behavior
Freeze
Macro
micro
0.32
3.58
1.67
1.39
2.02
1.34
1.23
0.13
7
1.606
0.92
7
0.3748
0.007
21
1.95
0.01
7
2.683
0.03
7
2.225
0.1
21
1.433
0.2
7
1.346
0.49
7
0.9
9.63
9.81
9.83
4.47
4.48
4.26
1.78
4.11
1.05
<0.0001
7
11.51
<0.0001
7
11.8
<0.0001
7
11.62
<0.0001
7
7.297
<0.0001
7
7.194
<0.0001
7
6.808
0.0005
7
3.783
0.0003
7
4.135
0.6
7
0.78
8.67
0.3
7.66
10.58
6.54
3.76
17.48
2.7
2.62
<0.0001
3
9.965
0.4
1
0.74
<0.0001
1
25.87
0.0005
3
6.585
0.001
1
11.35
0.02
1
6.113
<0.0001
3
9.133
0.2
1
1.664
0.16
1
2.01
20.3076
28.68
21.32
38.04
42.08
44.87
45.29
56.78
46.91
<0.0001
70
2.427
<0.0001
72
3.355
<0.0001
72
2.451
<0.0001
71
6.12
<0.0001
3
6.484
<0.0001
73
6.872
<0.0001
71
9.501
<0.0001
35
11.43
<0.0001
36
6.812
All
Alert
Macro
micro
2.62
1.33
0.4
21
1.046
Source of variation
Interaction
Time
Treatment
Subjects
(matching)
% total
variation
p-value
DF
F
% total
variation
p-value
DF
F
% total
variation
p-value
DF
F
% total
variation
p-value
DF
F
Table 3.8 Focal behavior duration
47
Hide or Attempt to hide
All
E+
E-
Alert
Treatment and Time p<0.0001
Alert micro enviornment
Treatment and Time p<0.0001
Day 2
300
250
250
200
M-m- (n=19)
M+m-(n=19)
150
M+m+ (n=17)
Day 1
300
Day 2
Day 1
Day 2
250
200
M+(n=36)
150
M-(n=40)
100
seconds
300
seconds
Seconds
Day 1
Alert macro environment
Time p<0.0001
200
m-(n=38)
150
m+(n=38)
100
100
50
M-m+ (n=21)
50
50
0
0
1
0
1
2
3
4
5
6
7
2
3
4
5
6
7
8
1
2
3
4
Time Point
8
5
6
7
8
Time Point
Time Point
Figure 3.15 Focal behavior- Mean duration of time spent alert, focal sampling; from left
to right all treatment groups, macro and micro environments
Freeze
Treatment p=0.0005 Time p<0.0001
Day 1
Day 2
Freeze macro environment
Treatment p=0.001 Time p<0.0001
300
M-m+ (n=21)
M-m- (n=19)
150
seconds
Seconds
M+m+ (n=17)
200
Day 1
300
Day 2
250
250
M-(n=40)
200
M+(n=36)
150
100
Day 1
Day 2
250
seconds
300
Freeze micro environment
Treatment p=0.02 Time p<0.0001
m+(n=38)
200
150
m-(n=38)
100
100
M+m- (n=19)
50
50
50
0
0
1
0
1
2
3
4
5
6
7
2
3
4
5
6
7
8
1
2
3
Time Point
8
4
5
6
7
8
Time Point
Time Point
Figure 3.16 Focal behavior- Mean duration of time spent freezing, focal sampling; from
left to right all treatment groups, macro and micro environments
Hide or Attempt to Hide
Treatment p<0.0001 Time p=0.005
Hide or Attempt to Hide in m+
Time p=0.0003
300
Day 2
150
M-m+ (n=21)
100
M+m+(n=17)
M+m-(n=19)
M+m- (n=19)
50
0
2
3
4
5
Time Point
6
7
8
seconds
seconds
200
1
Day 1
300
Day 2
250
250
Day 1
Day 2
250
200
150
M-m+(n=21)
100
M+m+(n=17)
50
seconds
Day 1
300
Hide or Attempt to Hide m-
200
150
100
M-m-(n=19)
50
0
M+m-(n=19)
0
1
2
3
4
5
Time Point
6
7
8
1
2
3
4
5
6
7
8
Time Point
Figure 3.17 Focal behavior- Mean duration of time spent hiding or attempting to hide,
focal sampling; from left to right all treatment groups, macro and micro environments
48
3.5 Stranger Approach Test: A summary table of the one way ANOVA results is
presented in table 3.9.
3.5.1 Latency to interact: Analysis of the four treatment groups showed that
latency to interact was significant at step one (p=0.002), step two (p=0.001) and step
three (p=0.025) (figure 3.18).
When assessing the effect of the macro environment on the latency to interact,
the results were not significant at step one (p=0.2) or step two (p=0.2) but step three was
significant (p=0.03). The same analysis to assess the effect of the micro environment
showed significance at step one (p=0.0009), step two (p=0.0002) and step three
(p=0.04) (figure 3.19).
3.5.2 Duration of interaction: Analysis of the four treatment groups showed that
duration of interaction was significant at step one (p=0.002), step two (p=0.004) and step
three (p=0.01) (figure 3.20).
When assessing the effect of the macro environment on the duration of
interaction, the results were not significant at step one (p=0.1) or step two (p=0.1) but
step three was significant (p=0.03). The same analysis to assess the effect of the micro
environment showed significance at step one (p=0.001), step two (p=0.0007) and step
three (p=0.009) (figure 3.21).
3.5.3 Approach score: Analysis of the four treatment groups showed that the
approach score was significant at step one (p=0.03), step two (p=0.02) and step three
(p=0.007) (figure 3.22).
When assessing the effect of the macro environment on mean approach score,
the results were significant at step one (p=0.05), step two (p=0.03) and step three
(p=0.008). The same analysis to assess the effect of the micro environment showed
significance at step one (p=0.03), step two (p=0.03) and step three (p=0.02) (figure
3.23).
49
Dunn's Multiple Comparison Test ns= not significant, *<0.05, **<0.01
Latency to interact
1
2
3
**
**
ns
ns
ns
ns
ns
ns
ns
**
**
*
ns
ns
ns
ns
ns
ns
M+m+ vs M+mM+m+ vs M-m+
M+m+ vs M-m
M+m- vs M-m+
M+m- vs M-mM-m+ vs M-m-
Duration of Interaction
1
2
3
ns
ns
ns
ns
ns
ns
ns
ns
ns
**
**
**
ns
ns
ns
ns
ns
ns
1
ns
ns
ns
*
ns
ns
Approach Score
2
3
ns
ns
ns
ns
ns
ns
*
**
ns
ns
ns
ns
Table 3.9 Approach Test Study One
Latency to Interact 2
p=0.001
Latency to Interact 1
p=0.002
Latency to Interact 3
p=0.03
40
40
Time (seconds)
30
20
30
Time (seconds)
Time (seconds)
40
20
10
30
20
10
10
0
M+m+
M+m-
M-m+
M-m-
0
M+m+
Treatment Group
0
M+m+
M+m-
M-m+
M+m-
M-m+
M-m-
Treatment Group
M-m-
Treatment Group
Figure 3.18 Latency to interact step 1, 2, and 3, all treatment groups; mean +/- SD
Latency to Interact 2
Latency to Interact 1
p=0.2
40
p=0.0009
p=0.2
20
10
p=0.0002
30
20
10
0
M+
M-
m+
Treatment Group
m-
p=0.04
p=0.03
Time (seconds)
30
Time (seconds)
Time (seconds)
Latency to Interact 3
40
40
30
20
10
0
0
M+
M-
m+
Treatment Group
m-
M+
M-
m+
m-
Treatment Group
Figure 3.19 Latency to interact step 1, 2, and 3, macro and micro environments; mean
+/- SD
50
Duration of Interaction 3
p=0.01
Duration of Interaction 2
P=0.004
40
30
30
20
40
Time (seconds)
40
Time (seconds)
Time (seconds)
Duration of Interaction 1
p=0.002
20
30
20
10
10
10
0
M+m+
0
0
M+m+
M+m-
M-m+
M+m+
M-m-
M+m-
M-m+
M+m-
M-m+
M-m-
Treatment Group
M-m-
Treatment Group
Treatment Group
Figure 3.20 Duration of interaction step 1, 2, and 3, all treatment groups; mean +/- SD
Duration of Interaction 3
Duration of Interaction 2
Duration of Interaction 1
40
40
40
p=0.03
p=0.1
p=0.001
20
10
p=0.009
p=0.0007
30
Time (seconds)
30
Time (seconds)
Time (seconds)
p=0.1
20
30
20
10
10
0
0
M+
0
M+
M-
m+
m-
M+
M-
Treatment Group
m+
m-
M-
m+
m-
Treatment Group
Treatment Group
Figure 3.21 Duration of interaction step 1, 2, and 3, macro and micro environments;
mean +/- SD
Approach Score 3
p=0.007
Approach Score 2
p=0.02
6
5
5
4
4
3
6
5
4
Score
6
Score
Score
Approach Score 1
p=0.03
3
3
2
2
2
1
1
1
0
M+m+
0
0
M+m+
M+m-
M-m+
Treatment Group
M-m-
M+m+
M+m-
M-m+
M-m-
M+m-
M-m+
Treatment Group
Treatment Group
Figure 3.22 Mean approach score step 1, 2, and 3, all treatment groups; mean +/- SD
51
M-m-
Approach Score 2
Approach Score 1
6
3
5
4
4
Score
Score
4
5
3
p=0.02
p=0.008
p=0.03
p=0.03
p=0.03
p=0.05
5
Score
Approach Score 3
6
6
3
2
2
1
1
2
1
0
0
M+
0
M+
M-
m+
m-
M-
m+
Treatment Group
m-
M+
M-
m+
Treatment Group
Treatment Group
Figure 3.23 Mean approach score step 1, 2, and 3, macro and micro environments;
mean +/- SD
4. Discussion
The results of this study supported the hypothesis that cats housed in the
enriched environments would be less distressed (as evidenced by their behavior) than
cats in the unenriched environments. First, all of the measures of cage condition (food
intake, cage use, urination, defecation, and total sickness behavior) suggested that all
cats experienced distress. Between 80-100% of cats in each treatment group exhibited
at least one sickness behavior (decreased food intake, no eliminations in the previous 24
hours, eliminating out of the litter pan, vomiting and diarrhea) on day one, and this
decreased slightly on day two. More cats housed in the unenriched room than in the
enriched room exhibited a sickness behavior. The results of this study agree with
previously published studies of sickness behaviors in response to environmental
disturbances in research cats housed in a laboratory (Stella et al., 2011; 2013).
Sickness behavior is a change from activities such as feeding, social contact, or
grooming to processes that conserve energy to boost immune function to fight
pathogens or in response to psychological stress (Raison and Miller, 2003; Dantzer et
al., 2008). This likely occurs by activation of the stress response system leading to
release of corticotrophin-releasing factor (CRF) (which controls the release of
glucocorticoids via ACTH). This in turn activates the sympathetic nervous system and
the immune system, causing the release of pro-inflammatory cytokines while increasing
vigilance behaviors and suppressing maintenance behaviors (Sapolsky, 2004; Marques-
52
m-
Deak et al., 2005). This is a well documented response that has been reported in many
species including rodents (Broom, 2006) and dairy cattle (Fogsgaard et al., 2012).
All cats had decreased food intake on day one, but cats in the enriched room
environment appeared to acclimate more quickly and were more likely to eat after the
initial acute stressor of placement in a novel environment than were cats housed in the
unenriched environment. Inhibition of food intake in response to human activity and
husbandry has been found in many species, including sea bass (Leal et al., 2011), and
decreases in total daily cumulative food intake has been found in rats following acute
restraint stress, which is a psychological stressor (Calvez et al., 2011). Although the
precise mechanisms underlying the inhibition of food intake are not fully understood, it
appears to be complex, and to involve CRF. The role of CRF is critical to the
relationship between food intake and stress and has been recognized to suppress food
intake (Krahn et al., 1986; Nakayama et al., 2011).
Eliminative behavior was unlikely to occur on day one regardless of the treatment
group. Thereafter, it was affected more by the macro than the micro environment.
Thirty six of the 76 cats did not have a BM during the study period and ten of these cats
did not urinate either. In homes, cats will normally urinate two to four times a day and
defecate once or twice a day (Overall and Dyer, 2005). The lack of eliminations
suggests distress in this subset of cats.
Few instances of vomiting and diarrhea were seen, which may have been related
to the low food intake, or to activation of the stress response system. A recent review
reported that activation of CRF signaling pathways mediates both the inhibition of upper
gastrointestinal and the stimulation of lower gastrointestinal motor function through
interaction with different CRF receptor subtypes (Stengel, 2009). Behavioral inhibition of
eating, drinking, and eliminating should be considered as evidence of distress. These
are behaviors that are easy to recognize and quantify, and have been found in
populations of owned cats in homes (Heidenberger, 1997) and in laboratory cats (Stella
et al., 2011; 2013). Daily inspection of the cage appears to be the most reliable way to
determine if cats are eating, drinking, and eliminating. Comparison with direct
observation data (focal and scan sampling) revealed how seldom one could actually
observe a cat engaging in these behaviors.
When looking at how the cats used their cages, disrupted use was seen much
less often than no use, suggesting that most cats suppressed active behaviors in this
53
environment. Studies of both domestic and wild felids in captivity have found decreased
exploratory behavior (Carlstead et al., 1993a) and increased attempts to hide (Carlstead
et al., 1993b; Rochlitz, 2000), particularly when the environment was unenriched or
unpredictable. However, when assessed in an open field test after exposure to a “nontraumatic” acute stressor (bright light or white noise) rats increased activity as their initial
stress response (Roth and Katz, 1979). Cage use is important because it is often used
as a criterion to determine if a cat is feral by rescue organizations. Slater et al. (2010)
reported that a common observation was that cats thought to be feral were more likely to
disrupt the cage than cats they described as frightened socialized cats. In the current
study, more cats disrupted the cage on day two than on day one, and they did so slightly
more often when housed in unenriched environments than enriched ones. These results
suggest that as cats acclimate they may actually be more active in attempts to hide or to
explore so that cage disruption may relate more to available resources and
environmental factors than to socialization to humans. Therefore, using this criterion to
asses “feralness” may be problematic. Additionally, whether a cat disrupts its cage or
shows no use may depend upon the coping style of the cat. Korte et al. (2005) have
proposed that individuals differ in their response to stressors by either being proactive or
reactive. A proactive response is characterized by fight-flight, aggression and high risk
behaviors whereas a reactive response is more likely to be in the form of freeze-hide,
low aggression and low risk behaviors. Although unstudied in cats, this framework could
offer an explanation for cage use in cats.
In support of previous published studies (Rochlitz, 2000; Gourkow and Fraser, 2006;
Kry and Casey, 2007), cats afforded the opportunity to hide or perch were likely to be
exhibiting these behaviors when observed, supporting the proclivity of cats to hide when
confronted with challenging environments. The cats in the enriched cages were
observed in the hide or in the perch area in 77.1% of the scan samples. In a study of
cats in shelters, Kry and Casey (2007) compared cats in an enriched group provided
Hide, Perch & Go boxes, with those in an unenriched group (not provided boxes). They
reported that the cats in the enriched group were found in either the hide or perch area
of the Hide, Perch & Go boxes 77% of the time as well. Additionally, in the current
study, the duration of hiding or attempting to hide was longer in the cats housed in the
unenriched room compared to those housed in the enriched room. Together this
indicates that cats may have a need to hide and that hiding allows them to partially
54
isolate from unfamiliar conspecifics and humans when confronted with a threatening
environment. Not providing cats the opportunity to hide is likely to adversely affect their
level of distress and in turn, their overall welfare, so all efforts should be made to meet
this need.
When looking at perching behavior, all cats had an increase in perching from day
one to day two but a significant difference was only identified in the cats housed in the
unenriched room environment. These results suggest that time spent engaged in
perching behavior may change in response to the environment along with acclimation to
confinement. Further research is needed to understand the motivation behind perching
behavior in this environment.
Results of the scan sampling of behavior suggests that cats housed in enriched
environments acclimated more quickly than those housed in unenriched environments
as seen by the higher percentage of cats exhibiting affiliative or maintenance behaviors
in the enriched environments. Additionally, the effect of the room appeared to be greater
than the effect of the cage. This suggests that aspects of confinement housing such as
lack of predictability, as well as noise and disruptions, may have an adverse effect on cat
welfare.
Analysis of the focal sampling data found few behaviors of statistical significance.
Differences between treatment groups were identified for three measures of the duration
of observation time; alert, freeze, and hide or attempt to hide (discussed above)
Additionally, the frequency of six behaviors with at least one statistically significant time
point were identified; alert, approach, rub, tail-up, rest, and lip lick. Alert behavior is an
expression of vigilance which has been found to be a reliable measure of fear in
animals. Results of a study of dairy cows suggest that cows alter their vigilance
behavior according to their level of fear toward humans and environmental stimuli.
Therefore, this measure may provide information about the degree of fear the individual
is experiencing (Welp et al., 2004) especially given that the Darwinian concept of
evolutionary adaptation would favor individuals by increasing the survival of those that
express greater vigilance to avoid threats from predators and conspecifics (Boissy,
1995). The cats in this study did alter the amount of time spent engaging in vigilance
behaviors with increases in the morning of day two which coincided with morning
husbandry, uncovering of the cages and resetting of the video equipment. When looking
at alert behavior, the micro environment appeared to affect the behavior of the cats more
55
than the macro environment with the cats that were in unenriched cages spending more
time alert. There are two possible explanations for this. First, it could be that the cats
that were not provided hide boxes were easier to see and therefore alert behavior was
coded as longer in duration and increased in frequency. Alternatively, the environment
may have been perceived as more challenging to cats that were not provided a hide box
resulting in the difference in time spent alert.
Freezing behavior is a type of motor inhibition or behavioral response to aversive
events and may also be indicative of the degree of fear being experienced by the
individual (Boissy, 1995). Freezing behavior in rodents is an anti-predator defensive
behavior that has been associated with fear, more than anxiety, directed towards an
unambiguous threat (Marks and Nesse, 1994). Freezing may be a beneficial response
by aiding in the location and assessment of the threat, concealment, and inhibition of the
predator’s attack reflex. When looking at the freezing behavior in this study there
appears to be an effect of both the macro and micro environment with the macro effect
being larger, so that cats housed in the unenriched room environment exhibited freezing
behavior for longer durations. Cats in unenriched rooms may have exhibited freezing
behavior perhaps due to the perception of the environment as threatening whereas cats
that had hide boxes may have been experiencing the same emotional state but were
better able to cope with the environment.
An interesting finding was that lip licking behavior was frequently seen in all groups.
Although this is a well documented behavior in dogs, this is the first time it has been
reported in a study of domestic cats. Lip licking in dogs has been defined as an
appeasement gesture to signal to others that it is feeling anxious, uncomfortable or
fearful (Beerda et al., 1997). Unlike in dogs where it is typically seen in circumstances
where the dog is signaling its intent to another (human or conspecifics), in the cats it was
seen on the videotaped observations when there was no direct target of the behavior.
Additionally, this behavior was not evenly distributed, being exhibited frequently in some
individuals but never in others. Further study is needed to understand lip licking
behavior in cats.
Some measures of positive welfare such as species typical behaviors including play
and grooming were rarely observed. One study of daily time budgets of group living cats
observed cats in laboratory housing for eight hours a day from 8:00-16:00h with the
following results; 36% of time was spent in maintenance behaviors (resting, sitting,
56
drinking, eating and eliminating), 30% in comfort behaviors, and 24.5% in locomotory
behaviors (Podberscek et al., 1991). Another study found grooming behavior consumed
about 4% of the time budget (Eckstein and Hart, 2000). Play has been shown to cause
an increase in opioidergic activity in the nucleus accumbans in both humans and
animals (Held and Špinka, 2011). Play is easy to recognize, closely linked to current
environmental conditions and is exhibited by most mammals in the absence of threats to
fitness, but will typically be absent from the behavioral repertoire in challenging
conditions. In the current study, only one of the 76 cats was observed to exhibit play
behavior. The lack of observation of behaviors associated with positive affect in the cats
in this study suggests that fear or anxiety may have been experienced that was intense
enough to disturb or inhibit some species typical behaviors, as has been documented in
other species (Boissy, 1995; Boissy et al., 2007).
Vocalizations were not common in this study environment. Most vocalizations
occurred in the morning of day two with cats in the enriched room environment more
likely to hiss, growl and meow than cats in the unenriched room environment. This was
a time of high arousal due to the researcher being present to conduct husbandry and
prepare for the day. Cats generally use vocalizations for short-range direct
communication (Bradshaw, 2000). For instance, purring is a ubiquitous vocalization
among domestic cats and has been found in a range of circumstances, most of which
involve cat-cat or cat-human contact, including when they are content, fearful, or in pain
(Beaver, 1992). It has been proposed that purring may function as a “manipulative”
contact- or care-soliciting signal possibly derived from its function in the neonate
(Bradshaw, 2000). Similarly, meow is rarely heard in cat-cat interactions and is thought
to be a learned response to elicit human attention (Bradshaw, 2000). This would seem to
explain the observation that feral cats are rarely heard meowing in confinement and the
frequency and pattern of meowing seen in this study. In contrast, a recent study showed
that feral cats both growled and hissed more frequently than socialized cats when
confronted with an agonistic encounter and concluded that socialization plays a role in
the type of vocalization a cat uses (Yeon et al., 2011). The increase in frequency of
auditory signals has been proposed as a measure of the short-term welfare of other
species including livestock species (Weary and Fraser, 1995; Manteuffel et al., 2004),
silver fox (Gogoleva et al., 2010) and rats (Knutson et al., 2002). However, vocalizations
may not be reliable indicators of affect in cats, possibly because their social structure as
57
a solitary predator does not include a need to vocalize to maintain group cohesion or
contact with a group and may therefore be more indicative of their degree of
socialization and attachment to humans. As such, vocalizations should be considered
within the context in which the cat is being observed and in conjunction with other
behaviors. Additionally, overall a small percentage of cats vocalized, which is in
agreement with the low percentage of cats exhibiting cage disruption and eliminations.
This may suggest that cats are more likely to act like prey (reactive coping) rather than
predator (proactive coping) in threatening situations, but further research is needed.
The stranger approach test showed that cats housed in the enriched cages had
greater latency to interact, had shorter duration of interaction, and had lower approach
scores than did cats in unenriched cages at all three steps. This result suggests that cats
with hide boxes were less likely to approach than were cats without hide boxes.
Additionally, cats housed in the unenriched room environment had a greater latency to
approach, had shorter duration of interaction, and had lower approach scores than cats
housed in the enriched room environment. This indicates that a macro by micro
environmental interaction existed with cats in the treatment M+m- having the shortest
latency to interact, the longest duration of interaction, and the highest approach score.
Response to human approach has been shown to be affected by the environment
(Søndergaard and Halekoh, 2003; Graml et al., 2008), paternity and socialization
(McCune, 1995), gender of the tester (Lore and Eisenberg, 1986), gender of the
individual being tested (Vandenheede and Bouissou, 1993; Wells and Hepper, 1999),
and temperament (Lansade and Bouissou, 2008) in species including cattle, pigs,
sheep, poultry, horses, dogs, and cats. This approach test aimed to assess the
fearfulness of the cats after 48 hours of housing in different environments with the
outcome predicted to show that housing in a more enriched environment would lead to
cats exhibiting shorter latencies and longer durations of interaction. The results showed
that an enriched room, but not an enriched cage, led to the expected results of shorter
latency to interact, longer duration of interaction, and higher approach scores. However,
other factors may be involved in cats’ responses to an unfamiliar human including
socialization to humans and conspecifics and temperament or personality. Further
investigation of stranger approach tests is needed to verify and validate the robustness
of this test as a tool to study fearfulness and socialization in domestic cats.
58
This study has limitations. First, the number of subjects was relatively small and
the groups were of unequal size. Several measures were approaching statistical
significance so larger sample sizes may have resulted in more definitive results.
Second, the only macro environmental factor assessed was noise. Other factors such
as the type and intensity of the lights as well as the light:dark cycle, ambient room
temperature, and odors may affect the welfare of confined cats and should be
investigated. Third, the enriched cage had hide, perch, consistent daily setup, canned
and dry food presented in separate bowls, and deep litter in the litter pan. To determine
which variable is most important to the cat, these factors should be studied separately in
the future. Finally, the stress response was not measured using a physiologic metric
such as cortisol or another product of the stress response system. Elevation of
corticosteroid in plasma levels is one of the most evolutionarily conserved markers of
arousal, and is often used as an indicator of the degree of stress experienced. However,
because of the potential stress induced by collection of blood or saliva needed to
measure cortisol levels, it was decided that the risks outweighed the benefits of such
metrics. Urine cortisol: creatinine ratios and fecal cortisol levels have also been used,
but this study was of a short duration and many cats did not eliminate. Salivary cortisol
has also been measured in cats but they were conditioned for collection, whereas the
cats in this study were not, therefore ruling it out as an option (Siegford et al., 2003).
Although physiologic measures are useful indicators of the stress response, in
agreement with Jenson and Toates (Jensen and Toates, 1997), stress may be
considered as primarily a behavioral/psychological result. Additionally, in previous
studies of laboratory cats the behavioral measures appeared to accurately reflect the
stress response (Carlstead et al., 1993b; Stella et al., 2013)
In conclusion, the macro environment appears to be at least as important to the
cat as the micro environment. This is an important finding, since most research and
interventions are aimed at enriching the cage. Without attention to the macro
environment the welfare of confined cats may never be adequate. Modifications to
husbandry protocols, management changes such as decreasing noise levels, and
housing cats in areas away from dog areas are cheap, easy and effective in managing
perceived threats to cats in cages. Cats housed in the enriched room environments
acclimated more quickly, exhibiting more behaviors indicative of positive affect, including
approach, object rub, and tail-up behaviors as well as resting behavior, by the end of
59
day one, whereas the cats in the unenriched room environments took longer to
acclimate, exhibiting these behavior only towards the end of day two. This indicates that
the disruptive environment was more likely to disturb the expression of affiliative and
maintenance behavior.
However, regardless of the housing environment, cats in all groups did show
increases in measures of food intake, eliminations, and affiliative and maintenance
behaviors with time. This suggests that the majority of cats acclimated to confinement
given enough time, supporting findings from earlier studies. For example, Kessler and
Turner (1997) assessed the ability of cats to acclimate to a boarding facility over a two
week period using the Cat Stress Score. The results indicated that two thirds of the cats
acclimated, one third found boarding distressful, and 4% never acclimated. The daily
Cat Stress Scores of the singly housed cats declined significantly from day one to day
five, and overall stress levels continued to decrease during the two weeks of boarding.
Similarly, Kry and Casey (2007) aimed to determine if adding a hide box improved the
cats’ ability to cope with the stressful environment and reported that cats appeared to be
coping, indicated by lower Cat Stress Scores, by day three. This is an important finding
for the management of cats admitted to shelters to ensure cats are afforded enough time
to acclimate before culling decisions are made. These results indicate 48 hours may be
the minimum amount of time that cats need to acclimate to novel environments.
All cats exhibited behaviors indicative of fear and anxiety at the beginning of the
study, which are negative emotional states. Negative emotions along with reductions in
growth, production, and reproduction in confined animals are generally thought of as
indicators of poor animal welfare (Désiré et al., 2002; Boissy et al., 2007). Gray (1979)
classifies fear-producing stimuli into five subdivisions: 1) dangers that are part of the
evolutionary history of the species, 2) novelty, 3) resulting from learning (conditioned
response), 4) the intensity of the stimuli (physical characteristics) associated with
predation, and 5) interactions with conspecifics. In this study threats included unfamiliar
conspecifics, humans, and recordings of barking dogs as well as a novel environment.
Additionally some cats may have had previous experiences at boarding kennels,
shelters, or veterinary clinics and therefore had a conditioned fear response to the
environment. Therefore, all of Gray’s fear-producing stimuli were present for at least
some of the cats in this study and are typically present in confinement housing of cats.
60
The results of this study suggest that cats may be distressed when confronted with a
novel confinement experience, particularly in the first 24 hours. Additionally, if provided
the opportunity to hide they spend much of their time doing so, suggesting this is an
important behavioral response when fearful or anxious. Therefore, providing an
environment with minimal disturbances and an opportunity to hide may be beneficial to
the welfare of cats confined to cages, especially in the first two days.
Future research should aim to identify obstacles to implementation, investigation of
other aspects of the macro environment, and the effect of the quality of human-animal
interactions on the welfare of confined cats.
61
Chapter 4
The Behavior and Welfare of Domestic Cats (Felis sylvestris catus)
Allotted More Cage Space than the Norm of U.S. Housing Facilities
1. Introduction
Optimal cage size is often discussed, debated and legislated for species that are
confined. For example, societal concern for farm animal welfare has led to legislation in
many Western countries to improve housing standards with respect to cage size and
space allocations for animals. Stocking densities and pen size are increasingly
important to public perceptions of farm animal welfare. Currently, fourteen states in the
U.S. have established space and movement limitations for selected species of farm
animals by citizen petitions and legislative bills, according to the National Agricultural
Law Center. In January 2012, this increasing pressure and patchwork of state
regulations led to the adoption of national standards by the United Egg Producers (UEP)
in the U.S. that will nearly double the size of layer hen cages.
A similar concern for cats housed in shelters has gained recent attention. Several
groups have established recommendations for cage size, but little research has been
conducted in this area. Studies suggest that cat housing in U.S. shelters is typically a
small, single cat cage of less than 0.56 square meters (6 square feet) of floor space
leading to an inability of the cat to express normal species-typical behaviors such as
lying in a full stretch, walking more than a few steps, running, jumping, playing, and
eating away from elimination areas (McCobb et al., 2005; Rees and Lubinski, 2008;
Tanaka et al., 2012). Results have suggested that cage size may impact cat health and
welfare in these facilities. In addressing these concerns, the Association of Shelter
Veterinarians, in its Guidelines for Standards of Care in Animal Shelters, states cats
should have “…a minimum of four square feet of floor space in its primary enclosure if it
is a cage. Vertical space should be a minimum of two feet. If the physical space
limitations make it necessary to have less than four square feet then the vertical space
should be increased to compensate. This can be accomplished by placing a solid
62
resting place in the cage that allows the cat to stand underneath, and rest comfortably
on top of it” (AoSV, 2012).
Requirements for the housing of laboratory cats in the U.S. are similar to those
proposed for shelter cats. The National Research Council recommendations for
laboratory housing are 0.28 square meters (3 square feet) of floor space, 61 cm (24
inches) high for cats less then 4kg and 0.37 square meters (4 square feet) of floor
space, 61 cm (24 inches) high for cats more then 4kg (NRC, 1996). Meanwhile
laboratory cats in Australia are afforded 0.3-0.5 square meters (3.2-6.4 square feet) per
cat (NHMRC, 2012) while in Great Britain the housing recommendations are “…sufficient
space to permit housing in socially compatible groups, allow separate areas for urination
and defecation, activity and resting/sleeping, and allow essential enrichment such that
cats can perform a wide range of normal behavior” (nc3rs, 2012).
What little research has been done on the effects of cage size on cat behavior
and welfare has been inconclusive. Kessler and Turner (Kessler and Turner, 1999)
used the Cat Stress Score to measure acclimation to confinement and reported that cats
housed singly in cages 0.7 square meters (7.5 square feet) exhibited higher Cat Stress
Scores than cats housed in cages 1.0 square meters (10.8 square feet) on days 1, 2 and
6 of a one week stay. They concluded that this study gave no indication of the minimum
size needed and that the results may reflect the qualitative aspects (hiding but not
perching was available) of the housing as well as the quantitative ones. Based on the
above study, a group in Japan recently assessed behavior and stress in cats in three
different size cages and compared the results to those of group housed cats in an
enriched enclosure (Uetake et al., 2012). Small, medium and large cages were
compared, with large cages also equipped with a wooden log for cats to jump on.
Results indicated that singly housed cats were less active than group housed cats
regardless of the size of the cage, but that cats were not distressed based on behavior
measures and urine cortisol:creatinine ratios when housed in small cages for short
periods of time (six days per cage size).
The aim of this study was to evaluate the behavior and welfare of cats housed in
enriched or unenriched macro (room) environments and enriched or unenriched micro
(cage) environments with greater than the typical space allowance 1.1 square meters
(11.8 square feet) of caged cats in the U.S. The hypothesis was that cats afforded 1.1
square meters (11.8 square feet) of floor space and housed in the enriched
63
environments would be less distressed (as evidenced by their behavior) than cats
afforded the same amount of floor space and in unenriched environments.
2. Materials & Methods
2.1. Subjects: Adult cats between 0.5-11 years of age (mean age 2.96 years)
were recruited from The Ohio State University faculty, staff and students of the College
of Veterinary Medicine. Fifty nine neutered cats, 34 male and 25 female, completed the
study. All cats were healthy and current on viral rhinotracheitis, calicivirus,
panleukopenia, and rabies vaccines at the time of participation. Informed consent was
collected from all owners who volunteered a cat for the study. Cats were randomly
assigned to one of four treatment groups that consisted of combinations of an enriched
(M+) or unenriched (M-) macro environment and an enriched (m+) or unenriched (m-)
micro environment (Appendix A, Table A.2). Participants were admitted to the study
between 17:00-20:00 hours on day 0 and placed in an individual stainless steel cage.
The cage door was covered by a towel that had been sprayed with Feliwaya, a
commercially available synthetic cat pheromone, and the cat was left to acclimate
overnight. Behavioral observations were collected for eight hours a day from 08:0016:00 on days 1 and 2, after which cats were released to their owners’ between 17:0020:00 hours on day 2. The Animal Care and Use Committee of The Ohio State
University and the Clinical Research Advisory Committee of the Veterinary Medical
Center approved all experimental procedures used in this study.
2.2 Macro Environment: Cats were housed in the Ohio State University
Veterinary Medical Center (OSUVMC) vivarium. A 14:10 light:dark schedule was
maintained to mimic length of day in Ohio at that time of year (May/June). A mean ± SD
room temperature of 22 ± 1.6°C (72 ± 4°F) was maintained throughout the vivarium.
The room in which cats were housed was 4.57 meters by 4.88 meters and had cages
along three walls (see Figure 3.1a & b) with a small attached ante room used as a
storage and office area. Ten cats were randomized to each replicate with enriched and
unenriched cages balanced between upper and lower cages. The enriched room
(treatments M+m+ & M+m-) was a managed environment with minimal disturbances
from people, barking dogs or other unpredictable noises and events. The husbandry
routine was temporally consistent, with cats being provided care at the same time and in
the same order each day to provide the cats some predictability about their environment.
64
Cats in the unenriched room (treatments M-m+ & M-m-) environment were confronted
with multiple, random, unpredictable disturbances each day that included loud
conversations, talk radio, recordings of barking dogs, and noise from cage doors and
equipment being moved or dropped (see Appendix B, table B.1). Routine husbandry
occurred at different times of day, and cats were cared for in a random order each day.
All cats were cared for by a single researcher to ensure consistent handling. Cages
were spot cleaned so as to minimize disruption to the cat and to ensure consistency.
The daily husbandry and test procedure schedule is outlined in Appendix B.
2.3 Micro Environment: Cats were housed in individual stainless steel cages
measuring 140 × 156 × 150 cm (1.1 square meters). The front half of the slatted cage
floor was covered by a matb. The enriched cage (treatments M+m+ & M-m+) contained
a two tiered cardboard Hide, Perch & Goc box that had a lower hiding area (53 x 30 x 22
cm) with two access openings and an open upper sitting area (53 x 30 x 9 cm), placed in
the left rear corner of the cage. Bedding (84 x 74 cm cage pad folded into quarters) was
provided in both the ‘hide’ and ‘perch’ areas. A plastic litter pan (32 x 22 x 8 cm) was
placed in the right rear cage corner filled with Sani Chipsd litter at a depth of 3 cm. One
cup of dry commercial cat foode, 1.5 oz of commercial canned cat foodf, and water was
provided in separate 0.6L (20 oz) stainless steel bowls (see figure 3.2a). Unenriched
cages (treatments M+m- & M-m-) contained bedding (84 x 74 cm cage pad folded into
quarters) and a plastic litter pan (32 x 22 x 8 cm) with Sani Chipsd litter at a depth of 1.5
cm; no Hide, Perch & Goc box was provided. Food (one cup of dry commercial cat foode
and 1.5 oz of commercial canned cat foodf) was provided in the same 0.6L (20 oz)
stainless steel bowl. Water was provided in a separate bowl (see figure 3.2b), and cage
items were rearranged daily (see figure 3.3).
2.4 Data Collection: All participating owners were asked to complete a prestudy questionnaire for each cat consisting of information about demographics, health
history and home environment (Appendix C). Prior to routine husbandry each day, one
researcher stood in front of each cage for 30-60 seconds recording food intake,
urination, defecation, cage use, and additional sickness behaviors (vomiting, diarrhea,
eliminating out of the litter pan) for each cat (see Table 3.3). During husbandry, anything
that could not be seen from outside the cage was recorded for further analysis (e.g.,
cached food, vomit, or eliminations out of the litter pan).
65
Behavioral observations were collected between 08:00-16:00 hours using two
sampling techniques. A scan sample was collected every two hours that included the
cat’s position in the cage, the type of behavior(s) it was exhibiting and vocalizations
based on an ethogram for cats in cages developed based on observations of cats in
cages in shelters, veterinary hospitals and research laboratories (see Table 3.4). The
observer stood quietly in the middle of the housing room and recorded these
parameters. Observation time for each scan sample was approximately three minutes.
On the alternate hours, a five minute continuous focal sample was video
recorded for later coding of the observed behaviors using the same ethogram for cats in
cages (see Table 3.4). Two cats were recorded simultaneously for five minutes for a
total of ten replicates per observation hour. Video cameras were placed on tri-pods, one
recording a cat housed in an upper cage and one recording a cat housed in a lower
cage, while the researcher left the housing room to minimize observer effects on the
cats’ behavior.
After the last scan sample collection on day two, all cat cage doors were covered
and a three- step stranger approach test was conducted on all cats beginning
approximately 30 minutes after doors were covered (Appendix D). Cats were tested in a
randomized order. Each cat cage was uncovered for the test immediately prior to
commencement of step one. One male unfamiliar to the cats served as the stranger
throughout the study. Data were recorded live as well as video recorded for further
analysis. During step one, the stranger stood quietly one meter from the cage for 30
seconds. Next, the stranger took a step closer and placed his hand on the cage door
and stood quietly for 30 seconds. In the last step, the stranger opened the cage door
and stood quietly with his hand in the cage for 30 seconds. Cage doors were re-covered
immediately after the completion of step three. Latency to interact, duration of
interaction, and a sociability score (1-5) was recorded for each cat at each step.
2.6 Statistical Analysis:
Cage condition: Food intake, urination, and defecation outcomes were analyzed
by generalized linear mixed models. The models included fixed effects of treatment,
day, the interaction of treatment and day, and subject as a random effect (modeled as a
random intercept). Simple effect comparisons of treatment by day using least squares
means was conducted on these parameters when indicated. Analyses were performed
66
in SAS 9.2g using the GLIMMIX procedure (Stroup, 2011). A second analysis was
performed to assess the effect of the macro environment. Treatment (M+) grouped
M+m+ and M+m- and treatment (M-) grouped M-m+ and M-m-. In this analysis, M+
contained 36 cats and M– contained 40 cats. Finally, a third analysis was done to
assess the effect of the micro environment. Treatment (m+) grouped M+m+ and M-m+
and treatment (m–) grouped M+m- and M-m-. In this analysis, m+ and m- each
contained 38 cats.
Comparisons of the cats’ cage use in all four treatment groups were analyzed
using Pearson’s chi square tests on the frequency of the three outcomes on day one and
day two using STATA 11i. This analysis was repeated to assess the effect of macro and
micro environments grouping treatments as described above. Finally, a Wilcoxon sign
rank test of each treatment group comparing the average number of cats that exhibited
normal cage use on day one to those that did on day two was performed using
GraphPad Prism 5h to assess the change in the number of cats exhibiting normal cage
use.
For comparison with earlier work (Stella et al., 2011; 2013), decreased appetite,
no eliminations for 24 hours, eliminating out of the litter pan, upper gastrointestinal (UGI)
and lower gastrointestinal (LGI) signs were dichotomized as either present or absent and
summed as total sickness behavior. Wilcoxon sign rank test of each treatment group
comparing day one to day two was performed using GraphPad Prism 5h.
Scan sample behavior data: Scan sampled behaviors were grouped for
analysis based on their association with affect or subjective emotional states.
Maintenance and affiliative behaviors comprised one group, reflecting positive affect,
whereas agonistic, vigilance and avoidant behaviors comprised the second group,
reflecting negative affect or distress. Maintenance and affiliative behaviors comprised
one group reflecting positive affect, whereas agonistic, vigilance and avoidant behaviors
comprised the second group reflecting distress. Data were dichotomized as either
present or absent and Pearson’s chi square tests were performed on the frequency data
at each sampling point using STATA 11i. Statistical significance was set at the level of
p=0.006 to account for multiple comparisons (p=0.05/9 sample points or comparisons;
p=0.006). This analysis was repeated to assess the effect of macro and micro
environments grouping treatments as described above.
67
Analysis of the scan samples of position in cage was complicated by the
presence of structural zeros in the data set meaning the cats in treatments M+m- and Mm- only had three choices for position in cage due to absence of hiding and perching
opportunities. Therefore, only hiding and perching behavior in the enriched cage groups
was further analyzed. Pearson’s chi square tests on the number of cats in the hide box
at each scan sample point were performed comparing treatment M+m+ to treatment Mm+ using STATA 11i. Additionally, an average score was calculated for each cat on day
one and day two by calculating the average number of scan samples the cat was
observed in the hide box. Wilcoxon sign rank tests of treatment M+m+ and treatment Mm+ comparing day one scores to day two scores and comparing the two treatments on
day one and on day two were conducted. This analysis was repeated with the perching
data.
Analysis of vocalizations was complicated due to small frequency counts in each
cell. A summary of the number and percentage of cats vocalizing at each time point is
presented. Pearson’s chi square tests were performed on the vocalization data at each
sampling point using STATA 11i. Statistical significance was at the level of p=0.006 to
correct for multiple comparisons. Additionally, Pearson’s chi square tests of the number
of meowing or hissing/growling at each scan sample point were performed comparing
the macro environments (M+ to M-). An average score was calculated for each cat on
day one and day two by calculating the average number of samples the cat was
observed meowing and hissing/growling. Wilcoxon sign rank tests of M+ (treatments
M+m+ & M=m-) and M- (treatments M-m+ & M-m-) comparing day one scores to day
two scores were conducted, as were comparisons of the two treatments on day one and
on day two.
Focal sample behavior data: Data were collected on the frequency and/or
duration of the behaviors of the ethogram. Data was summarized and behaviors that
were not exhibited in at least 5% of the samples were removed from further analysis.
The remaining frequency behaviors were approach, rub, tail-up, alert, eat/drink, groom,
rest, stretch, yawn, lip lick, turn away, and startle. These behaviors were dichotomized
as either present or absent and analyzed with a Pearson’s chi square test at each
sampling point using STATA 11i. Statistical significance was set at the level of p=0.006
to account for multiple comparisons. The remaining duration behaviors were tail-up,
alert, groom, rest, crouch, freeze, turn away, hide or attempt to hide, dilated pupils,
68
tense, and increased respiratory rate. Dilated pupils, tense and increased respiratory
rate were removed from further analysis due to the inability to quantitatively assess
these parameters. The remaining behaviors were analyzed using two way repeated
measures ANOVA in GraphPad Prism 5h.
Stranger approach data: Latency to interact, duration of interaction and the
mean approach score at each of the three steps of the approach test was analyzed
separately by one way ANOVA and unpaired t-tests with GraphPad Prism 5h.
3. Results:
3.1 Cage Condition: Results of the generalized linear mixed model are
presented in table 4.1.
3.1.1 Food Intake: The percentage of cats in all treatment groups that ate more
than 50% of the offered food is depicted in Figure 4.1. On day one, 20% or fewer of the
cats in each treatment group ate more than half of the offered food, whereas on day two
at least 40% of cats in all groups ate more than half of the offered food. No significant
effect of treatment (p= 0.7) was identified, but the effect of day was statistically
significant (p= 0.0009). Further analysis did not identify any significant effects of the
individual comparisons.
When considering the macro environment, a significant effect of treatment was
not identified (p= 0.3) but an effect of day was found (p= 0.0005) (figure 4.1). No other
significant differences were identified.
When considering the micro environment, no significant effect of treatment (p=
0.9) was found but the effect of day was significant (p= 0.0005) (figure 4.1). Further
analysis did not identify any other significant results.
69
Food Intake
Effect
Num df
Dem df
F-value
p-value
All treatments
Tx
3
55
0.54
0.7
Day
1
55
12.38
0.0009
Tx*Day
3
55
0.43
0.7
Tx
1
57
1.06
0.3
Day
1
57
13.71
0.0005
Tx*Day
1
57
0.5
0.5
Tx
1
57
0.00
0.96
Day
1
57
13.83
0.0005
Tx*Day
1
57
1.14
0.3
Tx
3
55
0.02
0.98
Day
1
55
19.99
<0.0001
Tx*Day
3
55
1.92
0.14
Tx
1
57
0.54
0.6
Day
1
57
18.15
<0.0001
Tx*Day
1
57
5.14
0.02
Tx
1
57
0.57
0.4
Day
1
57
20.93
<0.0001
Tx*Day
1
57
1.39
0.3
Tx
3
55
0.22
0.9
Day
1
55
9.82
0.002
Tx*Day
3
55
0.87
0.5
Tx
1
57
0.57
0.4
Day
1
57
9.99
0.002
Tx*Day
1
57
1.37
0.3
Tx
1
57
0.04
0.8
Day
1
57
9.65
0.002
Tx*Day
1
57
1.44
0.2
Macro Environment
Micro Environment
Urination
All treatments
Macro Environment
Micro Environment
Bowel Movement
All treatments
Macro Environment
Micro Environment
Table 4.1 Results of Generalized Linear Mixed Model Study Two
3.1.2 Urination: The number of cats in each treatment that did not urinate during
the study period and those that urinated out of the litter pan is shown in Appendix G,
table G.1. The number of cats that urinated out of the litter pan was too few to analyze.
Cats that did not urinate were compared to those that eliminated in the litter pan. The
percentage of cats that urinated in the litter pan is depicted in Figure 4.2. The effect of
70
treatment was not significant (p=0.9) but the effect of day was significant (p<0.0001)
meaning more cats in all treatment groups urinated in the litter pan on day two than on
day one. Further comparisons did not identify any other significant results.
Analysis of the macro environment showed that the effect of treatment was not
significant (p=0.6) but the effect of day was (p<0.0001) (figure 4.2). Further
comparisons did not identify any other significant results.
Analysis of the micro environment showed that the effect of treatment was not
significant (p=0.4) but that day was (p<0.0001) (figure 4.2). Further comparisons did not
identify any other significant results.
3.1.3 Bowel Movement (BM): The number of cats in each treatment that did not
defecate during the study period and those that defecated out of the litter pan is shown
in Appendix G, table G.2. The number of cats that had a BM out of the litter pan was too
few to analyze. Those that did not have a BM were compared to those that eliminated in
the litter pan. The percentage of cats that had a BM in the litter pan is depicted Figure
4.3. The effect of treatment was not significant (p=0.9), but the effect of day was
(p=0.002) meaning more cats in all treatment groups had a BM in the litter pan on day
two than on day one. Further comparisons did not identify any other significant results.
Analysis of the macro environment showed the effect of treatment was not
significant (p=0.4) but that day was (p=0.002) (figure 4.3). Further comparisons did not
identify any other significant results.
Analysis of the micro environment also showed no significant effect of treatment
(p=0.8) but the effect of day was significant (p=0.002) (figure 4.3). Further comparisons
did not identify any other significant results.
3.1.4 Cage Use: Analysis of how cats used the cage (none, normal, disrupted)
identified no statistically significant results for comparisons of all treatments, macro or
micro environments. Figures 4.4 illustrate cage use as the percentage of cats exhibiting
each outcome in all treatment groups as well as in the macro and micro environments on
day one and day two. The change in the number of cats that used the cage normally
from day one to day two in each treatment group was statistically significant for
treatments M+m+ (p=0.01) and M+m- (p=0.02) (figure 4.5). The change from day one to
day two in the enriched macro environment (M+) was statistically significant (p=0.0005)
as was the change in both micro environments (m+ p=0.04; m- p=0.005) (figure 4.5).
71
3.1.5 Sickness Behavior: Only six instances of additional sickness behavior
(vomiting and diarrhea) were recorded, which provided too few observations to analyze.
A summary is presented in Appendix G, table G.3. For comparison with earlier work
(Stella et al., 2011; 2013), decreased appetite, no eliminations for 24 hours, eliminating
out of the litter pan, upper gastrointestinal (UGI) and lower gastrointestinal (LGI) signs
were combined. Statistically significant decreases in sickness behavior from day one to
day two were identified for treatment M+m+ (p=0.001), treatment M+m- (p<0.0001) and
treatment M-m- (p=0.003) but not treatment M-m+ (figure 4.6).
Food Intake macro environment
Treatment p=0.3 Day p=0.0005
Food Intake
Treatment P=0.65 Day P=0.0009
60
M+m+ (n=14)
M+m- (n=14)
M-m+ (n=16)
M-m- (n=15)
40
20
0
80
100
M+ (n=28)
percentage of cats
80
M- (n=31)
60
40
20
0
Day 1
Food Intake micro environment
Treatment p=0.96 Day p=0.0005
100
percentage of cats
percentage of cats
100
Day 2
Day 1
80
m+ (n=30)
m- (n=29)
60
40
20
Day 2
0
Day 1
Day 2
Figure 4.1 Food Intake- Percentage of cats that ate more than half the offered food; from
left to right all treatment groups, macro and micro environments
Urinated in litter pan
Treatment P=0.9 Day P<0.0001
Urinated in the litter pan Macro Environment
Treatment P=0.6 Day P<0.0001
100
M+m- (n=14)
M-m+ (n=16)
M-m- (n=15)
percentage of cats
percentage of cats
80
60
40
20
80
100
M+ (n=28)
m+ (n=30)
M- (n=31)
percentage of cats
M+m+ (n=14)
100
60
40
20
0
Day 1
Day 2
Urinated in the litter pan micro environment
Treatment P=0.4 Day P<0.0001
80
m- (n=29)
60
40
20
0
Day 1
Day 2
0
Day 1
Day 2
Figure 4.2 Urinations- Percentage of cats that urinated in the litter pan; from left to right
all treatment groups, macro and micro environments
72
BM in litter pan
Treatment P=0.8 Day P=0.002
BM in the litter pan macro environment
Treatment P=0.4 Day P=0.002
100
M+m- (n=14)
60
M-m+ (n=16)
M-m- (n=15)
100
BM in the litter pan micro environment
Treatment P=0.8 Day P=0.002
100
M+ (n=28)
m+ (n=30)
M- (n=31)
80
40
20
60
40
20
0
Day 1
Day 2
80
percentage of cats
80
percentage of cats
percentage of cats
M+m+ (n=14)
m- (n=29)
60
40
20
0
Day 1
Day 2
0
Day 1
Day 2
Figure 4.3 Defecations- Percentage of cats that defecated in the litter pan; from left to
right all treatment groups, macro and micro environments
Cage Use
Cage Use
60
40
Cage Use
100
None
Normal
Disrupted
80
% of cats
80
None
Normal
Disrupted
80
60
40
20
60
40
20
20
0
0
M+_D1
2
1
M-_D1
M-_D2
m+_D1
m+_D2
m-_D1
m-_D2
Treatment Group and Day
M
-m
M
-m
M+_D2
Treatment Group and Day
-_
D
2
-_
D
1
M
-m
+_
D
2
-m
+_
D
1
D
-_
+m
M
M
D
2
D
-_
M
+m
+_
+m
M
M
+m
+_
D
1
0
Treatment Group and Day
Figure 4.4 Cage use- Percentage of cats that displayed normal cage use; from left to
right all treatment groups, macro and micro environments
Normal Cage Use day 1 vs day 2
Macro Environment
p=0.1
p=0.8
p=0.02
p=0.01
0.5
0.0
-0.5
p=0.0005
1.5
p=0.2
1.0
0.5
0.0
2
-_
D
1
p=0.005
p=0.04
1.0
0.5
0.0
-0.5
M+_D1
M+_D2
M-_D1
Treatment Group and Day
M-_D2
m+_D1
m+_D2
m-_D1
m-_D2
Treatment Group and Day
M
-m
-_
D
2
+_
D
-m
-m
M
M
1
2
-m
M
M
+m
-_
+_
D
D
1
D
D2
-_
+m
M
+_
+m
+_
D1
-0.5
M
+m
M
1.5
average # cats
1.0
Normal Cage Use day 1 vs day 2
Micro Environment
average # cats
Normal Cage Use day 1 vs day 2
1.5
average # cats
% of cats
100
None
Normal
Disrupted
% of cats
100
Treatment Group and Day
Figure 4.5 Change in cage use - The number of cats displaying normal cage use from
day 1 to day 2, mean +/- SD; from left to right all treatment groups, macro and micro
environments
73
Percent of Cats that Exhibited
at least one Sickness Behavior
Sickness behaviors per day
p=0.001
p=0.2
p<0.0001
p=0.003
3
100
percentage of cats
average SB/day
4
2
1
M-m- (n=15)
M+m- (n=14)
60
-_
2
-_
1
M
-m
2
M
-m
1
+_
M
-m
+_
2
M
-m
2
1
M
+m
-_
M
+m
-_
+m
+_
M
M
+m
+_
1
0
M-m+ (n=16)
M+m+ (n=14)
80
40
1
Treatment Group and Day
2
Day
Figure 4.6 Sickness behaviors- Change in the number of sickness behaviors from day 1
to day 2, mean +/- SD (left); percentage of cats exhibiting at least one sickness behavior
day 1 and day 2 (right)
3.2 Scan sample data Analysis
3.2.1 Behavior: Analysis of the number of cats exhibiting affiliative and
maintenance behaviors at each time point revealed that time points 4 (p<0.0001), 5
(p<0.0001), and 9 (p=0.002) were statistically significant (Figure 4.7; Table 4.2).
Analysis of the macro environment revealed that time points 4 (p<0.0001), 5 (p<0.0001),
7 (p=0.001) and 9 (p=0.005) were statistically significant (Figure 4.7; Table 4.3). Finally,
analysis of the micro environment revealed that time point 6 (p=0.006) was statistically
significant (Figure 4.7; Table 4.4).
Treatment
M+m+
M+mM-m+
M-mp-value
Time
1
1
2
2
2
0.937
Time
2
2
7
2
4
0.129
Day 1
Time Time 4
3
7
11
9
11
4
1
8
3
0.167 <0.0001*
Time 5
13
11
2
5
<0.0001*
Time
6
7
12
4
9
0.01
Day 2
Time Time
7
8
13
14
13
13
8
13
9
14
0.01 0.31
Time
9
14
13
8
13
0.002*
N
14
14
16
15
Table 4.2 Scan sampling behavior- Number of cats in each treatment group at each time
point exhibiting affiliative and maintenance behaviors. Statistical significance at p=0.006
74
Treatment
M+
Mp-value
Time
1
3
4
0.8
Time
2
10
6
0.2
Day 1
Time Time 4
3
16
22
12
4
0.2
<0.0001*
Time 5
24
7
<0.0001*
Time
6
19
13
0.05
Day 2
Time
Time
7
8
26
27
17
27
0.001* 0.2
Time
9
27
21
0.005*
N
28
31
Table 4.3 Scan sampling behavior- Number of cats in macro environment at each time
point exhibiting affiliative and maintenance behaviors. Statistical significance at p=0.006
Treatment
micro(+)
micro(-)
P value
Time
1
3
4
0.7
Time
2
5
11
0.07
Time
3
11
17
0.09
Time
4
12
14
0.05
Time
5
15
16
0.07
Time 6
11
21
0.006*
Time
7
21
22
0.06
Time
8
27
27
0.7
Time
9
22
26
0.1
N
30
29
Table 4.4 Scan sampling behavior- Number of cats in micro environment at each time
point exhibiting affiliative and maintenance behaviors. Statistical significance at p=0.006
3.2.2 Position in cage: A summary table and graphs of position in the cage are
presented in Appendix H, figure H.1 and table H.1. Comparisons of the number of cats
in the hide box at each scan sample point were performed comparing treatments M+m+
and M-m+ with no significant effect identified at any time point although time point 6
approached significance (p=0.007) (figure 4.8). Further analysis of treatment M+m+
and treatment M-m+ comparing day one scores to day two scores showed a statistically
significant decrease in score from day one to day two in treatment M+m+ only (p=0.005)
(figure 4.8).
A comparison of the number of cats perching at each scan sample point was
performed comparing treatments M+m+ and M-m+. No statistically significant
differences were identified (figure 4.9). Further analysis of treatment M+m+ and
treatment M-m+ comparing day one scores to day two scores also did not identify a
statistically significant difference between treatment groups or day (figure 4.9).
3.2.3 Vocalizations: A summary of the number and percentage of cats
vocalizing at each time point is presented in Appendix H, table H.2. The percentage of
cats that exhibited meowing or hissing/growling are depicted in figures 4.10 (combined
data focal and scan samples). A significant effect of treatment on meowing was
identified at time point 2 (p=0.005) but no significant effect was identified at any time
point for hissing/growling. Analysis comparing the average number of cats vocalized in
the M+ and M- environments on day one and day two (figure 4.11) showed a significant
75
effect in both M+ (p=0.01) and M- (p=0.006) for meow and M+ (p=0.03) for hiss/growl
(figure 4.11). Further comparisons did not identify any other statistically significant
effects for either meow or hiss/growl.
Affiliative and Maintenance Behaviors
Day 1
Day 2
100
M+m+ (n=14)
M=m- (n=14)
M-m+ (n=16)
M-m- (n=15)
80
Affiliative and Maintenance Behaviors Macro Environment
Day 2
Day 1
100
Day 2
Day 1
*
*
*
100
M+ (n=28)
*
m- (n=29)
*
80
*
80
m+ (n=30)
60
40
60
% of cats
% of cats
M- (n=31)
40
60
40
20
20
20
0
1
2
3
4
5
6
7
8
9
0
Sample
0
1
2
3
4
5
6
7
8
9
1
2
3
4
Sample
5
6
7
8
9
Sample
Figure 4.7 Scan sample behaviors- Percentage of cats exhibiting affiliative or
maintenance behaviors, scan sampling; from left to right all treatment groups, macro and
micro environments
Percentage of Cats in Hide Box
Day 1
100
Hide
Day 2
1.5
M+m+ (n=14)
average hide sore
M-m+ (n=16)
80
P=0.007
*
% of cats
% of cats
Affiliative and Maintenance Behaviors Micro Enviornment
*
*
60
40
20
0
1
2
3
4
5
6
7
8
P=0.005
P=0.9
1.0
0.5
0.0
9
M+m+_1 M+m+_2
Sample
M-m+_1
M-m+_2
Treatment Group and Day
Figure 4.8 Hide box use- Percentage of cats in the hide box (left); change in number of
cats in the hide box day 1 to day 2, mean +/- SD (right), scan sampling
76
Percentage of Cats Perching
Perch
1.5
Day 2
Day 1
average perch sore
100
% of cats
80
60
40
M-m+ (n=16)
M+m+ (n=14)
20
P=0.2
P=0.3
1.0
0.5
0.0
-0.5
M+m+_1 M+m+_2
0
1
2
3
4
5
6
7
8
9
M-m+_1
M-m+_2
Treatment Group and Day
Sample
Figure 4.9 Perching- Percentage of cats perching (left); change in number of cats
perching day 1 to day 2, mean +/- SD (right), scan sampling
Meowing
Day 2
50
*
Day 1
40
30
20
M-m- (n=15)
M+m- (n=14)
M+m+ (n=14)
M-m+ (n=16)
10
0
% of cats
% of cats
40
Hissing/growling
Day 1
50
Day 2
30
M-m+ (n=16)
M-m- (n=15)
M+m+ (n=14)
M+m- (n=14)
20
10
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Time Point
Time Point
Figure 4.10 Vocalizations- Percentage of cats meowing (left) and growling/hissing (right),
scan sampling
Hiss/growl
Meow
1.0
P=0.01
average meow score
average meow score
1.0
P=0.006
0.5
0.0
-0.5
M+_1
M+_2
M-_1
P=0.03
0.5
0.0
M+_1
M-_2
P=0.8
M+_2
M-_1
M-_2
Treatment Group and Day
Treatment Group and Day
Figure 4.11 Change in vocalizations- The number of cats meowing from day 1 to day 2
(left) and growling/hissing (right); mean +/- SD, scan sampling
77
3.3 Focal sample data analysis:
3.3.1. Frequency: Results of the analysis of the frequency behaviors approach,
rub, tail-up, alert, eat/drink, groom, rest, stretch, and lip lick, are presented in table 4.5.
Alert was statistically significant at time point one (P=0.006). No other significant
differences were identified. Further analysis to assess the macro and micro
environments did not identify any statistically significant effects at any time point for any
of the above behaviors.
Behavior
Time point
Day 1
Approach
Macro
Micro
Rub
Macro
Micro
Alert
Macro
Micro
Tail-up
Macro
Micro
Eat/drink
Macro
Micro
Groom
Macro
Micro
Stretch
Macro
Micro
Rest
Macro
Micro
Lip Lick
Macro
Micro
1
0.2
0.3
0.08
0.5
0.4
0.3
0.006
0.04
0.06
0.5
0.7
0.2
0.5
0.4
0.3
0.6
0.4
0.6
0.6
0.7
1.0
0.6
0.5
0.5
2
0.2
0.3
0.1
0.5
0.4
0.3
0.2
0.2
0.1
0.8
0.9
0.3
0.04
0.06
0.1
0.08
0.2
0.2
0.1
0.02
0.5
0.7
0.9
0.5
3
0.4
0.3
0.3
0.02
0.2
0.03
0.3
0.6
0.08
0.6
0.9
1.0
0.6
0.4
1.0
0.2
0.06
0.2
0.2
0.06
0.5
Day 2
4
0.4
0.3
0.3
0.07
0.1
0.07
0.5
0.1
1.0
0.2
0.06
0.6
0.8
0.5
0.8
0.1
0.02
0.4
0.4
0.3
0.3
5
0.7
0.9
0.5
0.6
0.8
0.9
0.3
0.6
0.08
1.0
0.9
1.0
0.6
0.9
0.3
0.2
0.9
0.04
0.03
0.6
0.2
0.6
0.4
1.0
1.0
1.0
0.9
6
0.6
0.6
0.7
0.3
0.5
0.08
0.1
0.7
0.07
0.9
0.6
0.7
0.3
0.3
0.2
0.6
0.6
0.3
0.6
0.9
0.6
0.8
0.8
0.6
0.3
0.1
0.5
Table 4.5 Focal sample frequency data.
78
7
0.2
0.9
0.9
0.3
0.5
0.5
0.05
0.05
0.1
0.6
0.4
1.0
0.3
0.2
0.2
0.9
0.9
1.0
0.08
0.3
0.6
0.1
0.1
0.5
0.4
0.1
0.4
8
0.4
0.3
0.3
0.3
0.3
0.3
0.1
0.08
0.3
0.5
0.1
1.0
0.8
0.6
0.6
0.1
0.03
0.8
0.5
0.8
0.02
0.8
0.6
0.4
0.5
0.9
0.3
3.3.2. Duration: Results of the analysis of the duration of the behaviors alert,
freeze and hide or attempt to hide are presented in table 4.6. Analysis of alert behavior
showed that the effect of both time (p<0.0001) and treatment (p=0.01) were significant
as was the interaction (p=0.02) (figure 4.12). Further analysis of the macro environment
showed that treatment (p=0.007), time (p<0.0001) and interaction (p=0.03) were
significant (figure 4.12) while both time (p<0.0001) and treatment (p=0.01) were
significant in the micro environment with the cats in the unenriched cage condition
exhibiting longer durations of alert behavior than those housed in an enriched cage
(figure 4.12).
Analysis of freeze (p<0.0001), groom (p=0.001), and rest (p<0.0001) showed that
time was significant (figures 4.13, 4.14 and 4.15). Further analysis showed that time
was significant in both the macro and micro environments for freeze (p<0.0001) (figure
4.13) and groom (p=0.0009 and p=0.001 respectively) (figure 4.14) while rest showed
that the effect of both time (p<0.0001) and treatment (p=0.02) were significant in the
macro environment and time (p=0.001) was significant in the micro environment (figure
4.15). Analysis of hiding or attempting to hide showed that the effect of both time
(p<0.0001) and treatment (p<0.0001) were significant (figure 4.16). Further analysis
showed that only time was significant in the enriched (p=0.004) and unenriched
(p=0.006) micro environments (figure 4.16). No statistical significant effects of
treatment were identified for duration of tail-up and crouching behaviors.
79
Source of
variation
All
1.2
Freeze
Macro
0.28
All
5.3
micro
0.61
micro
0.37
All
3.2
Groom
Macro
1.3
micro
0.9
All
3.6
Rest
Macro
1.6
micro
0.6
Hide or Attempt to hide
All
E+
E2.2
2.4
2.0
p-value
0.02
0.03
0.8
0.9
0.9
0.8
0.7
0.4
0.6
0.3
0.2
0.8
0.4
0.2
0.5
DF
F
% total
variation
21
1.8
7.2
7
2.2
7.2
7
0.6
6.9
21
0.59
13.2
7
0.42
13.2
7
0.6
13.2
21
0.83
4.5
7
1.1
4.5
7
0.8
4.4
21
1.1
9.8
7
1.8
9.8
7
0.57
9.6
21
1.0
4.04
7
1.5
5.0
7
0.09
6.4
p-value
<0.000
1
7
7.4
8.6
<0.00
01
7
7.2
4.2
<0.00
01
7
6.8
3.7
<0.00
01
7
19.6
0.7
<0.00
01
7
19.7
0.7
<0.00
01
7
19.9
0.01
0.001
0.009
0.001
7
3.5
0.23
<0.00
01
7
8.9
0.02
<0.00
01
7
5.5
18.3
0.006
7
3.6
1.4
<0.00
01
7
9.3
2.5
0.004
7
3.6
1.6
<0.00
01
7
9.3
2.6
7
3.1
0.13
7
3.0
0.57
0.001
0.007
0.01
0.9
0.4
0.9
0.2
0.06
0.4
0.2
0.02
0.8
0.8
0.5
DF
F
% total
variation
3
6.2
25.7
1
7.9
30.2
1
6.8
30.7
3
0.26
47.8
1
0.79
47.8
1
0.02
48.4
3
1.4
21.2
1
3.7
21.5
1
0.58
22.6
3
1.8
26.4
1
5.4
26.4
1
0.04
28.9
<0.00
01
3
9.6
35.1
1
0.8
47.5
1
0.5
32.9
p-value
<0.000
1
55
3.4
<0.00
01
57
3.7
<0.00
01
57
3.7
<0.00
01
55
8.9
<0.00
01
57
8.7
<0.00
01
57
8.9
<0.00
01
55
2.1
<0.00
01
57
2.2
<0.00
01
57
2.2
<0.00
01
55
3.2
<0.00
01
57
3.1
<0.00
01
57
3.1
<0.00
01
55
6.1
<0.00
01
28
7.4
<0.00
01
27
4.0
% total
variation
Interaction
Time
80
Treatment
Subjects
(matching)
Behavior
Alert
Macro
2.2
DF
F
% total
variation
p-value
DF
F
Table 4.6 Focal behavior duration- study two
80
Day 1
300
Alert macro environment
Treatment p=0.007 Time p<0.0001 Interaction p=0.03
Day 2
300
200
M-m+ (n=16)
M+m- (n=14)
M+m+ (n=14)
150
100
Day 1
300
Day 2
250
M-m- (n=15)
seconds
seconds
250
Alert micro environment
Treatment p=0.01 Time p<0.0001
Day 1
Day 2
250
200
M- (n=31)
150
M+ (n=28)
100
seconds
Alert
Treatment p=0.001 Time p<0.0001 Interaction p=0.02
200
m- (n=29)
m+ (n=30)
150
100
50
50
50
0
1
2
3
4
5
6
7
8
0
0
1
Time Point
2
3
4
5
6
7
8
1
2
3
Time Point
4
5
6
7
8
Time Point
Figure 4.12 Focal behavior- Mean duration of time spent alert; from left to right all
treatment groups, macro and micro environments
Freeze
Treatment p=0.9 Time p<0.0001
300
Day 1
Freeze macro environment
Treatment p=0.4 Time p<0.0001
Day 2
300
Day 1
150
M-m+ (n=16)
M+m- (n=14)
M+m+ (n=14)
M-m- (n=15)
100
50
0
1
2
3
4
5
6
7
300
Day 2
250
250
200
200
seconds
200
seconds
seconds
250
Freeze micro environment
Treatment p=0.9 Time p<0.0001
150
100
M- (n=31)
50
8
Time Point
M+ (n=28)
Day 1
Day 2
150
100
m+ (n=30)
m- (n=29)
50
0
0
1
2
3
4
5
6
7
8
1
2
3
Time Point
4
5
6
7
8
Time Point
Figure 4.13 Focal behavior- Mean duration of time spent freezing; from left to right all
treatment groups, macro and micro environment
Groom
Treatment p=0.2 Time p=0.001
60
Day 1
Groom macro environment
Treatment p=0.06 Time p=0.0009
Day 2
60
Day 1
M+m+ (n=14)
50
Groom micro environment
Treatment p=0.4 Time p=0.001
60
Day 2
Day 1
Day 2
20
M+m- (n=14)
10
M-m+ (n=16)
M-m- (n=15)
0
1
2
3
4
5
Time Point
6
7
8
40
M+ (n=28)
20
M- (n=31)
seconds
30
seconds
seconds
50
40
40
30
m+ (n=30)
20
10
m- (n=29)
0
1
2
3
4
5
Time Point
6
7
8
0
1
2
3
4
5
6
7
8
Time Point
Figure 4.14 Focal behavior- Mean duration of time spent grooming; from left to right all
treatment groups, macro and micro environments
81
Rest
Treatment p=0.2 Time p<0.0001
300
Day 1
Rest macro environment
Treatment p=0.02 Time p<0.0001
Day 2
300
Day 1
300
Day 2
250
250
200
200
Day 1
Day 2
150
M+m- (n=14)
M+m+ (n=14)
M-m+ (n=16)
M-m- (n=15)
100
50
0
1
2
3
4
5
6
7
150
M+ (n=28)
M- (n=31)
100
50
8
Time Point
seconds
200
seconds
seconds
250
Rest micro environment
Treatment p=0.8 Time p<0.0001
150
m+ (n=30)
m- (n=29)
100
50
0
0
1
2
3
4
5
6
7
8
1
2
3
Time Point
4
5
6
7
8
Time Point
Figure 4.15 Focal behavior- Mean duration of time spent resting; from left to right all
treatment groups, macro and micro environments
Hide or attempt to hide
Treatment andTime p<0.0001
300
Day 1
Hide or attempt to hide enriched micro environment
Time p=0.004
300
Day 2
Day 1
300
Day 2
M+m+ (n=14)
M-m+ (n=16)
150
100
50
M+m- (n=14)
M-m- (n=15)
0
1
2
3
4
5
6
7
8
Time Point
200
M+m+ (n=14)
M-m+ (n=16)
150
100
seconds
200
Day 1
Day 2
250
250
seconds
seconds
250
Hide or attempt to hide unenriched micro environments
Time p=0.006
200
150
100
M+m- (n=14)
M-m- (n=15)
50
50
0
1
0
1
2
3
4
5
6
7
2
3
8
4
5
6
7
8
Time Point
Time Point
Figure 4.16 Focal behavior- Mean duration of time spent hiding or attempting to hide;
from left to right all treatment groups, macro and micro environments
3.4 Stranger Approach Test:
3.4.1 Latency to interact: Analysis of the four treatment groups did not identify a
statistically significant effect of treatment for latency to interact at any step (figure 4.17).
Additionally, when assessing the effect of the macro and micro environments no
statistically significant differences were identified on latency to interact at any step (figure
4.18).
3.4.2 Duration of interaction: Analysis of the four treatment groups did not
identify a statistically significant effect of treatment for latency to interact at any step
(figure 4.19). Additionally, when assessing the effect of the macro and micro
environments no statistically significant differences were found for latency to interact at
any step (figure 4.20).
82
3.4.3 Approach score: Analysis of the four treatment groups did not identify a
statistically significant effect of treatment for latency to interact at any step (figure 4.21).
Additionally, when assessing the effect of the macro and micro environments no
statistically significant differences were identified on latency to interact at any step (figure
4.22).
Latency to Interact 1
p=0.3
40
40
Latency to Interact 3
P=0.9
30
30
20
10
Time (seconds)
30
Time (seconds)
Time (seconds)
Latency to Interact 2
P=0.7
40
20
20
10
10
0
M+m+
M+m-
M-m+
0
M-m-
M+m+
0
Treatment Group
M+m+
M+m-
M-m+
M+m-
M-m-
M-m+
M-m-
Treatment Group
Treatment Group
Figure 4.17 Latency to interact step 1, 2 and 3, mean +/- SD, all treatment groups
Latency to Interact 1
Latency to Interact 2
40
Latency to Interact 3
40
40
P=0.2
P=0.3
P=0.4
P=0.8
30
Time (seconds)
Time (seconds)
Time (seconds)
20
P=0.9
P=0.9
30
30
20
10
20
10
10
0
M+
0
M+
M-
m+
M-
m+
m-
0
Treatment Group
m-
Macro +
Macro -
Micro +
Micro -
Treatment Group
Treatment Group
Figure 4.18 Latency to interact step 1, 2 and 3, mean +/- SD; macro and micro
environments
Duration of Interaction 1
P=0.3
40
30
10
30
Time (seconds)
20
20
10
0
M+m+
M+m-
M-m+
Treatment Group
M-m-
Duration of Interaction 3
P=0.9
40
30
Time (seconds)
Time (seconds)
Duration of Interaction 2
P=0.3
40
20
10
0
M+m+
M+m-
M-m+
Treatment Group
M-m-
0
M+m+
M+m-
M-m+
M-m-
Treatment Group
Figure 4.19 Duration of interaction step 1, 2 and 3, mean +/- SD, all treatment groups
83
Duration of Interaction 2
Duration of Interaction 1
40
P=0.2
40
P=0.3
P=0.4
30
P=0.7
P=0.8
30
20
20
10
10
0
0
P=0.7
30
Time (seconds)
Time (seconds)
Time (seconds)
Duration of Interaction 3
40
20
10
M+
M-
m+
m-
M+
M-
Treatment Group
m+
m-
0
Treatment Group
M+
M-
m+
m-
Treatment Group
Figure 4.20 Duration of interaction step 1, 2 and 3, mean +/- SD; macro and micro
environments
Approach Score 1
P=0.4
6
Approach Score 2
P=0.7
6
Approach Score 3
P=0.8
6
5
5
5
4
4
2
Score
3
Score
Score
4
3
3
2
2
1
1
1
0
M+m+
M+m-
M-m+
M-m-
0
Treatment Group
M+m+
0
M+m+
M+m-
M-m+
M+m-
M-m+
M-m-
Treatment Group
M-m-
Treatment Group
Figure 4.21 Mean approach score step 1, 2 and 3, mean +/- SD; all treatment groups
Approach Score 1
P=0.2
P=0.5
5
P=0.6
5
4
P=0.6
P=0.3
2
P=0.4
5
4
Score
3
Approach Score 3
6
4
Score
Score
Approach Score 2
6
6
3
3
2
2
1
1
1
0
0
M+
M-
m+
Treatment Group
m-
M+
0
M+
M-
m+
m-
M-
m+
Treatment Group
Treatment Group
Figure 4.22 Mean approach score step 1, 2 and 3, mean +/- SD; macro and micro
environments
84
m-
4. Discussion
The most important finding of this study is that housing cats in cages that were
roughly double the average size used in shelters and labs (0.55 square meters) did not
seem to have an impact on the behavior of the cats, as predicted. Kessler and Turner
(1999) attempted to measure acclimation to boarding over two weeks and concluded
that qualitative factors other than cage size may influence the degree of stress a cat
experiences. Uetake et al (2012) assessed cage size as well and did not identify any
significant differences in the behavior of the cats in the different size cages and
concluded cats can cope with being housed in small cages for short periods of time.
However, these results were based on a small sample size, the large cages had logs for
enrichment and the singly housed cats were compared to group housed cats living in an
enriched enclosure. Therefore many factors may have influenced the results other then
cage size.
Recently, Tanaka, et al. (2012) tried to identify associations in shelter cats between
changes in body weight, food intake, behavioral stress scores (using the Cat Stress
Score) and incidence of upper respiratory infection. The authors concluded that the use
of small cages (0.53 square meters) was a stressor, and that this along with high
population density, lack of hiding space, frequent movement of cats, and audible dog
barking noise were probably common stressors in other shelters as well (Tanaka et al.,
2012). What has yet to be determined is which, if any, of these common stressors is
most relevant to the cat. The results of the current study suggest that cage size may be
of lesser importance to the cat, particularly in the first few days of confinement, than a
hiding opportunity and an enriched macro (room) environment would be.
Another aspect of cage size to be considered is how the amount of space provided
affects animal activity. Studies have indicated that the level of activity of a cat may
impact its adoptability (Gourkow and Fraser, 2006; Fantuzzi et al., 2010), so the size of
the cage could be important in this regard. What has not been considered in these
studies is what other factors may affect the activity level of the cat. For example, in dairy
cows, factors such as age, diet, foot health, degree of lameness, social groups, flooring
substrate, temporality of daily activities as well as individual differences all have been
found to affect dairy cow activity in addition to the type of housing (Krohn, 1994; Müller
85
and Schrader, 2005; Mattachini et al., 2011). Many of these factors could presumably
also affect cat activity levels and therefore need further study.
As this discussion has highlighted, cage size is only one factor in the well-being of
cats housed singly for extended periods of time. Previous studies by Stella et al. (2011,
2013) have reported that cats could be housed in stainless steel cages with 0.55 square
meters (5.88 square feet) of floor space for many years with few adverse behavioral or
physiological effects. In that research colony, cats lived in enriched cages and room
environments and had daily free play time out of the cages to socialize with conspecifics,
exercise, and explore. This suggests that these forms of enrichment may compensate
for or be more salient to the cat than increased space. Other studies of laboratory cats
have come to similar conclusions. For example, de Monte and Le Pape (1997)
concluded that for adult cats, single housing may not be considered a “totally unfortunate
housing situation”, especially if the cats have daily positive interactions with humans.
This suggests that the overall quality of the environment is important to cat welfare, and
that cage size is only one factor of the environment to be considered.
A final factor to consider in calculating usable space is the shape of the cage. For
example, a study of layer hen housing concluded that vertical cages provided more
usable space than did A-frame style cages, even though A-frame cages were twice as
common in production facilities in the United States (Kiess et al., 2012). Likewise, in the
case of cats, providing elevated resting areas may increase the usable cage space in a
manner that is relevant to the cat (Rochlitz, 2000).
The results of the sickness behaviors exhibited in the present study are in agreement
with previous published results of sickness behavior in response to environmental
disturbances in research cats housed in a laboratory (Stella et al., 2011; 2013), as well
as the results from study one. Between 90-100% of cats in each treatment group in this
study exhibited at least one sickness behavior on day one, with a larger decrease in the
percentage of cats exhibiting sickness behaviors on day two in the cats housed in the
enriched room compared with those housed in the unenriched room. Additionally, there
was a decrease in the average number of total sickness behaviors from day one to day
two in all treatment groups except M-m+. This study did not identify any statistically
significant differences between the treatment groups, possibly due to the small sample
size, but the results did identify significant differences between days in measures of food
intake, urination, defecation, cage use and total sickness behavior. One difference
86
identified in this study is that a slightly larger percentage of cats housed in the
unenriched room environment ate more than 50% of the offered food, eliminated in the
litter pan (both urine and BM), and used the cage normally on day one compared to the
cats housed in the enriched room environment whereas the reverse pattern was seen for
the micro environment. This suggests that in this cohort of cats both an enriched room
and cage may have been of similar importance. Alternatively, this could have been an
anomaly as the differences were not statistically significant.
When looking at how the cats used their cages, disrupted use was exhibited by a
larger percentage of cats than was seen in study one. More cats disrupted their cages
on day two than on day one, and this was seen more often in cats housed in the
unenriched environments, particularly the unenriched room. As in study one, this
suggests that as cats acclimate they actually may be more active in attempts to hide or
to explore, so that cage disruption relates more to available resources and
environmental factors than to socialization to humans.
Secondly, the average number of observations during scan sampling of cats in the
hide box decreased for the group housed in the enriched room, but no significant
decrease was seen in the group housed in the unenriched room. Further, there was a
decrease from day one to day two in the amount of time the cats spent in the hide box,
possibly indicating a degree of acclimation to the novel environment. However, an
average of 200 of the 300 seconds of observation time at each time point was spent in
the hide box, which reinforces the idea that cats need to hide and should be provided the
opportunity to do so. Additionally, there was a trend towards an increasing number of
cats perching on day two compared to day one. Future research should observe cats for
longer periods of time to determine if cats’ preferences for hiding and perching change
with acclimation to a novel environment or time spent in confinement. This is important
to understand for cats housed in cages for extended periods of time.
The results of the scan sampling of behavior showed a higher percentage of cats
exhibiting affiliative or maintenance behaviors in the enriched environments, with the
effect of the room appearing to be greater than that of the cage. This suggests that cats
housed in enriched environments may find it easier to cope with confinement housing.
In this study, analysis of the focal sampling data found few behaviors of statistical
significance across treatment groups; five measures of the duration of observation time
exhibiting alert, freeze, rest, groom, and hide or attempt to hide (discussed above). A
87
statistically significant difference between treatments was identified for time spent
exhibiting alert behavior that appeared to be affected by both the micro and macro
environments, with a greater percentage of cats housed in the unenriched conditions
exhibiting the behavior for a longer duration of the observations. As previously
discussed, alert behavior is an expression of vigilance that has been found to be a
reliable measure of fear in animals so this measure may provide information about the
degree of fear the individual is experiencing (Welp et al., 2004). Similarly, the duration
of observations spent exhibiting freezing behavior declined from day one to day two in all
groups. Together these results suggest that vigilance behavior may decrease in cats
that are able to cope with their environment.
Resting behavior appeared to be affected most by the macro environment but all
groups exhibited an increase in time spent resting from day one to day two. One study
of cats in homes found they spent on average 7.5 hours at night and 7.6 hours during
the day resting (Heidenberger, 1997). This suggests that an increase in the amount of
time cats spend resting may be an indicator that the cat is coping well with the
environment.
Finally, there was a significant effect of the macro environment on the amount of time
spent engaged in grooming behavior; cats in the enriched room environment spent more
time grooming. Presence of grooming behavior suggests a positive emotional state and
along with the absence of behaviors indicative of negative emotional states may be an
important indicator of good welfare. Alternatively, increased grooming behavior could be
a displacement behavior, but in this study the grooming observed appeared in the
absence of behaviors indicating distress so it seems unlikely this was a displacement
behavior in this case. All of the above results support the assertion that an enriched
macro environment as well as an enriched cage is essential for confined cats to
experience good welfare.
No effect of treatment, macro or micro environment was found on latency to interact,
duration of interaction, or mean approach score at any of the three steps of the stranger
approach test. One reason for these results may be a lack of power due to small sample
sizes so that no differences could be detected, indicating further study with larger groups
may be needed. Alternatively, the stranger approach test in cats may be affected by the
personality, temperament or early socialization of the cat more so than the housing
environment or level of fear the individual is experiencing at the time of the assessment.
88
These results suggest that there is a need for future studies to further explore the
stranger approach test methodology in cats.
All of these results are in general agreement with the results found in study one.
This strengthens the conclusion that the macro environment appears to be at least as
important to the cat as the micro environment and without attention to the macro
environment the welfare of confined cats may never be adequate. Those behaviors that
were not identified as statistically significant in this study but were found to be different in
study one may have been due to a lack of power from the smaller number of cats in
each treatment group.
Some limitations of this study exist. First, the number of subjects was relatively
small and the groups were of unequal sizes. Larger sample sizes may have resulted in
more definitive results. Second, cats were housed for only a short period of time in
cages. Over longer periods of confinement, cage size may be of more relevance to
individuals, especially if they have no free time outside of them. Thirdly, this was a
biased sample population since all cats were recruited from and volunteered by their
owners. In a less socialized population, larger cages might prove to be more relevant to
cats. Lastly, this study did not directly compare the response of cats housed in different
cage sizes to determine the relevance of cage size to confined cats.
In conclusion, the current study suggests that increasing floor space from 0.55 to 1.1
square meters (5.88 to 11.8 square feet), particularly in the first 48 hours of confinement,
may not be as relevant to the cat as the provision of an enriched macro environment and
an opportunity to hide in the cage. Discussions of animal enclosure space have typically
been focused on meeting minimum standards. These minimum standards do not
represent an extreme in the normal range of housing standards, but rather become the
mean, median and mode of the distribution, and often only address aspects of the
physical space, saying nothing about the quality of that space (Young, 2003). In
resource-poor environments, attention to environmental factors that can be
inexpensively adjusted to reduce the perception of threat to confined cats may be a
more useful allocation of limited resources than provision of larger cages. Hediger
recognized this in 1964 when he stated, “In reality the quality of the space at the
disposal of the animal is of the greatest importance for its welfare”.
Further, the results of this study support the conclusions drawn from study one,
which suggested that providing an enriched room environment and a hiding opportunity
89
may improve the welfare of confined cats. However, regardless of the housing
environment, cats in all groups did show increases in measures of food intake,
eliminations, and affiliative and maintenance behaviors by the end of the study
suggesting that the majority of cats acclimated to confinement given enough time. This
is an important finding for the management of cats admitted to shelters. It suggests that
prior to making culling decisions cats should be afforded enough time to acclimate, with
a minimum of 48 hours allotted to do so. In addition, a larger cage did not lead to an
increase in performance in the cats in this study, suggesting that other factors may be
more relevant to confined cats. This is important, especially for facilities such as
humane organizations that typically do not have the resources to replace current cages.
Studies one and two demonstrate some environmental enrichment that could be cost
effective alternative interventions that may improve cat welfare,
Future research should to be directed toward determining the minimum space
requirements for caged cats and how this is affected by length of stay, age of the cat and
quality of the space provided.
90
Chapter 5
Long Term Memory in the Domestic Cat (Felis sylvestris catus):
Does Housing Environment Affect Behavior during Re-exposure?
1. Introduction
The importance of an enriched environment to the ability of cats to cope with a
confinement experience has been demonstrated in studies one and two. Cats housed in
enriched macro or room environments exhibited behaviors that indicated they acclimated
more quickly than did cats housed in unenriched room environments. This suggests that
an enriched macro environment may be as important to the cat as an enriched micro or
cage environment, and that attention to the macro environment may be beneficial to the
welfare of confined cats.
One area of research that remains relatively unexplored, however, is the effect of the
cat’s cognitive abilities, specifically its memory, in regard to its responses to the cage
environment. Cognitive processes include perception, learning, memory and decision
making, which permit animals to acquire, process, retain and act on information
(Shettleworth, 2001). Understanding an animal’s cognitive abilities is important in its
management in confinement because of the potential for cognition to impact welfare.
For example, an animal’s ability to remember past experiences, both positive and
negative, could have significant effects on welfare (Nicol, 1996). Relatively unenriched
environments may result in animals, including cats, experiencing psychological distress
and negative emotional states if they are unable to cope with the environment and
experience chronic activation of the stress response system (Dantzer, 2002). Cat
cognitive abilities, particularly memory and its relationship to well-being, have not been
extensively studied. Nevertheless, during the course of a cat’s lifetime behavioral
changes will occur presumably through the process of learning and therefore, learning
91
should be considered in order to assess the ability of the cat to acclimate to and cope
with confinement housing.
Wechsler (1995) has defined coping as, “behavioral responses that aim at reducing
the effect of aversive stimuli on fitness or physiological measures related to fitness”.
Successful coping responses will result in learning. As the animal avoids or modifies
aversive situations this will lead to changes in the animal’s assessment of the situation
and its emotional response to it, altering future behavior. Animals will learn aspects of
the housing environment through information gathering and processing and considerable
importance has been placed on the first few days in a new environment as being critical
for understanding their ability to cope as reported in studies of laying hens and pigs
placed in semi-natural environments (Wood-Gush and Duncan, 1976; Stolba and WoodGush, 1989). This in turn leads to changes in behavior as the animal attempts to
optimize its behavioral responses to increase its fitness, experience positive emotional
states, and avoid suffering (Wechsler and Lea, 2007). Little research has been
conducted in this area in many species, especially cats.
Experimental studies of several farm animal species have found that farm animals
are able to discriminate between individual humans based on past experience. For
example, Miura et al (1996) (as reviewed in Wechsler and Lea, 2007) reported that
weanling pigs regularly handled by one experimenter for two weeks approached the
handler rather than a stranger. Additionally, Munksgaard et al. (2001) found that dairy
cows kept a larger distance between themselves and an aversive handler than from a
gentle handler (Munksgaard et al., 2001). Further, dairy cows also will maintain a larger
flight distance from a person who has minimal contact with them than from a person who
has positive contact with them (Bertenshaw and Rowlinson, 2008). It is likely that cats
that are frequently handled may likewise respond negatively or positively to handling and
confinement as a function of their memory of past experiences.
Therefore, the aim of the current study was to 1) indirectly assess the cat’s long-term
memory of a confinement housing experience by comparing their initial responses to a
controlled confinement experience to their responses one year later, and 2) to determine
whether cats respond consistently to the effects of the environment as a function of their
previous experience. The hypothesis was that cats would remember a previous
confinement experience, so that cats housed in the enriched environments would
92
habituate faster upon re-exposure than would cats housed in the unenriched
environments.
2. Materials and Methods
2.1 Subjects: Owners of all participants from study one (76 cats) were
contacted one year after completion of the first study in which they participated. Thirty
two cats were enrolled to repeat the study. Twenty five cats, 15 female and 10 male,
between 1.5-13 years of age (mean age 5.9 years) completed the study. All cats were
healthy and current on viral rhinotracheitis, calicivirus, panleukopenia, and rabies
vaccines at the time of participation. Informed consent was collected from all owners
who volunteered a cat for the study. Cats were assigned to the same treatment group
they were in during the first year, which was one of four treatment groups that consisted
of combinations of an enriched (M+) or unenriched (M-) macro environment and an
enriched (m+) or unenriched (m-) micro environment (Appendix A, Table A.3).
Participants were admitted to the study between 17:00-20:00 hours on day 0 and placed
in an individual stainless steel cage. The cage door was covered by a towel that had
been sprayed with Feliwaya, a commercially available synthetic cat pheromone, and the
cat was left to acclimate overnight. Behavioral observations were collected for eight
hours a day from 08:00-16:00 on days 1 and 2, after which cats were released to their
owners’ between 17:00-20:00 hours on day 2. The Animal Care and Use Committee of
The Ohio State University and the Clinical Research Advisory Committee of the
Veterinary Medical Center approved all experimental procedures used in this study.
2.2 Macro environment: Cats were housed in the Ohio State University
Veterinary Medical Center (OSUVMC) vivarium. A 14:10 light: dark schedule was
maintained to mimic length of day in Ohio at that time of year (May/June). A mean ± SD
room temperature of 22 ± 1.6°C (72 ± 4°F) was maintained throughout the vivarium.
The room housing the cats had dimensions of 4.57 meters by 4.88 meters and had
cages along three walls (see Figure 3.1a & b) with a small attached ante room used as a
storage and office area. Cats were placed in the same treatment group and cage (upper
or lower) to which they had been assigned in year one with 12 cats studied in the M+
replicate and 13 cats in the M- replicate. The enriched room (treatments M+m+ & M+m-)
was a managed environment with minimal disturbances from people, barking dogs or
other unpredictable noises and events. The husbandry routine was temporally
93
consistent, with cats being provided care at the same time and in the same order each
day to provide the cats some predictability about their environment. Cats in the
unenriched room (treatments M-m+ & M-m-) environment were confronted with multiple,
random, unpredictable disturbances each day that included loud conversations, talk
radio, recordings of barking dogs, and noise from cage doors and equipment being
moved or dropped (Appendix B, table B.1). Routine husbandry occurred at different
times of day, and cats were cared for in a random order each day. All cats were cared
for by a single researcher to ensure consistent handling. Cages were spot cleaned and
the cat was not removed to minimize disruption and ensure consistency.
2.3 Micro environment: Cats were housed in individual stainless steel cages
measuring 70 × 78 × 75 cm (0.55 square meters). The front half of the slatted cage floor
was covered by a matb. The enriched cage (treatments M+m+ & M-m+) contained a two
tiered cardboard Hide, Perch & Goc box that had a lower hiding area (53 x 30 x 22 cm)
with two access openings and an open upper sitting area (53 x 30 x 9 cm), placed in the
left rear corner of the cage. Bedding (84 x 74 cm cage pad folded into quarters) was
provided in both the ‘hide’ and ‘perch’ areas. A plastic litter pan (32 x 22 x 8 cm) was
placed in the right rear cage corner filled with Sani Chipsd litter at a depth of 3 cm. One
cup of dry commercial cat foode, 1.5 oz of commercial canned cat foodf, and water was
provided in separate 0.6L (20 oz) stainless steel bowls (see figure 3.2a). Unenriched
cages (treatments M+m- & M-m-) contained bedding (84 x 74 cm cage pad folded into
quarters) and a plastic litter pan (32 x 22 x 8 cm) with Sani Chipsd litter at a depth of 1.5
cm; no Hide, Perch & Goc box was provided. Food (one cup of dry commercial cat foode
and 1.5 oz of commercial canned cat foodf) was provided in the same 0.6L (20 oz)
stainless steel bowl. Water was provided in a separate bowl (see figure 3.2b), and cage
items were rearranged daily (see figure 3.3).
2.4 Data Collection: All participating owners were asked to complete a
shortened pre-study questionnaire for each cat consisting of information about
demographics, health history, home environment as well as any hospitalizations or
illnesses in the past year (Appendix I). Prior to routine husbandry each day, one
researcher (the same person from year one) stood in front of each cage for 30-60
seconds recording food intake, urination, defecation, cage use, and additional sickness
behaviors (vomiting, diarrhea, eliminating out of the litter pan) for each cat (see Table
94
3.3). During husbandry anything that could not be seen from outside the cage was
recorded (e.g., cached food, vomit, or eliminations out of the litter pan).
Behavioral observations were collected between 08:00-16:00 hours using two
sampling techniques. A scan sample was collected every two hours that included the
cat’s position in the cage, the type of behavior(s) it was exhibiting and vocalizations
based on an ethogram for cats in cages developed based on observations of cats in
cages in shelters, veterinary hospitals and research laboratories (see Table 3.4). The
observer stood quietly in the middle of the housing room and recorded these
parameters. Observation time for each scan sample was approximately three minutes.
On the alternate hours, a five minute continuous focal sample was video
recorded for later coding of the observed behaviors using the same ethogram for cats in
cages (see Table 3.4). Two cats were recorded simultaneously for five minutes with a
total of ten replicates per observation hour. Video cameras were placed on tri-pods, one
recording a cat housed in an upper cage and one recording a cat housed in a lower
cage, while the researcher left the housing room to minimize observer effects on the
cats’ behavior.
After the last scan sample collection on day two, all cat cage doors were covered
and a three- step stranger approach test was conducted on all cats starting
approximately 30 minutes after doors were covered (Appendix B). Cats were tested in a
randomized order. Each cat cage was uncovered for the test immediately prior to
commencement of step one. The same male tester as in year one served as the
stranger throughout the study. Data were recorded live as well as video recorded for
further analysis. During step one, the stranger stood quietly one meter from the cage for
30 seconds. Next, the stranger took a step closer and placed his hand on the cage door
and stood quietly for 30 seconds. In the last step, the stranger opened the cage door
and stood quietly with his hand in the cage for 30 seconds. The cage door was
recovered immediately after completion of step three. Latency to interact, duration of
interaction, and a sociability score (1-5) was recorded for each cat at each step.
2.5 Statistical Analysis: Due to the small sample size (n=25), data were
analyzed only based on the macro environment. The macro environment in the previous
studies accounted for the majority of the differences between groups and was chosen for
this reason. The exception was for analysis of hide and perch, since only cats in the
95
enriched cages had the opportunity to engage in these behaviors. Analyses were
conducted on both year one and year two data for both treatment groups.
Cage condition: Food intake, urination, and defecation outcomes were analyzed
by generalized linear mixed models. The models included fixed effects of treatment,
year, day (by year), the interaction of treatment and day (by year), and subject as a
random effect (modeled as a random intercept). Simple effect comparisons of
treatment by day using least squares means was conducted on these parameters when
indicated. Analyses were performed in SAS 9.2g using the GLIMMIX procedure (Stroup,
2011).
Comparison of the cats’ cage use in the two treatment groups was analyzed
using Pearson’s chi square tests on the frequency of the three outcomes on day one and
day two in year one and year two using STATA 11i. Finally, a Wilcoxon sign rank test of
each treatment group comparing the average number of cats that exhibited normal cage
use on day one to those that did so on day two in year one and year two was performed
using GraphPad Prism 5h to assess the change in the number of cats exhibiting normal
cage use.
For comparison with earlier work (Stella et al., 2011; 2013), decreased appetite,
no eliminations for 24 hours, eliminating out of the litter pan, upper gastrointestinal (UGI)
and lower gastrointestinal (LGI) signs were dichotomized as either present or absent and
summed as total sickness behavior for each cat. Wilcoxon sign rank test of each
treatment group comparing day one to day two on year one and year two was performed
using GraphPad Prism 5h. Additionally, Wilcoxon sign rank test of each treatment group
comparing year one to year two on day one and day two was performed as well.
Scan sample behavior data: Scan sampled behaviors were grouped for
analysis based on the affect of the cats by recording the observable manifestations of a
subjectively experienced emotion such as pain, fear, pleasure or contentment.
Maintenance and affiliative behaviors comprised one group reflecting positive affect,
whereas agonistic, vigilance and avoidant behaviors comprised the second group
reflecting distress. Data were dichotomized as either present or absent and Pearson’s
chi square tests were performed on the frequency data at each sampling point using
STATA 11i. Statistical significance was set at the level of p=0.006 to account for multiple
comparisons (p=0.05/9 sample points or comparisons; p=0.006).
96
Analysis of the scan samples of position in cage was complicated by the
presence of structural zeros in the data set, meaning that the cats in treatments M+mand M-m- only had three choices for position in cage due to absence of hiding and
perching opportunities. Therefore, only hiding and perching behavior in the enriched
cage groups was further analyzed. Pearson’s chi square tests on the number of cats in
the hide box at each scan sample point were performed comparing treatment M+m+
(n=6) to treatment M-m+ (n=8) using STATA 11i. Additionally, an average score was
calculated for each cat on day one and day two by calculating the average number of
scan samples the cat was observed in the hide box. Wilcoxon sign rank tests of
treatment M+m+ and treatment M-m+ comparing day one scores to day two scores and
comparing the two treatments on day one and on day two were conducted in year one
and year two. This analysis was repeated with the perching data.
Analysis of vocalizations was complicated due to small frequency counts in each
cell. A summary of the number and percentage of cats vocalizing at each time point is
presented. Pearson’s chi square tests were performed on the vocalization data at each
sampling point using STATA 11i. Statistical significance was at the level of p=0.006 to
correct for multiple comparisons.
An average score was calculated for each cat on day one and day two by
calculating the average number of samples the cat was observed meowing and
hissing/growling. Wilcoxon sign rank tests of M+ (treatments M+m+ & M+m-) and M(treatments M-m+ & M-m-) comparing day one scores to day two scores were
conducted, as were comparisons of the two treatments on day one and on day two in
both years.
Focal sample behavior data: Data were collected on the frequency and/or
duration of the behaviors of the ethogram. Data was summarized and behaviors that
were not exhibited in at least 5% of the samples were removed from further analysis.
The remaining frequency behaviors were rub, tail-up, alert, eat/drink, groom, rest,
stretch, yawn, and lip lick. These behaviors were dichotomized as either present or
absent and analyzed with a Pearson’s chi square test at each sampling point using
STATA 11e. Statistical significance was set at the level of p=0.006 to account for
multiple comparisons. The remaining duration behaviors were tail-up, alert, groom, rest,
crouch, stay, hide or attempt to hide, dilated pupils, tense, and increased respiratory
rate. Dilated pupils, tense and increased respiratory rate were removed from further
97
analysis due to the inability to quantitatively assess these parameters. The remaining
behaviors were analyzed using two way repeated measures ANOVA in GraphPad Prism
5d.
Stranger approach data: Latency to interact, duration of interaction and the
mean approach score at each of the three steps of the approach test was analyzed
separately by one way ANOVA and unpaired t-tests with GraphPad Prism 5h.
3. Results
3.1 Cage condition: Results of the generalized linear mixed model are
presented in table 5.1.
Food Intake
All treatments
Effect
Num df
Dem df
F-value
p-value
Tx
1
69
7.87
0.007
Year
1
69
0.53
0.5
Day(year)
2
69
6.14
0.004
Tx*Day(year)
3
69
0.52
0.7
Tx
1
69
2.32
0.2
Year
1
69
0.0
0.99
Day(year)
2
69
11.19
<0.0001
Tx*Day(year)
3
69
0.44
0.95
Tx
1
69
0.9
0.3
Year
1
69
0.03
0.8
Day(year)
2
69
6.74
0.002
Tx*Day(year)
3
69
0.5
0.7
Urination
All treatments
Bowel Movement
All treatments
Table 5.1 Results of generalized linear mixed model study three
3.1.1 Food Intake: The percentage of cats in both treatment groups that ate
more than 50% of the offered food is depicted in Figure 5.1. Cats housed in the M+
environment were more likely to eat on day one and day two in year one and year two
than cats housed in the M- environment (treatment p=0.007, day p=0.004). The effect of
year on day was not significant in either treatment.
98
Further analysis revealed that on day two a comparison of treatment groups was
statistically significant in year one (p=0.009) and year two (p=0.02) (Appendix J, Table
J.1). No differences were found for day one in either year one or year two and additional
comparisons found no other significant differences.
3.1.2 Urination: The number of cats in each treatment that did not urinate
during the study period and those that urinated out of the litter pan is shown in Appendix
J, table J.2. The number of cats that urinated out of the litter pan was too few to
analyze. Cats that did not urinate were compared to those that eliminated in the litter
pan. The percentage of cats that urinated in the litter pan is depicted in Figure 5.1.
Treatment was not significant (p=0.1) but day was significant (p<0.0001). The effect of
year on day was not significant in either treatment. Further comparisons did not identify
any other significant results.
3.1.3 Bowel Movement (BM): The number of cats in each treatment that did not
defecate during the study period and those that defecated out of the litter pan is shown
in Appendix J, table J.3. The number of cats that had a BM out of the litter pan was too
few to analyze. Those that did not have a BM were compared to those that eliminated in
the litter pan. The percentage of cats that had a BM in the litter pan is depicted figure
5.1. Treatment was not significant (p=0.3), but day was significant (p=0.002). The effect
of year on day was not significant in either treatment. Further comparisons did not
identify any other significant results.
Percentage of cats that ate
> 50% of offered food
Treatment p=0.007 Day p=0.004
100
M+2
M-2
M+1
M-1
Percentage of cats had a BM in the litter pan
Treatment p=0.3 Day p=0.002
100
M+2
M-2
80
percentage of cats
80
percentage of cats
M+1
M-1
60
40
60
40
20
20
0
0
1
2
80
percentage of cats
100
Percentage of cats urinated in the litter pan
Treatment p=0.2 Day p<0.0001
M+1
M-1
M+2
M-2
60
40
20
0
1
Day
2
Day
1
2
Day
Figure 5.1 Food intake, urinations and defecations- Percentage of cats that ate more
than half of the offered food, urinated or defecated in the litter pan, from left to right.
(M+1= enriched room year one, M+2 = enriched room year two, M-1= unenriched room
year one, M-2 = unenriched room year two)
99
3.1.4 Cage Use: Analysis of how cats used the cage (none, normal, disrupted)
identified no statistically significant results. Depicted in figures 5.2 is how the cats used
their cage presented as the percentage of cats exhibiting each outcome in each
treatment group on day one and day two in both year one and year two. The change in
the number of cats that used the cage normally on day one from year one to year two
and on day two from year one to year two is depicted in figure 5.2. The change in the
number of cats that used the cage normally on from day one to day two in both year one
and year two is depicted in figure 5.2.
Cage Use
100
Normal Cage Use
Normal Cage Use
Day two
Day one
p=0.3
p=0.8
p=0.07
p=0.5
1.5
Disrupted
Normal
None
1.5
Year one
p=0.5
p=0.006
Year Two
p=0.2
p=0.4
1.0
40
20
average # cats
average # cats
1.0
60
0.5
0.5
0.0
0.0
-0.5
-0.5
Treatment Group and Year
-_
D
2
M
2
-_
D
1
M
1
+_
D
M
-_
D
2
+_
D
M
M
2
-_
D
1
M
1
+_
D
+_
D
M
M
-_
Y2
M
Y2
-_
Y1
M
Y1
+_
M
-_
Y2
+_
M
M
Y2
-_
Y1
M
Y1
+_
+_
M
Y2
-_
D
2_
Y2
M
Treatment Group and Day
M
Y2
-_
D
1_
M
M
+_
D
2_
Y2
Y1
D
1_
M
+_
Y1
-_
D
2_
M
Y1
-_
D
1_
M
D
2_
M
+_
D
1_
Y1
0
M
+_
% of cats
80
Treatment Group and Day
Figure 5.2 Cage use- as a percentage of cats in each treatment group on each day
exhibiting the behavior; change in the number of cats using the cage normally from year
one to year two on day 1 and day 2 (mean +/- SD); change in the number of cats using
the cage normally from day one to day in year 1 and year 2 (mean +/- SD). (M+1=
enriched room year one, M+2 = enriched room year two, M-1= unenriched room year
one, M-2 = unenriched room year two)
3.1.5 Sickness Behavior: Only six instances of additional sickness behavior
(vomiting and diarrhea) were recorded, which provided too few observations to analyze.
A summary is presented in Appendix J, Table J.4. For comparison with earlier work,
decreased appetite, no eliminations for 24 hours, eliminating out of the litter pan, UGI
and LGI signs were combined. Statistically significant decreases in sickness behavior
from day one to day two were identified for treatment M+ in year one (p=0.003) and in
year two (p=0.002) and treatment M- in year one (p=0.02) but not in year two (p=0.09)
(figure 5.3). No statistical significance was found for comparisons of day one or two in
year one versus year two for either treatment M+ or M- (figure 5.3). The percentage of
cats exhibiting at least one sickness behavior on day one and day two is depicted in
figure 5.3.
100
3
2
1
Tx 2_Y2
100
2
1
-_
Y2
M
Y2
-_
Y1
M
Y1
+_
M
-_
Y2
+_
M
M
Y2
-_
Y1
M
Y1
+_
+_
M
M
-_
D
2
-_
D
1
M
+_
D
2
+_
D
1
M
M
-_
D
2
M
M
-_
D
1
M
+_
D
2
Percent of Cats that Exhibited
at least one Sickness Behavior
3
0
M
+_
D
1
0
M
Sickness Behavior per day
Day 1
Day 2
p=1.0
p=0.4
p=0.1
p=0.6
4
percentage of cats
Sickness Behaviors per day
Year 1
Year 2
p=0.09
p=0.003 p=0.02
p=0.002
average SB/day
average SB/day
4
Treatment Group and Year
Treatment Group and Day
Tx 2_Y1
80
60
Tx1_Y1
Tx 1_Y2
40
1
2
Day
Figure 5.3 Sickness behaviors- per day comparison by year, mean +/- SD; Sickness
behaviors per day comparison by day, mean +/- SD; Percent of cats exhibiting at least
one sickness behavior.
3.2 Scan sample data analysis
3.2.1 Behavior: Analysis of the number of cats exhibiting affiliative and
maintenance behaviors in treatment M+ and M- in year one and year two revealed
statistical significance at time points one (p<0.0001), two (p=0.004), four (p<0.0001), 5
(p<0.0001), 6 (p<0.0001), and eight (p=0.001) (figure 5.4, tables 5.2-5.6).
Day 1
Time
Time 4
3
9
9
Day 2
Time
Time
7
8
9
10
Treatment
Time 1
M+ Y 1
6
Time
2
7
9
9
M+ Y 2
8
11
12
12
12
12
12
M- Y 1
1
5
15
1
0
3
M- Y 2
0
3
8
12
13
p-value
<0.0001*
0.004*
0.01
<0.0001*
<0.0001*
Time 5
Time
9
10
12
12
12
12
5
5
10
13
3
10
12
13
13
<0.0001*
0.007
0.001*
0.12
Time 6
Table 5.2 Scan sample of behavior- Comparison of treatment M+ vs treatment M- year
one and two. Number of cats in each treatment at each time point exhibiting affiliative
and maintenance behaviors. Statistical significance at p=0.006 level.
101
N
Treatment
M+
Mp-value
Time
1
6
1
0.019
Time
2
6
5
0.305
Day 1
Time
Time
3
4
9
9
5
1
0.066 0.001*
Time 5
9
0
<0.0001*
Day 2
Time
Time
7
8
9
10
5
5
0.066 0.022
Time
6
8
3
0.007
Time
9
10
10
1.0
N
12
13
Table 5.3 Scan sample of behavior- Comparison of treatment M+ vs treatment M- year
one. Number of cats in each treatment at each time point exhibiting affiliative and
maintenance behaviors. Statistical significance at p=0.006 level.
Treatment
Time 1
1 (M+)
2 (M-)
P value
8
0
<0.0001*
Day 1
Time
Time
2
3
11
12
3
8
0.001* 0.016
Time
4
12
12
0.327
Time
5
12
13
Time 6
12
3
<0.0001*
Day 2
Time Time
7
8
12
12
10
12
0.076 0.327
Time
9
12
13
N
12
13
Table 5.4 Scan sample of behavior- Comparison of treatment M+ vs treatment M- year
two. Number of cats in each treatment at each time point exhibiting affiliative and
maintenance behaviors. Statistical significance at p=0.006 level.
Treatment
M+Y 1
M+ Y 2
p-value
Time
1
6
8
0.408
Time
2
7
11
0.059
Day 1
Time
3
9
12
0.064
Time
4
9
12
0.064
Time
5
9
12
0.064
Time
6
9
12
0.064
Day 2
Time
Time
7
8
9
10
12
12
0.064 0.14
Time
9
10
12
0.14
N
12
12
Table 5.5 Scan sample of behavior- Comparison of treatment M+ year one vs. two.
Number of cats in each treatment at each time point exhibiting affiliative and
maintenance behaviors. Statistical significance at p=0.006 level.
Treatmen
t
M- Y 1
Time
1
1
Time
2
5
Day 1
Time Time 4
3
5
1
0
Tim
e6
3
Day 2
Time Time
7
8
5
5
Time
9
10
M- Y 2
0
3
8
12
13
3
10
12
13
p-value
0.30
8
0.39
5
0.23
9
<0.0001
*
<0.0001
*
1.0
0.04
7
0.004
*
0.06
6
Time 5
N
1
3
1
3
Table 5.6 Scan sample of behavior- Comparison of treatment M- year one vs. two.
Number of cats in each treatment at each time point exhibiting affiliative and
maintenance behaviors. Statistical significance at p=0.006 level.
102
Percentage of Cats Exhibiting Affiliative or Maintenance Behavior
Year 1 vs. Year 2
Day 2
Day 1
*
100
*
*
*
*
M+ Y2 (n=12)
M- Y 2 (n=13)
M+ Y1 (n=12)
M- Y 1 (n=13)
% of cats
80
60
*
40
20
0
1
2
3
4
5
6
7
8
9
Sample
Figure 5.4 Scan sample behavior- Percentage of cats exhibiting affiliative
and maintenance behaviors, scan sampling.
3.2.2 Position in cage: Summary tables of position in the cage in year one and
year two are presented in Appendix K (tables K.1 and K.2). Analysis of the number of
cats in the hide box at each scan sample point was performed comparing treatment
M+m+ (n=6) to treatment M-m+ (n=8) for both year one and year two with no significant
effect identified at any time point (figure 5.5). Comparisons of the average number of
cats in the hide box on day one versus day two did not result in statistical significance in
either year one or year two in either treatment group. A comparison of the average
number of cats in the hide by year showed a statistically significant increase (p=0.03)
from year one to year two for the cats housed in the M- environment on day one but not
on day two and no difference was seen for the cats housed in the M+ environment on
either day (figure 5.6).
A comparison of the number of cats perching at each scan sample point was
performed comparing treatments M+m+ and M- m+ in year one and year two. Results
did not identify any significant time points (figure 5.7). Further analysis of treatment
M+m+ and treatment M-m+ did not identify any significant differences (figure 5.8).
103
Percentage of Cats in Hide Box
100
Day 2
Day 1
M- Y 2 (n=8)
% of cats
80
M- Y 1 (n=8)
60
40
M+ Y 2 (n=6)
M+ Y 1 (n=6)
20
0
1
2
3
4
5
6
7
8
9
Sample
Figure 5.5 Hide box use- Percentage of cats in the hide box
Hide
Hide
2
Y
1
M
-D
2
Y
2
2
M
-D
2
Y
D
+
M
M
+
D
2
Y
1
2
Y
1
-D
1
Y
Y
1
D
1
M
+
M
-D
D
1
Y
2
Y
2
0.0
2
Y
2
1
M
-D
2
Y
D
M
+
D
1
Y
2
1
Y
M
+
M
-D
2
Y
1
1
Y
M
-D
2
Y
D
1
M
+
D
M
+
1
0.0
0.2
M
0.2
0.4
2
0.4
0.6
1
0.6
0.8
-D
0.8
Day 2
p=0.2
p=1.0
p=0.03
p=1.0
1.0
M
p=0.6
p=0.5
+
p=0.8
Day 1
M
1.0
Year 2
average # cats in hide
p=0.5
1
average # cats in hide
Year 1
Treatment and year
Treatment and Day
Figure 5.6 Change in hide box use- Change in number of cats who exhibited hiding
behavior from day one to day two by year (left) and from year one to year two by day
(right).
104
Percentage of Cats Perching
80
Day 2
Day 1
% of cats
60
40
M- Y 1 (n=8)
M+ Y 2 (n=6)
20
M- Y 2 (n=8)
M+ Y 1 (n=6)
0
1
2
3
4
5
6
7
8
9
Sample
Figure 5.7 Perching- Percentage of cats perching
Perch
Perch
0.2
2
Y
1
M
-D
2
Y
2
2
M
-D
2
Y
D
+
M
M
+
D
2
Y
1
2
Y
1
Y
Y
1
1
Y
D
1
+
M
2
-D
+
D
2
Y
2
2
Y
1
Sample
M
M
-D
2
Y
D
+
M
M
+
D
1
Y
2
1
Y
1
-D
M
-D
M
2
Y
1
1
Y
1
2
Y
D
1
+
D
M
+
M
1
0.0
0.0
-D
0.2
0.4
M
0.4
0.6
2
0.6
0.8
1
0.8
-D
p=0.9
p=0.5
M
p=0.7
p=0.3
Day 2
p=0.1
p=0.5
p=0.1
p=0.5
1.0
M
average # cats in hide
1.0
Year 2
average # cats in hide
Year 1
Day 1
Sample
Figure 5.8 Change in perching- Change in number of cats who exhibited perching
behavior from day one to day two by year (left) and from year one to year two by day
(right).
3.2.3 Vocalization: A summary of the number and percentage of cats vocalizing
at each time point is presented in Appendix K, table K.3. The percentage of cats that
exhibited meowing or hissing/growling (treatment M+ and M- during year one and two)
are depicted in figures 5.9 and 5.9 (combined data focal and scan samples). A
significant difference for meowing was identified at time points 1(p<0.0001), 2 (p=0.001)
and 3 (p=0.002) but no significant effect was identified at any time point for
hissing/growling. Comparisons of the average number of vocalizations per day in M+
and M- on day one versus day two showed no statistically significant effect for meow in
year one or year two (figure 5.10). Additionally, comparisons of day one in year one and
105
year two and of day two in year one and year two showed no statistically significant
differences for either treatment (5.10). No statistical significance was identified for
hissing/growling in either treatment or year.
Meow
100
Day 1
20
Day 2
*
Day 1
Day 2
15
* *
60
40
M+ Y1 (n=12)
M+ Y2 (n=12)
M- Y1 (n=13)
M- Y2 (n=13)
20
0
1
2
3
4
5
6
7
8
% of cats
80
% of cats
Hiss/growl
10
M+ Y1 (n=12)
M+ Y2 (n=12)
M- Y1 (n=13)
M- Y2 (n=13)
5
0
9 10 11 12 113 14 15 16 17
1
2
3
4
5
6
7
8
Sample
9 10 11 12 113 14 15 16 17
Sample
Figure 5.9 Vocalizations- Percentage of cats who exhibited meowing (left) and
hissing/growling (right).
Meow
Meow
1.0
Year 2
p=0.6
p=0.02
p=0.7
average meow score
Year 1
0.8
p=0.09
0.6
0.4
0.2
0.0
Day 1
0.8
Day 2
p=0.6
p=0.9
p=0.1
p=0.3
0.6
0.4
0.2
0.0
2
-Y
Y
2
1
D
D
2
1
M
M
-Y
M
M
2
1
Y
2
Y
+
M
D
D
1
2
D
2
D
1
M
+
Y
2
M
+
M
+
Y
Y
1
2
2
D
M
-Y
2
M
-Y
D
1
2
2
Y
+
M
M
+
Y
Y
2
D
1
D
D
2
1
D
1
1
M
-
Y
M
-
+
M
+
Y
Y
1
1
D
D
1
2
D
D
1
-0.2
1
-0.2
M
average meow score
1.0
Sample
Sample
Figure 5.10 Change in vocalizations- Change in number of cats who exhibited meowing
from day one to day two by year (left) and change in number of cats who exhibited
meowing from year one to year two by day (right), mean +/- SD.
3.4 Focal sample data analysis:
3.4.1. Frequency: Results of the analysis of the frequency approach, rub, tailup, alert, lip lick, groom, stretch, and yawn are presented in table 5.7. The frequency of
rub was statistically significant at time point five (p=0.003), tail-up at time point one
(p=0.003), and alert (p<0.0001) and rest (p<0.0001) at time point five (figure 5.9). No
other differences were identified.
106
Rub
Tail-up
% of cats
% of cats
Day 2
*
40
M+ Y1 (n=12)
M+ Y2 (n=12)
M- Y2 (n=13)
M- Y1 (n=13)
20
0
1
2
3
4
5
6
7
40
Day 2
Day 1
60
Day 1
60
*
M+ Y1 (n=12)
M+ Y2 (n=12)
M- Y1 (n=13)
M- Y2 (n=13)
20
0
1
8
2
3
4
Alert
Day 1
*
100
7
8
Day 2
Day 1
100
80
M+ Y1 (n=12)
M- Y1 (n=13)
M- Y2 (n=13)
M+ Y2 (n=12)
60
40
% of cats
% of cats
6
Lip Lick
Day 2
80
20
60
40
M+ Y1 (n=12)
M- Y1 (n=13)
M- Y2 (n=13)
M+ Y2 (n=12)
20
0
0
1
2
3
4
5
6
7
8
1
2
3
4
Sample
5
6
7
8
Sample
Rest
Groom
Day 2
Day 1
Day 2
Day 1
100
60
80
M+ Y2 (n=12)
60
*
% of cats
% of cats
5
Sample
Sample
M- Y2 (n=13)
M- Y1 (n=13)
M+ Y1 (n=12)
40
40
M+ Y2 (n=12)
20
M+ Y1 (n=12)
M- Y1 (n=13)
M- Y2 (n-13)
20
0
0
1
2
3
4
5
6
7
8
1
2
3
Sample
Stretch
5
6
7
8
Yawn
Day 2
Day 1
4
Sample
Day 2
Day 1
20
30
10
% of cats
% of cats
15
M+ Y1 (n=12)
M+ Y2 (n=12)
M- Y1 (n=13)
5
M- Y2 (n=13)
0
1
2
3
4
5
6
7
20
M+ Y1 (n=12)
M- Y1 (n=13)
10
M+ Y2 (n=12)
M- Y2 (n=13)
0
8
1
Sample
2
3
4
5
6
7
8
Sample
Figure 5.11 Frequency focal behaviors- Percentage of cats that exhibited rub, tail-up,
alert, lip lick, resting, grooming, stretching, and yawning behaviors in each treatment
group. *denotes statistically significant time point at the level of p=0.006.
107
Behavior
Time point
Day 1
Frequency
Approach
Rub
Alert
Tail-up
Eat/drink
Groom
Stretch
Yawn
Rest
Lip Lick
1
0.5
0.009
0.8
0.003*
0.2
0.08
0.8
0.2
0.6
0.02
2
0.4
0.02
0.1
0.2
0.1
0.4
0.3
0.1
0.03
3
0.09
0.6
0.5
0.02
0.4
0.3
0.08
0.2
0.2
Day 2
4
0.09
0.02
0.4
0.2
0.09
0.02
0.07
0.03
5
0.3
0.003*
<0.0001*
0.009
0.5
0.3
0.4
0.4
<0.0001*
0.008
6
0.09
0.2
0.4
0.09
0.2
0.3
0.6
0.8
0.3
0.09
7
0.09
0.02
0.5
0.07
0.2
0.03
0.07
0.4
1.0
0.8
8
0.4
0.09
0.8
0.09
0.09
0.06
0.8
0.6
0.2
0.1
Table 5.7 Focal sample data
3.4.2. Duration: Results of the analysis of the duration of the behaviors tail-up,
rest, alert, crouch, hide or attempt to hide, and freeze are presented in table 5.8.
Analysis of tail-up showed time (P=0.002) was significant, treatment was not significant
(p=0.07) but the interaction was significant (p=0.0006) (figure 5.12). Analysis of rest
showed treatment (P=0.03), time (p<0.0001) and interaction (p=0.001) were significant
(figure 5.10). Analysis of alert showed treatment (P=0.4) was not significant but time
(p<0.0001) and interaction (p=0.03) were significant (figure 5.12). Analysis of hiding or
attempting to hide showed treatment (p=0.007) but not time (p=0.1) was significant
(figure 5.12) while analysis of freezing behavior showed both treatment (p<0.0001) and
time (p=0.01) were significant (figure 5.12). No significant results were identified for
crouch (figure 5.12).
108
Source of
variation
Behavior
Tail-up
% total
variation
p-value
DF
F
% total
variation
p-value
DF
F
% total
variation
p-value
DF
F
% total
variation
p-value
DF
F
Interaction
Time
Treatment
Subjects
(matching)
Alert
Rest
Crouch
Freeze
10.5
6.95
7.11
5.05
1.9
Hide/attempt
to hide
1.37
0.0006*
21
2.42
4.9
0.03*
21
1.7
6.36
0.005*
21
2.04
11.01
0.4
21
1.08
2.48
0.2
21
1.22
1.39
0.6
21
0.89
0.88
0.002*
7
3.4
2.58
<0.0001*
7
4.7
1.49
<0.0001*
7
9.4
4.86
0.1
7
1.6
2.27
0.01*
7
2.68
34.3
0.1
7
1.7
17.04
0.07
3
2.5
15.92
0.4
3
0.98
23.27
0.03*
3
3.2
23.5
0.1
3
1.9
18.4
<0.0001*
3
13.65
38.53
0.007*
3
4.6
57.04
0.006*
46
1.7
<0.0001*
46
2.6
<0.0001*
46
3.1
0.002*
46
1.8
<0.0001*
46
11.3
<0.0001*
46
16.9
Table 5.8 Focal behavior durations
Day 1
Day 2
40
30
20
M+ Y1 (n=12)
M+ Y2 (n=12)
M- Y2 (n=13)
M- Y1 (n=13)
10
0
2
3
4
5
6
7
200
M+ Y2 (n=12)
150
M- Y2 (n=13)
100
M- Y1 (n=13)
M+ Y1 (n=12)
50
M+ Y2 (n=12)
M- Y2 (n=13)
M- Y1 (n=13)
10
0
Sample
6
7
8
Time (seconds)
Time (seconds)
20
5
50
1
300
Day 2
4
100
2
3
4
5
6
7
2
3
8
4
5
6
7
8
Sample
Freeze
Treatment P<0.0001 Time P=0.01
Hide
Treatment P=0.007 Time P=0.1
30
3
M+ Y1 (n=12)
M- Y1 (n=13)
M- Y2 (n=13)
M+ Y2 (n=12)
150
Sample
M+ Y1 (n=12)
2
200
0
1
40
1
Day 2
250
0
8
Crouch
Treatment P=0.14 Time P=0.14
Day 1
Day 1
300
250
Sample
50
Day 2
Day 1
Day 2
Day 1
300
Day 2
250
200
M- Y2 (n=13)
M- Y1 (n=13)
150
100
M+ Y1 (n=12)
M+ Y2 (n=12)
50
0
1
2
3
4
5
Sample
6
7
8
Time (seconds)
1
Day 1
300
Time (seconds)
Time (seconds)
50
Alert
Treatment P=0.4 Time P<0.0001
Rest
Treatment P=0.03 Time P<0.0001
Time (seconds)
Tail-up
Treatment P=0.07 Time P=0.002
Tx 2 Y 1
250
200
Tx 1 Y 1
150
Tx 2 Y 2
100
50
Tx 1 Y 2
0
1
2
3
4
5
6
7
8
Sample
Figure 5.12 Focal behaviors- Mean duration of time spent engaging in (clockwise from
upper left) tail-up, resting, alert, crouching, hiding or attempting to hide, and freeze
behaviors in each treatment group and year.
109
3.5 Stranger Approach Test:
3.5.1 Latency to interact: Analysis of latency to interact was not significant at
any step comparing year one to year two (figure 5.13) or when comparing M+ to M(data not shown).
Latency to Interact 1
20
10
30
20
p=0.2
p=1.0
40
Time (seconds)
30
p=0.08
p=0.4
40
Time (seconds)
Time (seconds)
Latency to Interact 2
Latency to Interact 2
p=0.9
p=0.2
40
30
20
10
10
0
0
M+ Y1
M+ Y2
M- Y1
M- Y2
M+ Y1
0
M+ Y1
Treatment Group
M+ Y 2
M- Y1
M+ Y2
M- Y2
M- Y1
M- Y2
Treatment Group
Treatment Group
Figure 5.13 Latency to interact step 1, 2 and 3. Mean +/- SD
3.5.2 Duration of interaction: Analysis of duration of interaction was not significant at
step one or three comparing year one to year two (figure 5.14) or when comparing
treatment one to treatment two (data not shown). But in step two treatment group Mshowed a significant (p=0.047) increase in the duration of interaction in year two
compared to year one.
Duration of Interaction 1
20
10
0
30
20
10
0
M+ Y1
M+ Y2
M- Y1
Treatment Group
M- Y2
p=0.2
p=1.0
40
Time (seconds)
30
Duration of Interaction 3
p=0.047
p=0.4
40
Time (seconds)
Time (seconds)
p=0.6
p=0.2
40
Duration of Interaction 2
30
20
10
0
M+ Y1
M+ Y2
M- Y1
Treatment Group
M- Y2
M+ Y1
M+ Y2
M- Y1
Treatment Group
Figure 5.14 Duration of interaction step 1, 2 and 3. Mean +/- SD
3.5.3 Approach score: A comparison of the year one to year two approach
score in each treatment was not significant at any step (figure 5.15). But a comparison
of M+ to M-, identified a statistically significant differences at step one in year two
110
M- Y2
(p=0.03), step two in year one (p=0.04) and year two (p=0.05), while no significant
difference was identified for step three in either year (Figure 5.16).
Approach Score 1
2
0
4
2
0
M+ Y1
M+ Y2
M- Y1
M- Y2
p=0.3
p=0.6
6
Score (1-5)
4
Approach Score 3
p=0.6
p=0.5
6
Score (1-5)
Score (1-5)
6
Approach Score 2
p=0.9
p=0.2
4
2
0
M+ Y1
M+ Y2
Treatment Group
M- Y1
M- Y2
M+ Y1
M+ Y2
Treatment Group
M- Y1
M- Y2
Treatment Group
Figure 5.15 Mean approach score step 1, 2 and 3; within group between year
comparisons, mean +/- SD
Approach Score 1
6
4
2
p=0.05
p=0.04
6
Score (1-5)
p=0.07
Score (1-5)
Score (1-5)
6
Approach Score 3
Approach Score 2
p=0.03
4
2
0
p=0.3
p=0.07
4
2
0
M+_Y1
M-_Y1
M+_Y2
Treatment Group
M-_Y2
M+_Y1
0
M+_Y1
M-_Y1
M+_Y2
M-_Y2
M-_Y1
M+_Y2
M-_Y2
Treatment Group
Treatment Group
Figure 5.16 Mean approach score step 1, 2 and 3, between groups within year
comparison, mean +/- SD
4. Discussion
The results of this study agree with the hypothesis that cats would remember a
previous confinement experience, and that those housed in the enriched environments
would habituate faster upon re-exposure than would cats housed in the unenriched
environments. First, the measure of food intake identified a significant difference
between the treatment groups, with more cats housed in the enriched room environment
consuming more than 50% of the offered food on both days and in both years.
Measures of urination and defecation identified a greater percentage of cats in the
enriched macro environment eliminating on day two in both years but the difference was
not statistically significant. A larger sample size might have revealed a significant
difference.
111
When looking at the total sickness behaviors (decrease food intake, decreased
eliminations, eliminations out of the litter pan, vomiting, diarrhea), the percentage of cats
housed in the enriched macro environment that exhibited sickness behaviors decreased
from 100% on day one in both year one and year two to 60% and 40% on day two, year
one and year two respectively. In contrast, the percentage of cats housed in the
unenriched macro environment that exhibited a sickness behavior decreased from 100%
to 85% in year one and remained at 100% both days in year two. Additionally, the
average number of sickness behaviors exhibited by each cat increased on day two from
year one to year two in the cats housed in the unenriched macro environment, whereas
it decreased in both years in the cats housed in the enriched environment. Finally, cage
use showed similar results with the group housed in the enriched environment exhibiting
an increase in the number of cats showing normal cage use on day one from year one to
year two that was approaching statistical significance. Taken together, these results
suggest that the cats housed in the enriched macro environment acclimated faster than
did those housed in the unenriched macro environment in year one and that the effect
was exaggerated in year two. The cats housed in the unenriched environment showed
little improvement in their behavioral outcomes from year one to year two, which
suggests that they found it equally difficult to cope in that environment both at the time of
the initial exposure as well as during re-exposure one year later.
Results of the scan sampling of behavior identified differences between the
treatment groups in the percentage of cats exhibiting affiliative and maintenance
behaviors in both year one and year two. Starting at the first data point on day one in
year one, cats housed in the enriched environment were more likely to exhibit these
behaviors which increased until close to 80% of the cats were engaging in affiliative and
maintenance behaviors. Subsequently, these behaviors plateaued and remained
consistent until the end of the study. In year two, the cats housed in the enriched room
started at the same level but by time point three all cats were exhibiting affiliative and
maintenance behaviors and continued to do so until the end of the study. In contrast,
the cats housed in the unenriched room showed fewer affiliative and maintenance
behaviors in both years. Although there was improvement from year one to year two,
the cats in the unenriched room did not fully reach the level of the cats in the enriched
room until the last time point in year two. These results suggest that the cats may have
remembered the environment and that those housed in the enriched room were able to
112
habituate more rapidly upon re-exposure than did the cats housed in the unenriched
room. These results also agree with those from study one and two, which demonstrated
the importance of the quality of the macro environment to cats when confined.
Results for the number of cats that were in the hide box showed that fewer cats were
in the hide box and the number decreased from day one to day two in both years for cats
housed in the enriched room, whereas the cats housed in the unenriched room exhibited
more hiding behavior in both year one and year two, compared to the cats in the
enriched room. Most importantly, the comparison of year one to year two showed that
the cats in the unenriched room had an increase in hiding behavior on both day one and
day two, although statistical significance was identified for day one only. This suggests
that the cats housed in the unenriched room may have had a memory of the
confinement experience as well, but that they perceived it to be an aversive or
unpleasant experience, and upon re-exposure they exhibited more avoidance behavior.
These results agree with the conclusions from study one and two, suggesting that it
appears to be important to provide a hiding opportunity for cats; maybe even more
important to cats housed in an unenriched room environment.
Although results from the focal frequency behavior data identified only a few
significant differences between the cats’ year one and two responses, six of the eight
behaviors that were exhibited in more than 5% of the samples were affiliative or
maintenance behaviors. This suggests that the cats acclimated more quickly in year
two, exhibiting more behaviors indicative of positive affect earlier in the confinement
period. Also, of these eight behaviors, there was an increase from year one to year two
for the group housed in the enriched room environment in the frequency of affiliative and
maintenance behaviors (e.g., rub, tail-up, rest, groom, stretch, and yawn) and these
were seen earlier in the observation period in year two. In contrast, cats housed in the
unenriched room did not exhibit these behaviors at a similar frequency in either year one
or year two. Interestingly, both alert and lip licking behavior appeared to be exhibited at
similar frequency in all cats, regardless of housing environment, in both years. This
suggests these behaviors may indicate acute distress or anxiety in cats, but needs
further study.
Examination of focal behavior duration data suggested that some habituation
occurred in the cats housed in both room environments. Both duration of freezing and
crouching behavior decreased for cats housed in both enriched and unenriched room
113
environments in year two compared to year one. In addition, time spent resting
increased in both groups in year two. These results suggest that cats housed in the
unenriched room environment may have experienced some habituation, but to a lesser
extent than what was identified in the group housed in enriched environment.
Although the results of the stranger approach test appear promising, there are
limitations to their interpretation. Steps one and two revealed that the cats in the
enriched room in year two had a shorter latency to interact than in year one and when
compared to cats housed in the unenriched room. Similarly, in steps one and two,
duration of interaction in year two was longer for cats housed in the enriched room than
in year one and when compared to cats housed in the unenriched room. Finally, in
steps one and two the approach score was higher in year two for cats housed in the
enriched room environment than in year one and when compared with the cats housed
in the unenriched room. These results would suggest that the cats may have perceived
the approach of a stranger as less fearful in year two than in year one when housed in
an enriched environment as predicted. However, the “stranger” in the year two test was
the same person that tested the cats in year one. It is possible that the cats
remembered the tester in year two, even though the total amount of time each cat was
exposed to the tester was 90 seconds and there was no exposure to this person in the
time between the two tests. The trend in response was similar for the cats housed in the
unenriched room environment from year one to year two (decreased latency to interact,
longer duration of interaction and higher approach score) which would also seem to
support this conclusion despite no significant differences in the results. The stranger
approach test in cats needs further study and in the future two different unfamiliar people
should be employed for further validation of the results.
The most important finding of this study is the suggestion that cats may form longterm memories of a confinement housing experience that persist for at least one year
after initial exposure. The differences identified here suggest that the cats in the
enriched room environment habituated faster upon re-exposure. Habituation is said to
occur with the waning of response after repeated presentation of an eliciting stimulus. In
this case, cats were exposed to a single confinement of 48 hours duration. Evidence of
habituation was best illustrated in the scan sample results where the pattern of response
was similar in year one and year two with the difference being that in year two the curve
shifted up so that all cats housed in the enriched room were exhibiting affiliative and
114
maintenance behaviors by time point three. The response of the cats in the unenriched
room was not as drastic and was less consistent which suggests more difficulty or more
variation in coping with the environment despite evidence that some acclimation to the
environment may have occurred. The importance of these findings is that the quality of
the environment affected the behavior of the cats both during the initial exposure in year
one with cats in the enriched environment seeming to find it easier to cope with
confinement and, because they appeared to remember this experience, they may have
habituated more quickly upon re-exposure. It has been proposed that animals that are
able to learn based on unique environmental situations to which they are exposed and to
represent and predict events in their environment have an adaptive advantage (Dantzer,
2002). Therefore it is important to consider learning processes to assess an animals’
ability to acclimate to and cope with a given housing system. Successful coping
responses will result in learning, particularly in the first few days after introduction, to a
new housing system, which will then lead to alterations in behavior as information is
acquired and processed by the animal (Wechsler and Lea, 2007).
The results of this study suggest that the cats were distressed upon initial
confinement, so consideration of how activation of the stress response system affects
learning is important. The stress effects on memory are part of a generally adaptive
mechanism that allows the animal to focus on coping with the current stressor and to
form a lasting, easily accessible memory of it for future use.
This study has several limitations. First, the number of cats studied was rather small
and may have resulted in less definitive results than would have been found with a larger
sample size. Secondly, all cats were volunteered by their owners both in year one and
year two, which may have resulted in a biased sample. As discussed above, the
“stranger” in the approach test was the same person in both years, which may have
affected the results of the test in year two. Finally, the backgrounds of the cats involved
in this study varied, as did their previous experiences with confinement housing.
Information was collected pertaining to any major household disturbances, health
problems or hospitalizations in the year between the two studies did not reveal any
confinement experiences that could have been confounding yet this can not be ruled out
completely.
Few studies have been conducted on the cognitive abilities of cats. Those that have
been investigated include working memory through an object permanence test (Fiset
115
and Doré, 2006), spatial coding (Fiset and Doré, 1996), cognitive flexibility in decision
making (Dumas et al., 2006) and quantity discrimination ability (Pisa and Agrillo, 2009).
Considering the paucity of research on the cognitive abilities of cats, the current study
should be viewed as a small step towards assessing how cats learn and modify their
behavior in response to certain environmental factors. More research is necessary to
understand the cat’s cognitive abilities and how in turn this may affect its welfare.
Future research should aim to repeat this study in a larger group of cats. Research
should also aim to determine if cats generalize the confinement experience so that if
placed in a different room or cage they respond in the same way. Additionally, research
aimed at directly studying memory and learning in cats to assess their capabilities may
be useful if applied to environments, housing and human-animal interactions and could
lead to improvements in welfare if implemented.
The findings in this study are important because the results indirectly suggest that
cats may possess the ability to form long-term memories of a confinement experience. If
so, then this is important to cat welfare both in the short term as well as in the long term
as the effects on welfare may be either positive or negative depending on the quality of
the environment. How an animal perceives its environment and the resulting affective
state of the animal has been proposed as the most important aspect of animal welfare
(Dawkins, 1988; Mason, 1993; Mason and Mendl, 1993; Duncan, 2006; Broom, 2010).
Based on this assumption then, providing an environment for cats likely to increase the
probability that they will experience positive affective states is essential for good cat
welfare. Therefore, efforts should be made to provide a consistent, predictable,
environment with minimal disturbances and an opportunity to hide to all confined cats.
116
Chapter 6
Conclusions and Applications
The goal of this dissertation project was to scientifically evaluate aspects of the
environment that a cat confined to a cage may find aversive, have difficulty coping with
and that may, ultimately impact its welfare. The specific aims of the study were 1) to
evaluate the behavior of cats housed in enriched or unenriched macro and micro
environments; 2) to evaluate the behavior and welfare of cats housed in enriched or
unenriched macro (room) environments and enriched or unenriched micro (cage)
environments with space allowance of 1.1 square meters (11.8 square feet) per cat; 3) to
indirectly assess the cat’s long-term memory of a confinement housing experience by
comparing their initial responses to a controlled confinement experience to their
responses one year later and 4) to determine whether cats respond consistently to the
effects of the environment as a function of their previous experience. The implications
and applications of the results of the project are discussed along with proposed future
areas of investigation.
6.1 Study one: environmental factors
Study one assessed aspects of the macro and micro environment that may be
perceived as aversive to confined cats. The results suggest that the majority of cats
experienced negative affect when acutely confined in a novel environment. Cats
appeared to respond adversely to factors in the macro or room environment that they
may have perceived as threatening, and these factors were at least as relevant to the
cats as were factors in the micro or cage environment. A limitation of this study is that
the only macro environmental factor studied was noise in the form of recordings of
barking dogs, loud music and loud conversations, as well as noise associated with
husbandry (e.g., dropping of equipment, loudly closely cage and room doors). These
noises were assumed to be stress inducing stimuli, but other stimuli not easily perceived
by humans that may have been aversive to cats also could have resulted in the
117
observed behaviors. Examples of other stimuli in the environment include pheromones
and odors from unfamiliar conspecifics, ambient temperature, visual contact with
unfamiliar cats, and aspects of the lighting, all of which also need to be studied to
ascertain how they may affect cat welfare.
A second important finding is that the cats that were given an opportunity to hide
did so initially, and use of the hide box decreased more so for those cats housed in the
enriched compared to those housed in the unenriched macro environment. Hiding while
confined in a novel, potentially aversive environment may be a behavior that cats are
highly motivated to do since in the wild hiding would reduce risk and increase fitness.
Therefore cats that cannot perform this behavior may experience a negative emotional
state (Dawkins, 1990). This suggests that cats may use hiding behavior as a way to
cope in an unfamiliar environment and that providing cats with the opportunity to hide
may improve cat welfare. Further, the motivation of cats to hide and perch may change
over time so that the motivation to hide decreases and to perch increases, and that this
may be dependent on the quality of the environment. Results of this study suggest that
cats tended to engage in hiding initially, but more cats engaged in perching behavior on
day two. This is an area that needs further research especially for cats confined for
longer periods of time. Nonetheless, efforts should be made to provide hiding and
perching opportunities for cats in confinement.
These findings have implications for the management of cats in cages. Although
the macro environment is often inadequately considered in cat housing areas, these
results suggest that cat welfare may be compromised until these factors are addressed.
Additionally, cats are often denied access to hiding areas, which also may compromise
welfare particularly in the face of unenriched macro environments.
Future research should aim to study how other aspects of the macro
environment, such as factors pertaining to ambient temperature, lighting and odors,
might affect cat behavior and welfare. In this study the enriched macro environment
provided a predictable schedule, which may have affected the results observed. In the
future, predictability should be assessed separately to determine its relative importance
to confined cats. Also, this study focused on removing potentially aversive factors,
rather than on addition of potentially pleasant environmental factors e.g., classical music,
toys, olfactory enrichment, which may impact the behavior of cats. Finally, the micro
environment was either enriched or unenriched, but which factors were most important
118
(e.g., consistent placement of cage items, aspects of elimination area, hide and perch
opportunities) was not studied. A more refined study of cage factors could also prove
beneficial to the welfare of caged cats.
Another aspect of the environment that was not assessed in this project was the
area of human-animal interactions. In this study, cat handling involved minimal contact
with the caretaker, although cats that solicited attention were interacted with briefly
during husbandry. Some handling that animals experience in laboratories, shelters, and
production facilities has been shown to be perceived as aversive and fear inducing to
them, affecting their welfare (Rushen et al., 1999; Wielebnowski et al., 2002; Waiblinger
et al., 2006; Morgan and Tromborg, 2007). Human beings are perhaps the dominant
feature of many cats’ environment. As repeated interactions between the cat and
human occur, eventually each is able to make predictions about the other’s behavior.
The quality, positive or negative, of the resulting human-animal relationship likely affects
the cat’s quality of life. Moreover, the human mostly determines the number and nature
of interactions and hence the quality of the relationship. Unfortunately, most research on
the human-animal relationship in companion animals to date has focused on the
outcome of the relationship for the owner/caretaker rather than for the animal (Walsh,
2009; Virues-Ortega & Buela-Casal, 2006; Bernstein, 2005). Future research should
aim to determine the effects of neutral versus positive interactions, as well as varying
quantities of interaction on the behavior and welfare of confined cats both for cats
housed for short periods as well as over longer periods of time.
The stranger approach test was one aspect of the present study that was difficult
to interpret. There are several possible explanations for the finding that the cats housed
in the enriched environments did not consistently show the shortest latencies to interact,
longest durations of interaction and highest approach scores as predicted. The first is
that it is possible that cats, as a species, may be less motivated to interact with an
unfamiliar person because they are typically more independent and less gregarious than
livestock, horses or dogs, species that have previous been tested (Hemsworth et al.,
1996; Hausberger et al., 2008; Győri et al., 2010). A second explanation is that the
provision of a hiding and perching opportunity enabled the cats to relax more easily so
that they were less aroused at the time of testing and therefore the latency to interact
was longer as they “woke up”. In this case the longer latency to approach could be
viewed as an indicator of good welfare. Thirdly, the samples sizes were small and the
119
study was of short duration which may have resulted in less robust results. Finally, the
aim of the test was to measure fearfulness/anxiousness in response to the quality of the
environment. What may in actuality have been measured is sociability of the cat, which
is affected by such factors as past experience, socialization during the sensitive period
and genetics, particularly paternal influence as friendliness has been shown to be
inherited mainly from the tom (McCune, 1995; Siegford et al., 2003). For these reasons
this is an area that needs more research in the future.
6.2 Study two: cage size
The aim of study two was to evaluate the behavior and welfare of confined cats
when allocated roughly double the typical amount of floor space. The results suggested
that, at least acutely, a larger cage did not affect the behavior of the cats. The
implications of these results are important in resource-poor facilities such as those often
found in animal shelters. Although others have proposed that cats need more than 1.1
square meters (6 square feet) of floor space to have good welfare, this study, as well as
earlier publications from a laboratory environment (Stella et al., 2011; 2013), did not
support this conclusion. It seems that the quality of the environment is more important to
confined cats than is the cage size.
Beyond the quantity of space provided, the functionality of the space also needs
to be considered. Large enclosures are of little value to the animals if they cannot or will
not make use of the available space. This is an area that has not been well assessed in
cats, but a study of chimpanzees and gorillas in zoos found that much of the three
dimensional space available was not utilized (50% in chimpanzees and 70% for gorillas).
The authors concluded that the selectivity of space utilization reflected the quality rather
than the quantity of space (Ross and Lukas, 2006; Ross et al., 2011). Preferences for
particular elements of an enclosure and the probable uneven distribution of use are
important factors when allocating scarce resources to housing systems in confinement
housing of all species. Understanding these factors and providing functional, usable
space may lead to improvements in animal welfare. Areas requiring future research are
those that aim to determine the minimum space requirements for cats and how length of
confinement affects this parameter. Alternatively, the addition of time out of the cage for
exploration and play as an enrichment should be studied as this may be more relevant to
the cat as well as more cost effective for many shelters and other facilities.
120
6.3 Study three: long term effects of short term housing
The final study of this project aimed to indirectly assess the ability of cats to form
and retrieve long-term memories for previous confinement experiences. The results
suggest that they may form memories and that those previously housed in an enriched
macro environment habituated more quickly upon re-exposure to that environment. The
cats housed in the enriched room environment exhibited behaviors indicative of positive
affect (affiliative and maintenance) for the majority of their confinement in year two.
These results are particularly important for veterinarians providing veterinary care to
cats, and for investigators using cats in biomedical research. They suggest that
providing enriched housing and decreasing perceptions of threat during initial exposure
to confinement may have long term welfare benefits for cats. Lack of preventive
veterinary care for pet cats is a concern in the United States, so these findings may be
useful in combination with low stress handling techniques to improve cats’ experiences
in clinics, which might encourage owners to bring their cats for more regular veterinary
visits. In relation to cats used in research, it has been well documented that distress
from inadequate environments negatively impacts laboratory animals, and often the
results of biomedical research (Poole, 1997; Gaskill et al., 2009; Stella et al., 2011;
2013).
Future research should aim to directly assess the cognitive abilities of cats,
particularly their ability to form long-term memories of salient experiences. It has been
proposed that understanding an animal’s cognitive abilities is important in the
management and training of domestic and zoo species and may have welfare
implications because the cognitive ability of an animal will, to some extent, determine its
ability to suffer when deprived of certain stimuli (Nicol, 1996). Better understanding of
cognitive abilities may be beneficial in terms of husbandry and training. For example,
when humans have expectations that an animal “understands” what is expected, they
are less likely to give appropriate signals resulting in deleterious behavioral changes in
the animal from the lack of consistency and predictability of the human’s behavior that
often manifest as redirected, ambivalent and displacement behaviors, stereotypies and
injurious behaviors (McLean, 2001). Additionally, overestimating the mental ability of
animals may lead to a breakdown in the human-animal bond. For example, behavioral
reasons are commonly cited as the reason for relinquishment of dogs and cats to
121
shelters (Patronek, 1996; Salman, 2000), which can be viewed as a major contemporary
welfare issue.
Finally, the measures of food intake, cage use, eliminations and sickness
behaviors appeared to be among the most reliable measures to assess the well-being of
cats. These metrics are quantitative, easily collected and non-invasive, making their use
applicable in a variety of environments including shelters, veterinary clinics, biomedical
laboratories, and homes. Caretakers need little training in cat behavior to recognize
changes in these metrics, and they can be assessed and recorded quickly. Therefore
these measures could be implemented in cat housing facilities to monitor the welfare of
confined cats.
This study has added to our understanding of the behavior of cats in cages in
response to environmental factors that they may perceive as threatening and the
possibility that cats may form long-term memory for these environments. The results
also indicate areas that need future research to further our understanding of the
behavior of confined cats. Understanding cat behavior and providing the highest quality
environment possible based on research indicating what is relevant to confined cats
should be a goal of cat caretakers. Many welfare scientists, including Dawkins, Mason,
Mendl and Duncan, have argued that how an animal perceives and experiences stimuli
is the most important determinant of its welfare; thus affective states are really the
essence of welfare, and indeed are what gives welfare its moral urgency (Dawkins,
1988; Mason and Mendl, 1993; Duncan, 2006). Cats that are provided an opportunity to
cope with their environment will presumably be less fearful or anxious and experience a
more positive emotional state. This, along with good physical health, could lead to
improvements in the welfare of cats in cages no matter what type of facility houses them,
potentially improving the quality of life for millions of cats each year.
122
References
Adamec, R., Kent, P., Anisman, H., Shallow, T., Merali, Z., 1998. Neural plasticity,
neuropeptides and anxiety in animals-Implications for understaning and treating affective
disorder following traumatic stress in humans. Neurosci. Biobehav. Rev. 23, 301-318.
Adamec, R.E., 1991. Anxious personality in the cat: Its ontogeny and physiology, in:
Carroll, B.J., Barrett, J.E. (Eds.), Psychopathology and the Brain, Raven Press, New York,
pp. 153-168.
Afonso, E., Thulliez, P., Pontier, D., Gilot-Fromont, E., 2007. Toxoplasmosis in prey
species and consequences for prevalence in feral cats: not all prey species are equal.
Parasitology. 134, 1963-1971.
Amat, M., de la Torre, J.L.R., Fatjo, J., Mariotti, V.M., Van Wijk, S., Manteca, X., 2009.
Potential risk factors associated with feline behaviour problems. Appl. Anim. Behav. Sci.
121, 134-139.
Animal Legal Defense Fund, 2012. Animal Hoarding Facts.
AoSV., 2012. Association of Shelter Veterinarians, Guidelines for Standards of Care in
Animal Shelter, www.sheltervet.org
ASPCA, 2012. American Society for the Prevention of Cruelty to Animals Pet Statistic,
www.aspca.org
Baldwin, A.L., Schwartz, G.E., Hopp, D.H., 2007. Are Investigators Aware of
Environmental Noise in Animal Facilities and that this Noise May Affect Experimental
Data? J. Am. Assoc. Lab. Anim. Sci. 46, 45-51.
Bassett, L., Buchanan-Smith, H.M., 2007. Effects of predictability on the welfare of captive
animals. Appl. Anim. Behav. Sci. 102, 223-245.
Beaver, B.V., 1992. Feline Behavior: a guide for veterinarians. C.V. Mosby, St. Louis.
Beerda, B., Schilder, M.B.H., van Hooff, J.A.R.A.M., de Vries, H.W., 1997. Manifestations
of chronic and acute stress in dogs. Appl. Anim. Behav. Sci. 52, 307-319.
Bernstein, P. L. (2005). The Human-Cat Relationship. In: Rochlitz, I. (ed.) The Welfare of
Cats. Dordrecht:Springer.
Bertenshaw, C.E., Rowlinson, P., 2008. Exploring heifers' perception of 'positive'
treatment through their motivation to pursue a retreated human. Anim. Welf. 17, 313-319.
123
Boissy, A., 1995. Fear and Fearfulness in Animals. Q. Rev. Biol. 70, 165-191.
Boissy, A., Manteuffel, G., Jensen, M.B., Moe, R.O., Spruijt, B., Keeling, L.J., Winckler,
C., Forkman, B., Dimitrov, I., Langbein, J., Bakken, M., Veissier, I., Aubert, A., 2007.
Assessment of positive emotions in animals to improve their welfare. Physiol. Behav. 92,
375-397.
Bradshaw, J.W.S., Horsfield, G.F., Allen, J.A., Robinson, I.H., 1999. Feral cats: their role
in the population dynamics of Felis catus. Appl. Anim. Behav. Sci. 65, 273-283.
Bradshaw, J.W.S.and Cameron-Beaumont, C., 2000. The signaling repertoire of the
domestic cat and its undomesticated relatives, in: Turner and Bateson (Eds.), The
Domestic Cat: The biology of its behavior, Cambridge University Press, Cambridge, pp.
67-93.
Broom, D., 1991. Animal welfare: concepts and measurements. J. Anim. Sci. 69, 41674175.
Broom, D., 1996. Animal welfare defined in terms of attempts to cope with the
environment. Acta Agric. Scand. A. Anim. Sci. 27, 22-28.
Broom, D.M. and Johnson, K., 2000. Stress and Animal Welfare. 2nd ed. Kluwer
Academic Publishers, Boston.
Broom, D.M., 1988. The scientific assessment of animal welfare. Appl. Anim. Behav. Sci.
20, 5-19.
Broom, D.M., 2006. Behaviour and welfare in relation to pathology. Appl. Anim. Behav.
Sci. 97, 73-83.
Broom, D.M., 2010. Cognitive ability and awareness in domestic animals and decisions
about obligations to animals. Appl. Anim. Behav. Sci. 126, 1-11.
Calvez, J., Fromentin, G., Nadkarni, N., Darcel, N., Even, P., Tomé, D., Ballet, N.,
Chaumontet, C., 2011. Inhibition of food intake induced by acute stress in rats is due to
satiation effects. Physiol. Behav. 104, 675-683.
Cameron-Beaumont, C., Lowe, S.E., Bradshaw, J.W.S., 2002. Evidence suggesting
preacclimateation to domestication throughout the small Felidae. Biol. J. Linn. Soc. Lond.
75, 361-366.
Cameron-Beaumont, C., Lowe, S.E., Bradshaw, J.W.S., 2002. Evidence suggesting
preacclimateation to domestication throughout the small Felidae. Biol. J. Linn. Soc. Lond.
75, 361-366.
124
Carlstead, K., Brown, J.L., Seidensticker, J., 1993a. Behavioral and adrenocortical
responses to environmental changes in leopard cats (Felis bengalensis). Zoo Biol. 12,
321-331.
Carlstead, K., Brown, J.L., Strawn, W., 1993. Behavioral and physiological correlates of
stress in laboratory cats. Appl. Anim. Behav. Sci. 38, 143-158.
Carlstead, K., Brown, J.L., Strawn, W., 1993b. Behavioral and physiological correlates of
stress in laboratory cats. Appl. Anim. Behav. Sci. 38, 143-158.
Casey, R., Vandenbussche, S., Bradshaw, JWS, Roberts, MA, 2009. Reasons for the
Relinquishment and Return of Domestic Cats (Felis Sylvestris Catus) to Rescue Shelters
in the UK. Anthrozoos. 22, 347-358.
CDC, 2012. Morbidity and Mortality Weekly Report. 60.
Claxton, A.M., 2011. The potential of the human–animal relationship as an environmental
enrichment for the welfare of zoo-housed animals. Appl. Anim. Behav. Sci. 133, 1-10.
Clutton-Brock, J., 1992. The process of domestication. Mamm. Rev. 22, 79-85.
Clutton-Brock, J., 1999. A natural history of domesticated mammals 2nd ed. Cambridge
University Press, Cambridge.
Coleman, G.J., Hemsworth, P.H., Hay, M., 1998. Predicting stockperson behaviour
towards pigs from attitudinal and job-related variables and empathy. Appl. Anim. Behav.
Sci. 58, 63-75.
Coppola, C.L., Enns, R.M., Grandin, T., 2006. Noise in the Animal Shelter Environment:
Building Design and the Effects of Daily Noise Exposure. J. Appl. Anim. Welf. Sci. 9, 1-7.
Curtis, S., What constitutes animal well-being, in: Moberg, G. (Ed.), Animal Stress,
American Physiological Society, Bethesda, MD, pp. 1-11.
Dantzer, R., 2002. Can farm animal welfare be understood without taking into account the
issues of emotion and cognition? J. Anim. Sci. 80, E1-E9.
Dantzer, R., O'Connor, J.C., Freund, G.G., Johnson, R.W., Kelley, K.W., 2008. From
inflammation to sickness and depression: when the immune system subjugates the brain.
Nat. Rev. Neurosci. 9, 46-56.
Davis, H., Taylor, A., 2001. Discrimination between individual humans by domestic fowl
(Gallus gallus domesticus). Br. Poult. Sci. 42, 276-279.
Dawkins, M.S., 1988. Behavioural deprivation: A central problem in animal welfare. Appl.
Anim. Behav. Sci.20, 209-225.
125
Dawkins, M.S., 1990. From an animal's point of view: Motivation, fitness, and animal
welfare. Behav. Brain Sci. 13, 1-9.
de Monte, M., Pape, G.L., 1997. Behavioural Effects of Cage Enrichment in Single-Caged
Adult Cats. Anim. Welf. 6, 53-66.
Désiré, L., Boissy, A., Veissier, I., 2002. Emotions in farm animals: a new approach to
animal welfare in applied ethology. Behav. Processes 60, 165-180.
Driscoll, C.A., Macdonald, D.W., O'Brien, S.J., 2009. From wild animals to domestic pets,
an evolutionary view of domestication. Proc. Natl. Acad. Sci. 106, 9971-9978.
Driscoll, C.A., Menotti-Raymond, M., Roca, A.L., Hupe, K., Johnson, W.E., Geffen, E.,
Harley, E.H., Delibes, M., Pontier, D., Kitchener, A.C., Yamaguchi, N., O'Brien, S.J.,
Macdonald, D.W., 2007. The Near Eastern Origin of Cat Domestication. Science. 317,
519-523.
Dumas, C., St-Louis, B., Routhier, L., 2006. Decision making and interference in the
domestic cat (Felis catus). J. Comp. Psychol. 120, 367-377.
Duncan, I.J.H., 2006. The changing concept of animal sentience. Appl. Anim. Behav. Sci.
100, 11-19.
Dybdall, K., Strasser, R., Katz, T., 2007. Behavioral differences between owner surrender
and stray domestic cats after entering an animal shelter. Appl. Anim. Behav. Sci. 104, 8594.
Eckstein, R.A., Hart, B.L., 2000. The organization and control of grooming in cats. Appl.
Anim. Behav. Sci. 68, 131-140.
Fantuzzi, J.M., Miller, K.A., Weiss, E., 2010. Factors Relevant to Adoption of Cats in an
Animal Shelter. J. Appl. Anim. Welf. Sci. 13, 174-179.
Feaver, J., Mendl, M., Bateson, P., 1986. A method for rating the individual distinctiveness
of domestic cats. Anim. Behav. 34, 1016-1025.
Fiset, S., Doré, F., 2006. Duration of cats' (Felis catus) working memory for disappearing
objects. Anim. Cogn. 9, 62-70.
Fiset, S., Doré, F.Y., 1996. Spatial encoding in domestic cats (Felis catus). J. Exp.
Psychol: Anim. Behav. Proc. 22, 420-437.
Fitzgerald, B. and Turner, DC, 2000. Hunting behaviour of domestic cats and their impact
on prey populations, in: Turner and Bateson (Eds.), The Domestic Cat: The biology of its
behaviour Cambridge University Press, Cambridge, pp. 151-175.
126
Fogsgaard, K.K., Røntved, C.M., Sørensen, P., Herskin, M.S., 2012. Sickness behavior in
dairy cows during Escherichia coli mastitis. J. Dairy Sci. 95, 630-638.
Gaskill, B.N., Rohr, S.A., Pajor, E.A., Lucas, J.R., Garner, J.P., 2009. Some like it hot:
Mouse temperature preferences in laboratory housing. Appl. Anim. Behav. Sci.116, 279285.
Gogoleva, S., Volodina, E., Volodin, I., Kharlamova, A., Trut, L., 2010. The gradual vocal
responses to human-provoked discomfort in farmed silver foxes. Acta ethologica 13, 7585.
Goldstein, L., 2011. 17 reasons the Economic Impact of the Domestic Cat as a a NonNative Species in the U.S. Does Not Cost $17 Billion, straypetadvocacy.org.
Gottlieb, G., Halpern, C.T., 2002. A relational view of causality in normal and abnormal
development. Dev. Psychopathol. 14, 421-435.
Gourkow, N., Fraser, D., 2006. The effect of housing and handling practices on the
welfare, behaviour and selection of domestic cats (Felis sylvestris catus) by adopters in
an animal shelter. Anim. Welf. 15, 371-377.
Graml, C., Niebuhr, K., Waiblinger, S., 2008. Reaction of laying hens to humans in the
home or a novel environment. Appl. Anim. Behav. Sci. 113, 98-109.
Gray, J.A., 1979. Emotionality in male and female rodents: A reply to Archer. Br. J.
Psychol. 70, 425-440.
Győri, B., Gácsi, M., Miklósi, Á., 2010. Friend or foe: Context dependent sensitivity to
human behaviour in dogs. Appl. Anim. Behav. Sci. 128, 69-77.
Hausberger, M., Roche, H., Henry, S., Visser, E.K., 2008. A review of the human-horse
relationship. Appl. Anim. Behav. Sci. 109, 1-24.
Hediger, H., 1964. Wild Animals In Captivity: An outline of the Biology of Zoological
Gardens. Dover Publications, Inc, New York.
Held, S.D.E., Špinka, M., 2011. Animal play and animal welfare. Anim. Behav. 81, 891899.
Hemsworth, P.H., 2007. Ethical stockmanship. Aust. Vet. J. 85, 194-200.
Hemsworth, P.H., Barnett, J.L., Hansen, C., 1987. The influence of inconsistent handling
by humans on the behaviour, growth and corticosteroids of young pigs. Appl. Anim.
Behav. Sci. 17, 245-252.
127
Hemsworth, P.H., Barnett, J.L., Tilbrook, A.J., Hansen, C., 1989. The effects of handling
by humans at calving and during milking on the behaviour and milk cortisol concentrations
of primiparous dairy cows. Appl. Anim. Behav. Sci. 22, 313-326.
Hemsworth, P.H., Verge, J., Coleman, G.J., 1996. Conditioned approach-avoidance
responses to humans: the ability of pigs to associate feeding and aversive social
experiences in the presence of humans with humans. Appl. Anim. Behav. Sci. 50, 71-82.
HSUS, Humane Society of the United States, 2012, www.humanesociety.org
Inglis, I.R., 2000. Review: The Central Role of Uncertainty Reduction in Determining
Behaviour. Behaviour. 137, 1567-1599.
Jensen, P., Toates, F.M., 1997. Stress as a state of motivational systems. Appl. Anim.
Behav. Sci. 53, 145-156.
Jongman, E.C., 2007. Acclimateation of domestic cats to confinement. J. Vet. Behav. 2,
193-196.
Kass, P.H., New, J.C., Scarlett, J.M., Salman, M.D., 2001. Understanding Animal
Companion Surplus in the United States: Relinquishment of Nonadoptables to Animal
Shelters for Euthanasia. J. Appl. Anim.
Kessler, M.R., Turner, D.C., 1997. Stress and Acclimateation of Cats (Felis silvestris
catus) Housed Singly, in Pairs and in Groups in Boarding Catteries. Anim. Welf. 6, 243254.
Kessler, M.R., Turner, D.C., 1999. Effects of Density and Cage Size on Stress in
Domestic Cats (Felis silvestris catus) Housed in Animal Shelters and Boarding Catteries.
Anim. Welf. 8, 259-267.
Kiess, A.S., Hester, P.Y., Mench, J.A., Newberry, R.C., Garner, J.P., 2012. A
standardized cage measurement system: A versatile tool for calculating usable cage
space1. J. Appl. Poult. Res. 21, 657-668.
Knutson, B., Burgdorf, J., Panksepp, J., 2002. Ultrasonic vocalizations as indices of
affective states in rats. Psychol. Bull. 128, 961-977.
Korte, S.M., Koolhaas, J.M., Wingfield, J.C., McEwen, B.S., 2005. The Darwinian concept
of stress: benefits of allostasis and costs of allostatic load and the trade-offs in health and
disease. Neurosci Biobehav Rev 29, 3-38.
Krahn, D.D., Gosnell, B.A., Grace, M., Levine, A.S., 1986. CRF antagonist partially
reverses CRF- and stress-induced effects on feeding. Brain Res. Bull. 17, 285-289.
128
Krohn, C.C., 1994. Behaviour of dairy cows kept in extensive (loose housing/pasture) or
intensive (tie stall) environments. III. Grooming, exploration and abnormal behaviour.
Appl. Anim. Behav. Sci. 42, 73-86.
Kry, K., Casey, R., 2007. The effect of hiding enrichment on stress levels and behaviourof
domestic cats (Felis sylvestris catus) in a shelter setting and the implications for adoption
potential. Anim. Welf. 16, 375-383.
Lansade, L., Bouissou, M.-F., 2008. Reactivity to humans: A temperament trait of horses
which is stable across time and situations. Appl. Anim. Behav. Sci. 114, 492-508.
Leal, E., Fernández-Durán, B., Guillot, R., Ríos, D., Cerdá-Reverter, J., 2011. Stressinduced effects on feeding behavior and growth performance of the sea bass
(Dicentrarchus labrax): a self-feeding approach. J. Comp. Physiol. B, Biochem. Syst.
Environ. Physiol. 181, 1035-1044.
Lepper, M., Kass, P.H., Hart, L.A., 2002. Prediction of Adoption Versus Euthanasia
Among Dogs and Cats in a California Animal Shelter. J. Appl. Anim. Welf. Sci. 5, 29-42.
Leyhausen, P.a.L., Konrad, 1973. Motivation of Human and Animal Behavior. Van
Nostrand Reinhold Company, New York.
Lockwood, R., 2005. Cruelty toward Cats: Changing Perspectives, in: Salam and Rowan
(Eds.), The State of Animals III, Humane Society Press, Washington DC.
Lore, R.K., Eisenberg, F.B., 1986. Avoidance reactions of domestic dogs to unfamiliar
male and female humans in a kennel setting. Appl. Anim. Behav. Sci. 15, 261-266.
Manteuffel, G., Puppe, B., Schön, P.C., 2004. Vocalization of farm animals as a measure
of welfare. Appl. Anim. Behav. Sci. 88, 163-182.
Marks, I.f., Nesse, R.M., 1994. Fear and fitness: An evolutionary analysis of anxiety
disorders. Ethol. Sociobiol. 15, 247-261.
Marques-Deak, A., Cizza, G., Sternberg, E., 2005. Brain-immune interactions and disease
susceptibility. Mol. Psychiatry 10, 239-250.
Mason, G., Mendl, M., 1993. Why is there no simple way of Measuring Animal Welfare?
Anim. Welf. 2, 301-319.
Mattachini, G., Riva, E., Provolo, G., 2011. The lying and standing activity indices of dairy
cows in free-stall housing. Appl. Anim. Behav. Sci. 129, 18-27.
McCobb, E.C., Patronek, G.J., Marder, A., Dinnage, J.D., Stone, M.S., 2005. Assessment
of stress levels among cats in four animal shelters. J. Am. Vet. Med. Assoc. 226, 548-555.
129
McCune, S., 1995. The impact of paternity and early socialisation on the development of
cats' behaviour to people and novel objects. Appl. Anim. Behav. Sci. 45, 109-124.
McEwen, B.S., 2008. Central effects of stress hormones in health and disease:
Understanding the protective and damaging effects of stress and stress mediators. Eur. J.
Pharmacol. 583, 174-185.
McLean, A.N., 2001. Cognitive abilities — the result of selective pressures on food
acquisition? Appl. Anim. Behav. Sci. 71, 241-258.
Mellen, J.D., 1991. Factors influencing reproductive success in small captive exotic felids
(Felis spp.): A multiple regression analysis. Zoo Biol. 10, 95-110.
Mendl, M.a.H., Robert, 2000. Individuality in the domestic cat: origins, development and
stability, in: Turner and Bateson (Eds.), The Domestic Cat: The biology of its behavior,
Cambridge University Press, Cambridge, pp. 47-64.
Morgan, K.N., Tromborg, C.T., 2007. Sources of stress in captivity. Appl. Anim. Behav.
Sci.102, 262-302.
Morgan, M., Houpt, K.A., 1989. Feline Behavior Problems: The Influence of Declawing.
Anthrozoos. 3, 50-53.
Müller, R., Schrader, L., 2005. Individual Consistency of Dairy Cows’ Activity in Their
Home Pen. J. Dairy Sci. 88, 171-175.
Munksgaard, L., DePassillé, A.M., Rushen, J., Herskin, M.S., Kristensen, A.M., 2001.
Dairy cows’ fear of people: social learning, milk yield and behaviour at milking. Appl.
Anim. Behav. Sci. 73, 15-26.
Nakayama, N., Suzuki, H., Li, J.-B., Atsuchi, K., Tsai, M., Amitani, H., Asakawa, A., Inui,
A., 2011. The role of CRF family peptides in the regulation of food intake and anxiety-like
behavior, Biomol. Concepts, p. 275.
nc3rs, National Centre for the Replacement, Refinement and Reduction of Animals in
Research, 2012. http://www.nc3rs.org.uk/
New, J.C., Salmon, M.D., King, M., Scarlett, J.M., Kass, P.H., Hutchison, J.M., 2000.
Characteristics of Shelter-Relinquished Animals and Their Owners Compared With
Animals and Their Owners in U.S. Pet-Owning Houselholds. J. App. Anim. Wel. Sci, 3,
179-201.
NCPPSP, 2012. The National Council on Pet Population Study and Policy,
www.petpopulation.org
130
NHMRC, A., National Health and Medical Research Council, 2012. Australia.
www.nhmrc.gov.au
Nicol, C.J., 1996. Farm animal cognition. Anim. Sci. 62, 375-391.
NRC, National Research Council, 1996. Guide for the Care and Use of Laboratory
Animals. National Academy Press, Washington, D.C.
OIE, 2012. Animal Welfare page, www.oie.int
Ottway, D.S., Hawkins, D.M., Cat housing in rescue shelters: a welfare comparison
between communal and discrete-unit housing. Anim. Welf. 12, 173-189.
Overall, K.L., Dyer, D., 2005. Enrichment strategies for laboratory animals from the
viewpoint of clinical veterinary behavioral medicine: emphasis on cats and dogs. ILAR J.
46, 202-215.
Pacák, K., Palkovits, M., 2001. Stressor Specificity of Central Neuroendocrine
Responses: Implications for Stress-Related Disorders. Endocr. Rev. 22, 502-548.
Patronek GJ, G.L., Beck AM, et al, 1996. Risk factors for the relinquishment of cats to an
animal shelter. J. Am. Vet. Med. Assoc., 582-588.
Patronek, G., 1998. Free-roaming and feral cats-their impact on wildlife and human
beings. J. Am. Vet. Med. Assoc. 212, 218-226.
Pisa, P., Agrillo, C., 2009. Quantity discrimination in felines: a preliminary investigation of
the domestic cat (Felis silvestris catus). J. Ethol. 27, 289-293.
Podberscek, A.L., Blackshaw, J.K., Beattie, A.W., 1991. The behaviour of laboratory
colony cats and their reactions to a familiar and unfamiliar person. Appl. Anim. Behav. Sci.
31, 119-130.
Poole, T., 1997. Happy animals make good science. Lab. Anim. 31, 116-124.
Price, E.O., 1984. Behavioral Aspects of Animal Domestication. Q. Rev. Biol. 59, 1-32.
Raison, C.L., Miller, A.H., 2003. When Not Enough Is Too Much: The Role of Insufficient
Glucocorticoid Signaling in the Pathophysiology of Stress-Related Disorders. Am. J.
Psychiatry 160, 1554-1565.
Rees, T.M., Lubinski, J.L., 2008. Oral supplementation with l-lysine did not prevent upper
respiratory infection in a shelter population of cats. J. Feline Med. Surg. 10, 510-513.
Rochlitz, I., 2000. Recommendations for the housing and care of domestic cats in
laboratories. Lab. Anim. 34, 1-9.
131
Rochlitz, I., Podberscek, A.L., Broom, D.M., 1998. Welfare of cats in a quarantine cattery.
Vet. Rec. 143, 35-39.
Rochlitz, I., 2000. Feline welfare issues, in: Turner and Bateson (Eds.), The Domestic
Cat: The biology of its behavior Cambridge university Press, Cambridge, pp. 207-226.
Rogers, K., 1998. The Cat and the Human Imagination Feline Images from Bastet to
Garfield. The University of Michigan Press, Ann Arbor Michigan.
Ross, S.R., Calcutt, S., Schapiro, S.J., Hau, J., 2011. Space use selectivity by
chimpanzees and gorillas in an indoor–outdoor enclosure. Am. J. Primatol. 73, 197-208.
Ross, S.R., Lukas, K.E., 2006. Use of space in a non-naturalistic environment by
chimpanzees (Pan troglodytes) and lowland gorillas (Gorilla gorilla gorilla). Appl. Anim.
Behav. Sci. 96, 143-152.
Roth, K.A., Katz, R.J., 1979. Stress, behavioral arousal, and open field activity- A reexamination of emotionality in the rat. Neurosci. Biobehav. Rev. 3, 247-263.
Rushen, J., Taylor, A.A., de Passillé, A.M., 1999. Domestic animals' fear of humans and
its effect on their welfare. Appl. Anim. Behav. Sci. 65, 285-303.
Salman MD, H.J., Ruch-Gallie R, et al, 2000. Behavioral reasons for relinquishment of
dogs and cats to 12 shelters. J. Appl. Anim. Welf. Sci. 93-106.
Sapolsky, R., 2004. Why Zebras Don't Get Ulcers. 3rd ed. Henry Holt and Company, LLC,
New York.
Serpell, J.A., 2000. Domestication and history of the cat, in: Turner and Bateson (Eds.),
The Domestic Cat: The biology of its behaviour, Cambridge University Press, Cambridge,
pp. 179-192.
Shettleworth, S.J., 2001. Animal cognition and animal behaviour. Anim. Behav. 61, 277286.
Siegford, J.M., Walshaw, S.O., Brunner, P., Zanella, A.J., 2003. Validation of a
temperament test for domestic cats. Anthrozoos. 16, 332-351.
Slater, M.R., Miller, K.A., Weiss, E., Makolinski, K.V., Weisbrot, L.A.M., 2010. A survey of
the methods used in shelter and rescue programs to identify feral and frightened pet cats.
J. Feline Med. Surg. 12, 592-600.
Søndergaard, E., Halekoh, U., 2003. Young horses’ reactions to humans in relation to
handling and social environment. Appl. Anim. Behav. Sci. 84, 265-280.
Stella, J., Croney, C., Buffington, T., 2013. Effects of stressors on the behavior and
physiology of domestic cats. Appl. Anim. Behav. Sci. 143, 157-163.
132
Stella, J.L., Lord, L.K., Buffington, C.A.T., 2011. Sickness behaviors in response to
unusual external events in healthy cats and cats with feline interstitial cystitis. J. Am. Vet.
Med. Assoc. 238, 67-73.
Stengel, A.a.Y.T., 2009. Neuroendocrine Control of the Gut During Stress: CorticotropinReleasing Factor Signaling Pathways in the Spotlight. Annu. Rev. Physiol. 71, 219-239.
Stolba, A., Wood-Gush, D.G.M., 1989. The behaviour of pigs in a semi-natural
environment. Anim. Sci. 48, 419-425.
Stroup, W., 2011. Living with Generalized Linear Mixed Models, SAS Global Forum 2011,
p. 349.
Tanaka, A., Wagner, D.C., Kass, P.H., Hurley, K.F., 2012. Associations among weight
loss, stress, and upper respiratory tract infection in shelter cats. J. Am. Vet. Med. Assoc.
240, 570-576.
Tarjei, T., 1989. Coping with confinement — Features of the environment that influence
animals' ability to acclimate. Appl. Anim. Behav. Sci. 22, 139-149.
Tufts University, 2012. The Hoarding of Animals Research Consortium,
vet.tufts.edu/hoarding
Turner, D., 2000. The human-cat relationship, in: Turner and Bateson (Eds.), The
Domestic Cat: the biology of its behaviour, Cambridge University Press, Cambridge, pp.
193-206.
Uetake, K., Goto, A., Koyama, R., Kikuchi, R., Tanaka, T., 2012. Effects of single caging
and cage size on behavior and stress level of domestic neutered cats housed in an
animal shelter. Anim. Sci. J., 1-3.
USDA, United States Department of Agriculture, 2012, www.usda.gov
Valentin, V.V., Dickinson, A., O'Doherty, J.P., 2007. Determining the Neural Substrates of
Goal-Directed Learning in the Human Brain. J. Neurosci. 27, 4019-4026.
van der Harst, J and Spruijt, B., 2009. The importance of reward-evaluating mechanisms
for animal welfare. Veterinary Science Tomorrow.
Vandenheede, M., Bouissou, M.F., 1993. Sex differences in fear reactions in sheep. Appl.
Anim. Behav. Sci. 37, 39-55.
Virues-Ortega, J. V. and Buela-Casel, G., 2006. Psychophysiological effects of humananimal interaction: Theoretical issues and long-term interaction effects. J. Nerv. Ment.
Dis., 194, 52-57.
133
Waiblinger, S., Boivin, X., Pedersen, V., Tosi, M.V., Janczak, A.M., Visser, E.K., Jones,
R.B., 2006. Assessing the human-animal relationship in farmed species: A critical review.
Appl. Anim. Behav. Sci. 101, 185-242.
Walsh, F., 2009. Human-animal bonds II: the role of pets in family systems and family
therapy. Fam. Process, 48, 481-99.
Weary, D.M., Fraser, D., 1995. Signalling need: costly signals and animal welfare
assessment. Appl. Anim. Behav. Sci. 44, 159-169.
Wechsler, B., 1995. Coping and coping strategies: a behavioural view. Appl. Anim.
Behav. Sci. 43, 123-134.
Wechsler, B., Lea, S.E.G., 2007. Acclimateation by learning: Its significance for farm
animal husbandry. Appl. Anim. Behav. Sci. 108, 197-214.
Weiss, J.M., 1971. Effects of coping behavior in different warning signal conditions on
stress pathology in rats. J. Comp. Physiol. Psychol. 77, 1-13.
Weiss, J.M., 1972. Psychological factors in stress and disease. Scientific American 226,
104-113.
Wells, D.L., Hepper, P.G., 1999. Male and female dogs respond differently to men and
women. Appl. Anim. Behav. Sci. 61, 341-349.
WHO, World Health Organization, 2012. Rabies Fact Sheet, www.who.int
Wielebnowski, N.C., Fletchall, N., Carlstead, K., Busso, J.M., Brown, J.L., 2002.
Noninvasive assessment of adrenal activity associated with husbandry and behavioral
factors in the North American clouded leopard population. Zoo Biology 21, 77-98.
Wood-Gush, D.G.M., Duncan, I.J.H., 1976. Some behavioural observations on domestic
fowl in the wild. Appl. Anim. Behav. Sci. 2, 255-260.
Yeon, S.C., Kim, Y.K., Park, S.J., Lee, S.S., Lee, S.Y., Suh, E.H., Houpt, K.A., Chang,
H.H., Lee, H.C., Yang, B.G., Lee, H.J., 2011. Differences between vocalization evoked by
social stimuli in feral cats and house cats. Behav. Processes 87, 183-189.
Young, R., 2003. Environmental Enrichment for Captive Animals. Blackwell Science LTD,
Oxford UK.
Zawistowski, S., Morris, J., Salman, M.D., Ruch-Gallie, R., 1998. Population Dynamics,
Overpopulation, and the Welfare of Companion Animals: New Insights on Old and New
Data. J. Appl. Anim. Welf. Sci. 1, 193-206.
134
Footnotes
a
Feliway, Ceva Animal Health Inc., Lenxa, KS
b Sporttime yoga mat, School Specialty, Appleton, WI
c Hide, Perch & Go, OurPet’s Company, Fairport Harbor, OH
d Sani Chips, Harlan Teklad Aspen bedding, Haslett, MI
e Iams maintenance chicken-based dry formula, Dayton, OH
f Purina ProPlan adult chicken and rice entrée in gravy, Nestle, Vevey, Switzerland
g SAS 9.2, SAS Institute Inc., Cary NC
h GraphPad Prism 5, GraphPad Software Inc., La Jolla, CA
i STATA 11, StataCorp LP, College Station, TX
135
Appendix A: Demographics
136
Treatment
Cage
Age (years)
Breed
Sex
Color
M+m+
M+ mM- m+
M- mTop
Bottom
mean=4.1(0.75-13)
DSH
DMH
DLH
Other
Black
White
brown & black tiger
Orange
calico/tortoise shell
black & white
Grey
Other
Total
17
19
21
19
38
37
73
51
6
14
5
76
14
2
14
9
3
10
8
15
Female
9
10
7
8
17
17
33 (mean=4.2)
22
4
6
1
35
5
2
8
2
3
3
4
7
Male
8
9
14
11
21
20
40 (mean=4.0)
29
2
8
4
41
9
0
6
7
0
7
4
8
Table A.1 Cat and Housing Information Study One
Treatment
Cage
Age (years)
Breed
Sex
Color
M+m+
M+mM-m+
M-mtop
bottom
mean=2.96(0.5-11)
DSH
DMH
DLH
other
black
white
brown & black tiger
orange
calico/tortoise shell
black & white
grey
other
Total
14
14
16
15
31
28
58
43
7
5
4
59
10
0
8
17
6
5
4
9
Female
5
6
10
4
12
13
25 (Mean=3.55)
17
4
1
3
25
3
0
5
1
6
3
3
4
Table A.2 Cat and Housing Information Study Two
137
Male
9
8
6
11
19
15
33 (Mean=2.51)
26
3
4
1
34
7
0
3
16
0
2
1
5
Treatment
Cage
Age (years)
Breed
Sex
Color
M+m+
M+mM-m+
M-mtop
bottom
Mean= 5.9 (1.5-13)
DSH
DMH
DLH
other
1=black
2=white
3=br/blk tiger
4=orange
5=calico/tortie
6=black/white
7=grey
8=other
Total
6
6
8
5
11
14
25
19
2
4
0
25
2
0
7
3
3
2
4
4
Female
5
3
4
3
5
10
15 (mean=6.2)
12
2
1
0
15
2
0
5
1
3
0
2
2
Table A.3 Cat and Housing Information Study Three
138
Male
1
3
4
2
6
4
10 (mean=5.5)
7
0
3
0
10
0
0
2
2
0
2
2
2
Appendix B: Macro Environment Schedule
139
Enriched macro environment (M+)
Unenriched macro environment (M-)
Day
Time
0
17:0020:30
Intake; cage doors covered
21:00
Lights off
07:00
Lights on
1
Event
Event
08:00
Uncover cages; scan 1
Uncover cages; scan 1
08:15-08:30 barking dogs (75-85 dB)
09:00
Focal 1; husbandry
Focal 1
10:00
Scan 2
Scan 2
10:15 noise- dropped brooms, banged trash
can, doors
11:00
Focal 2
12:00
Scan 3
13:00
Focal 3; husbandry
13:15-14:00 talk radio (75-80 dB)
Focal 3
14:00
2
Scan 4
15:00
Focal 4
16:00
Scan 5; cover cages
Focal 4
Rock music 15:00-16:00 (70-75 dB)
Scan 5
16:15-16:30 barking dogs (75-85 dB); 16:30
cover cages
21:00
Lights off
07:00
Lights on
08:00
Uncover cages; focal 1
Uncover cages; focal 1
08:15-08:30 barking dogs (75-80 dB); 08:30
rock music (75-80 dB); husbandry
09:00
Scan 1; husbandry
Scan 1
10:00
11:00
Focal 2
Scan 2
11:00- 11:45 talk radio (75-80 dB)
Focal 3
Scan 3
13:00-13:20 barking dogs (75-85 dB)
Scan 2
12:00
13:00
Scan 3
14:00
15:00
15:15
17:0019:00
Focal 4
Scan 4; cover cages
Scan 4; cover cages
noise- dropped brooms, banged trash can,
doors
Stranger approach test
Release to owners
Table B.1 Daily Schedule
140
Appendix C: Cat and Client History Form Study One and Two
141
Owner name______________________ Cat’s name______________ Date____________
Contact Information: Phone #____________________________ E-mail __________________
Breed_________________________ Date of Birth_________________ Weight _________lb/kg
Sex: (circle one) FI FS
Owned How Long?
MI MN Declawed? No___ Yes___ If yes, Front___ All___
__ __years _____ months
Total Cats________ Total Dogs______ Other Pets___________ Other people____________
Housing: Apartment: studio, 1-2 bedrooms, 3 or more bedrooms, attached house/twin
duplex, attached house, 3 or more units, single house, other________________________
The cat’s condition today is
___________
Previous Illnesses or Surgeries
Please check the box that best applies to your cat
Diet – wet food
(name_______________________)
Diet – dry food
(name_______________________)
How many hours each day, on average,
does your cat spend indoors? (check
one)
None
25%
50%
75%
100%
None
25%
50%
75%
100%
0-6
6-12
18-24
12-18
Indoor Only
If you have more than one cat, what is their relationship?
Not Related Littermate
 Sibling
Parent-Offspring
Single Cat Household
Other
Where did you obtain your cat (source)?
Shelter
Offspring from a pet I already own(ed)
Purchased from a friend
Stray/orphan Purchased from a breeder Purchased from a pet shop
____________
142
Other
Gift
Directions: For items below, please use the following choices to describe how many times you
have seen your pet experience the symptom, adding comments/explanation – as appropriate
Score=
0 = I have NEVER seen it
1 = I have seen it at least ONCE
2 = I see it at least ONCE per YEAR
Score
3 = I see it at least ONCE per MONTH
4 = I see it at least ONCE per WEEK
5 = I see it DAILY
How often does your cat:
Comments/explanation
Cough
Sneeze
Have excessive appetite
Have little appetite
Have difficulty breathing
Vomit (food, hair, bile, other)
Have hairballs
Have diarrhea
Have constipation
Defecate outside the litter box
Strain to urinate
Have frequent attempts to urinate
Urinate outside the litter box
Have blood in the urine
Spray urine
Grooms excessively
Have excessive hair loss
Scratch excessively
Have discharge from eyes
Seem nervous (anxious)
Seem fearful
Seem Aggressive
Seem “needy” of contact or attention
143
Environmental History
Just like people, some cats may be more sensitive to changes in their environment than others.
Please review the attached checklist of common "life events" that can happen in the homes of
indoor-housed cats, and place a check mark next to any event your cat has experienced during
the past 12 months. Please also indicate the approximate date of the event. If you noticed that
any of the events affected the cat’s behavior (↓↓= much worse, ↓ = worse, → = no change, ↑ =
better, ↑↑ = much better). Please put a  in the appropriate box in the last columns next to any
events that affected your cats behavior.
Event
During the past 12 months, my cat has experienced
 Death or departure of a pet family member
 Death or departure of a human family member
 Serious hassle in the household (injury, illness, other)
 New human in the household (spouse, baby, friend,
child, other relative)
 New pet(s) in the household
 Change in schedule (work, school, travel, vacation,
retirement)
 Visitors (friends, relatives, etc.)
 Construction around the house (inside or outside)
 Changes of season
 Weather changes/Severe storm/Earthquake
 New house/apartment
 Frequent loud noises (house/car alarms, neighbors,
etc.)
 Boarding
 Remodeling
 Moving/rearranging furniture
 Neighborhood cats outdoors
 Exam time (for students)
 Holidays
 Change in diet
 Change in litter
 Travel (car, train, plane)
 Other (please describe below)
144
Approximate
Date
Change in behavior
↓↓
↓
→
↑
↑↑
Client Resource Checklist
The Ohio State Veterinary Medical Center
The following questions ask about your cat’s resources because we want to learn more
about your cat’s environment. Please  DK if you don’t know, NA if a question does not
apply to your home, or Yes or No after each question.
#
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Space
DK
Does each cat have its own resting area in a
convenient location that provides some
privacy?
Are resting areas are located such that
another animal cannot sneak up on the cat
while it rests?
Are resting areas are located away from
appliances or air ducts (machinery) that could
come on unexpectedly while the cat rests?
Are perches provided so each cat can look
down on their surroundings?
Can each cat move about freely, explore,
climb, stretch, and play if it chooses to?
If a new bed is provided, is it placed next to
the familiar bed so the cat can choose to use
it if it wants to?
Does each cat have the opportunity to move
to a warmer or cooler area if it chooses to?
Is a radio or TV left playing when the cat is
home alone?
Food and Water
Does each cat have its own food bowl?
Does each cat have its own water bowl?
Are the bowls located in a convenient location
that provides some privacy while it eats or
drinks?
Are bowls located such that another animal
cannot sneak upon this cat while it eats or
drinks?
Are bowls washed regularly (at least weekly)
with a mild detergent?
Are bowls located away from machinery that
could come on unexpectedly?
Litter boxes
Does each cat have its own box in a
convenient, well-ventilated location that still
gives the cat some privacy while using it (1
litter box per cat + 1)?
Are boxes located on more than one level in
multi-level houses?
Are boxes located so another animal cannot
sneak up on the cat during use?
145
NA
Yes
No
Other/Comments
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Are boxes located away from machinery that
could come on unexpectedly during use?
Is the litter kept clean and scooped as soon
after use as possible (just like we flush after
each use – or at least daily)?
Are boxes washed regularly (at least monthly)
with a mild detergent (like dishwashing liquid),
rather than strongly scented cleaners?
Is unscented clumping litter used?
Is the brand or type of litter purchased
changed infrequently (less than monthly)?
Is different litter put in a separate box so the
cat can choose to use it if it wants to?
Social Contact
Does each cat have the opportunity to
engage in play with other animals or the
owner if it chooses to on a daily basis?
Does each cat have the option to disengage
from other animals or people in the household
at all times?
Do any cats interact with outdoor cats through
windows?
How many hours a day are you in sight of
your cat?
How many minutes a do you spend petting
your cat?
How many minutes a do you spend playing
with your cat?
Body Care and Activity
Are horizontal scratching posts provided?
Are vertical scratching posts provided?
Are chew items (e.g., cat-safe grasses)
provided?
Does each cat like to play with toys?
Does each cat have toys to chase that mimic
quickly moving prey?
Does each cat have toys that can be picked
up, carried, and tossed in the air?
Are toys rotated on a regular basis (at least
weekly) to provide novelty?
______(h/day)
____(min/day)
____(min/day)
If you have additional comments on any of the questions, please write them
below, including the question #.
In order to care for your cat in the best possible way we would like to know a few of
his/her likes and dislikes. Please complete to the best of your knowledge.
My cat’s favorite toy is:_______________________________________
My cat’s favorite treat is:___________________________________________
My cat dislikes:___________________________________________________
My cat is afraid of:________________________________________________
Other:___________________________________________________________
146
Appendix D: Approach Test*
147
Study ID:______________________
Date:_____________________________
Divide each cage into 2 equal sized zones (from front to back). An unfamiliar person will
then perform the following test.
1. Stand one meter in front of the cage and stand quietly for 30 seconds. Score
the cat’s response.
2. Take one step towards the cage, place hand on the door, and stand quietly
for 30 seconds. Score the cat’s response.
3. If the cat is not showing defensive aggression (lunging, hissing, growling,
attempting to escape), open the door of the cage, stand quietly, and allow the
cat to investigate or interact with you for 30 seconds. Score the cat’s
response.
4. At each step record, latency to interact with tester, time spent interacting with
tester and time in each zone.
5. Rate the sociability (fearfulness) of the cat using the following scale
6.
①
②
③
④
⑤
Actively Avoidant
No Approach but interacting
Actively seeks interaction
1. Actively avoidant, aggressive, or displaying other signs of distress.
2. Avoidant but slightly less distressed.
3. Remains in the same position (not approaching) in the cage but may be
interacting (e.g., purring, rubbing, kneading paws).
4. Responds positively, approaches observer.
5. Actively seeks interaction with observer; rubs cage door, rolls, purrs,
meows, solicits play, etc
latency to
interact
time spent
interacting
time spent in
zone
Step 1
1
2
Step 2
1
2
Step 3
1
2
Sociability
Score
Notes:
*Adapted from Marston, L.C., Bennett, P.C., 2009. Admissions of Cats to Animal Welfare
Shelters in Melbourne, Australia. J. App. Anim. Welf. Sci. 12, 189-213.
148
Appendix E: Cage Condition Study One Results
149
(in all tables Tx 1= M+m+, Tx 2= M+m-, Tx 3= M-m+, Tx 4= M-m-)
Simple
Tx
Tx
Estimate
Effect
Standard
DF
t Value
Pr >
Error
Alpha
Lower
Upper
|t|
Odds
Lower
Upper
Ratio
Odds
Odds
Ratio
Ratio
Level
Day 1
1
2
1.0196
0.8556
72
1.19
0.2373
0.05
-0.6860
2.7252
2.772
0.504
15.260
Day 1
1
3
1.1402
0.8482
72
1.34
0.1831
0.05
-0.5507
2.8311
3.127
0.577
16.964
Day 1
1
4
1.8797
0.9936
72
1.89
0.0625
0.05
-0.1011
3.8604
6.551
0.904
47.483
Day 1
2
3
0.1206
0.8850
72
0.14
0.8920
0.05
-1.6436
1.8848
1.128
0.193
6.585
Day 1
2
4
0.8600
1.0252
72
0.84
0.4043
0.05
-1.1836
2.9037
2.363
0.306
18.242
Day 1
3
4
0.7395
1.0190
72
0.73
0.4704
0.05
-1.2919
2.7709
2.095
0.275
15.972
Day 2
1
2
-0.7114
0.8275
72
-0.86
0.3928
0.05
-2.3611
0.9382
0.491
0.094
2.555
Day 2
1
3
1.6220
0.8189
72
1.98
0.0514
0.05
-0.01041
3.2545
5.063
0.990
25.906
Day 2
1
4
0.9579
0.7993
72
1.20
0.2346
0.05
-0.6354
2.5512
2.606
0.530
12.823
Day 2
2
3
2.3335
0.8289
72
2.82
0.006
0.05
0.6810
3.9859
10.314
1.976
53.834
Day 2
2
4
1.6694
0.8095
72
2.06
0.043
0.05
0.05561
3.2832
5.309
1.057
26.660
Day 2
3
4
-0.6641
0.8007
72
-0.83
0.4096
0.05
-2.2603
0.9321
0.515
0.104
2.540
Table E.1 Food Intake Study One- Simple Effect Comparisons of Tx*Day Least Squares
Means By Day
150
Label
Estimate
Standard Error
DF
t Value
Pr > |t|
Alpha
Lower
Upper
Exponentiated
Exponentiated
Exponentiated
Estimate
Lower
Upper
Tx 2 vs 3
by Day
-2.2129
1.1145
72
-1.99
0.051
0.05
-4.4345
0.008733
0.1094
0.01186
1.0088
Tx 2 vs 4
by Day
-0.8093
1.2122
72
-0.67
0.5065
0.05
-3.2258
1.6071
0.4452
0.03972
4.9883
Table E.2 Estimates, Food Intake Study One
152
Simple
Effect
Level
macro_grp
_macro_grp
Estimate
Standard
Error
DF
t Value
Pr >
|t|
Alpha
Lower
Upper
Odds
Ratio
Lower
Odds
Ratio
Upper
Odds
Ratio
Day 1
Macro (+)
Macro (-)
0.9440
0.6311
74
1.50
0.1390
0.05
-0.3135
2.2015
2.570
0.731
9.039
Day 2
Macro (+)
Macro (-)
1.6276
0.5586
74
2.91
0.005
0.05
0.5146
2.7407
5.092
1.673
15.498
Table E.3 Simple Effect Comparisons of macro_grp*Day Least Squares Means By Day, Food Intake Study One
151
Treatment
M+m+
M+mM-m+
M-m-
No U x 2 days
2
2
3
3
U out day 1
1
0
0
0
U out day 2
1
4
1
1
Table E.4 No Urine or urine out of box- Number of cats that did not urinate during study
period or urinated out of the litter pan
Treatment
M+m+
M+mM-m+
M-m-
No BM x 2 days
6
7
11
12
BM out day 1
2
0
1
0
BM out day 2
0
1
0
0
Table E.5 No BM or BM out od box- Number of cats that did not have a BM during the
study period or had BM out of the litter pan
152
Simple
Effect
Level
Tx
Tx
Estimate
Standard
Error
DF
t Value
Pr >
|t|
Alpha
Lower
Upper
Odds
Ratio
Lower
Odds
Ratio
Upper
Odds
Ratio
Day 1
1
2
0.2513
1.4584
68
0.17
0.8637
0.05
-2.6589
3.1615
1.286
0.070
23.607
Day 1
1
3
-0.9045
1.2098
68
-0.75
0.4573
0.05
-3.3185
1.5096
0.405
0.036
4.525
Day 1
1
4
-0.4990
1.2768
68
-0.39
0.6972
0.05
-3.0468
2.0489
0.607
0.048
7.759
Day 1
2
3
-1.1558
1.2032
68
-0.96
0.3402
0.05
-3.5567
1.2452
0.315
0.029
3.474
Day 1
2
4
-0.7503
1.2706
68
-0.59
0.5568
0.05
-3.2857
1.7851
0.472
0.037
5.960
Day 1
3
4
0.4055
0.9752
68
0.42
0.6789
0.05
-1.5405
2.3514
1.500
0.214
10.500
Day 2
1
2
0.1335
0.6840
68
0.20
0.8458
0.05
-1.2314
1.4984
1.143
0.292
4.475
Day 2
1
3
1.5198
0.7109
68
2.14
0.0361
0.05
0.1013
2.9384
4.571
1.107
18.885
Day 2
1
4
1.1299
0.6975
68
1.62
0.1099
0.05
-0.2619
2.5216
3.095
0.770
12.449
Day 2
2
3
1.3863
0.6982
68
1.99
0.0511
0.05
-0.00697
2.7796
4.000
0.993
16.112
Day 2
2
4
0.9963
0.6845
68
1.46
0.1501
0.05
-0.3696
2.3623
2.708
0.691
10.615
Day 2
3
4
-0.3900
0.7114
68
-0.55
0.5854
0.05
-1.8095
1.0296
0.677
0.164
2.800
Table E.6 Simple Effect Comparisons of Tx*Day Least Squares Means By Day,
Bowel Movement Study One
153
Simple
Effect
Level
macro_grp
_macro_grp
Estimate
Standard
Error
DF
t Value
Pr >
|t|
Alpha
Lower
Upper
Odds
Ratio
Lower
Odds
Ratio
Upper
Odds
Ratio
Day 1
Macro (+)
Macro (-)
-1.5796
1.1224
70
-1.41
0.1637
0.05
-3.8181
0.6589
0.206
0.022
1.933
Day 2
Macro (+)
Macro (-)
1.2571
0.4920
70
2.56
0.0128
0.05
0.2758
2.2383
3.515
1.318
9.378
Table E.7 Simple Effect Comparisons of macro_grp*Day Least Squares Means By Day,
Bowel Movement Study One
155
Tx
M+m+
M+mM-m+
M-m-
Day 1
Day 2
0 (0%)
2 (11%)
0 (0%)
2 (11%)
0 (0%)
1 (5%)
1 (5%)
2 (11%)
Total # in Tx
group
17
19
21
19
Table E.8 Summary of additional sickness behaviors
154
Appendix F: Scan sample data Study One
155
Tx
Position
M+m+
Time
1
4
0
0
2
11
10
7
2
9
0
0
3
9
10
9
0
1=Front half
2=Rear half
3=LB
4=Elevated
5=Hide
1=Front half
2=Rear half
3=LB
1=Front half
2=Rear half
3=LB
4=Elevated
5=Hide
1=Front half
2=Rear half
3=LB
M+m-
M-m+
M-m-
2
6
0
0
1
9
11
6
1
4
0
0
7
10
7
12
0
Day 1
3
2
0
0
6
9
9
10
0
3
1
0
7
10
7
12
0
4
3
0
0
3
11
9
10
0
0
0
0
8
13
9
10
0
5
3
0
0
4
10
11
7
0
0
0
0
6
14
1
17
1
Day 2
7
8
5
4
0
0
2
0
5
8
5
5
13 13
6
6
0
0
1
3
0
0
0
0
13
9
7
8
8
5
10 11
1
1
6
9
0
0
0
6
14
4
0
5
1
0
9
6
9
8
1
9
4
0
0
7
6
11
8
0
2
0
0
10
9
8
10
1
N
17
19
21
19
Table F.1 Number of cats in each treatment group in each cage position by sample
point.
Position in cage M+m+ (n=17)
20
Day 1
Position in cage M-m+ (n=21)
20
Day 2
10
Perch
Hide
Front
LB
Rear
5
0
1
2
3
4
5
6
7
8
9
Perch
Hide
10
5
1
2
3
4
5
6
7
8
9
Time Point
Position in cage M-m- (n=19)
Position in cage M+m- (n=19)
20
Front
LB
Rear
0
Time Point
20
Day 2
Day 1
Day 2
Day 1
15
# of cats
15
# of cats
Day 2
15
# of cats
# of cats
15
Day 1
Front
10
Rear
LB
1
2
3
4
5
6
7
8
Front
5
5
0
Rear
10
LB
0
1
9
2
3
4
5
6
Time Point
Time Point
Figure F.1 Position in cage for each treatment group.
156
7
8
9
Treatment
Time
Vocalization
Day 1
3
1
2
16
(94%)
0 (0%)
0 (0%)
16
(94%)
1 (6%)
0 (0%)
16
(84%)
0 (0%)
2 (11%)
18
(86%)
0 (0%)
3 (14%)
16
(84%)
1 (5%)
2 (11%)
19
(90%)
1 (5%)
1 (5%)
14
(74%)
2 (11%)
3 (16%)
17
(89%)
1 (5%)
1 (5%)
None
Growl, hiss
Meow
15 (88%)
0 (0%)
2 (12%)
M+m-
None
Growl, hiss
Meow
17 (89%)
0 (0%)
2 (11%)
M-m+
None
Growl, hiss
Meow
17 (81%)
1(5%)
3 (14%)
M-m-
None
Growl, hiss
Meow
17 (89%)
2 (11%)
0 (0%)
158
M+m+
Day 2
4
5
6
7
8
9
N
17
(100%)
0 (0%)
0 (0%)
17
(100%)
0 (0%)
0 (0%)
17
14 (82%)
2 (18%)
0 (0%)
16 (94%)
0 (0%)
1 (6%)
13 (76%)
0 (0%)
2 (12%)
17
(100%)
0 (0%)
0 (0%)
15 (79%)
1 (5%)
3 (16%)
16 (84%)
0 (0%)
3 (16%)
13 (68%)
0 (0%)
6 (32%)
16 (84%)
0 (0%)
3 (16%)
17 (89%)
0 (0%)
2 (13%)
18 (95%)
0 (0%)
1 (5%)
19
21
(100%)
0 (0%)
0 (0%)
21
(100%)
0 (0%)
0 (0%)
20 (95%)
0 (0%)
1 (5%)
19 (90%)
0 (0%)
2 (10%)
20 (95%)
0 (0%)
1 (5%)
21
(100%)
0 (0%)
0 (0%)
21
14 (74%)
3 (16%)
2 (11%)
17 (89%)
1 (5%)
1 (5%)
13 (68%)
1 (5%)
5 (26%)
13 (68%)
1 (5%)
5 (26%)
16 (84%)
1 (5%)
2 (11%)
17 (89%)
1 (5%)
1 (5%)
19
Table F.2 Vocalizations- Number and percentage of cats in each treatment vocalizing at each time poin
157
Appendix G: Cage Condition Results Study Two
158
Treatment
No U
M+m+
M+mM-m+
M-m-
1
0
2
1
U out (day
1)
0
0
0
2
U out (day
2)
0
0
0
0
Table G.1 Number of cats that did not urinate during the study period or
urinated out of the litter pan
Treatment
No BM
M+m+
M+mM-m+
M-m-
8
5
9
6
BM out (day
1)
0
0
0
0
BM out (day
2)
0
0
0
1
Table G.2 Number of cats that did not have a BM during the study period
or had a BM out of the litter pan
Tx
Day 1
Day 2
M+m+
1 (7%)
0 (0%)
Total # in
Tx group
14
M+mM-m+
M-m-
1 (7%)
1 (6%)
1 (7%)
0 (0%)
2 (13%)
0 (0%)
14
16
15
Table G.3 Summary of Sickness behaviors
159
Appendix H: Scan Sample Data Study Two
160
Position in cage M+m+ (n=14)
15
# of cats
Hide
10
Hide
Perch
5
Perch
Front
5
0
Rear
0
2
3
4
5
6
7
8
9
Front
Rear
1
2
3
Time Point
15
4
5
6
7
8
9
Time Point
Position in cage M+m- (n=14)
15
Position in cage M-m- (n=15)
Day 2
Day 1
Day 1
Day 2
Rear
10
# of cats
# of cats
Day 2
Day 1
10
1
Position in cage M-m+ (n=16)
Day 2
Day 1
# of cats
15
5
10
Front
Rear
5
Front
0
1
2
3
4
5
6
7
8
0
9
1
Time Point
2
3
4
5
6
7
8
9
Time Point
Figure H.1 Position in cage by treatment. Clockwise from top left treatment one, three,
four, two.
161
Tx
M+m+
M+m-
M-m+
M-m-
Position
Front half
Rear Half
LB
Perch
Hide
Front half
Rear Half
LB
Front half
Rear Half
LB
Perch
Hide
Front half
Rear Half
LB
Time
1
n
4
0
0
1
9
6
8
0
6
0
0
1
9
6
9
0
2
n
3
0
0
2
9
6
8
0
6
0
0
1
9
6
9
0
Day 1
3
n
3
1
0
2
8
7
7
0
5
0
0
3
8
7
8
0
Day 2
4
N
2
0
0
2
10
3
11
0
3
1
0
2
10
3
11
1
5
n
1
0
0
3
10
1
13
0
1
1
0
2
12
0
12
3
6
n
9
3
0
0
2
7
7
0
4
0
0
2
10
6
9
0
7
n
2
0
0
7
5
5
9
0
6
0
0
2
8
8
7
0
Table H.1 Summary statistics for position in cage; scan behavior.
162
8
n
3
0
0
4
7
3
11
0
3
0
0
5
8
8
7
0
9
n
3
0
0
3
8
3
11
0
1
1
0
5
9
8
7
0
Total
14
14
16
15
Time
Treatment
Vocalization
1
None
Growl, hiss
Meow
10 (71%)
3 (21%)
1 (7%)
M+m-
None
Growl, hiss
Meow
10 (71%)
1 (7%)
3 (21%)
M-m+
None
Growl, hiss
Meow
13 (81%)
3(19%)
0 (0%)
M-m-
None
Growl, hiss
Meow
12 (80%)
1 (7%)
2 (13%)
164
M+m+
Day 1
3
2
13
(93%)
1 (7%)
0 (0%)
11
(79%)
0 (0%)
3 (21%)
15
(94%)
1 (6%)
0 (0%)
9 (60%)
0 (0%)
6 (40%)
Day 2
4
5
6
7
8
9
N
12 (86%)
0 (0%)
2 (14%)
12 (86%)
1 (7%)
1 (7%)
14 (100%)
0 (0%)
0 (0%)
9 (64%)
0 (0%)
5 (36%)
10 (71%)
0 (0%)
4 (29%)
13 (93%)
0 (0%)
1 (7%)
14 (100%)
0 (0%)
0 (0%)
14
12 (86%)
0 (0%)
2 (14%)
13 (93%)
0 (0%)
1 (7%)
14 (100%)
0 (0%)
0 (0%)
12 (86%)
0 (0%)
2 (14%)
10 (71%)
0 (0%)
4 (29%)
12 (86%)
0 (0%)
2 (14%)
14 (100%)
0 (0%)
0 (0%)
14
15 (94%)
1 (6%)
0 (0%)
16 (100%)
0 (0%)
0 (0%)
16 (100%)
0 (0%)
0 (0%)
16 (100%)
0 (0%)
0 (0%)
14 (88%)
2 (13%)
0 (0%)
15 (94%)
0 (0%)
1 (6%)
16 (100%)
0 (0%)
0 (0%)
16
11 (73%)
0 (0%)
4 (27%)
1 (87%)
0 (0%)
2 (13%)
14 (93%)
0 (0%)
1 (7%)
10 (67%)
0 (0%)
5 (33%)
9 (60%)
0 (0%)
6 (40%)
12 (80%)
0 (0%)
3 (20%)
13 (87%)
0 (0%)
2 (13%)
15
Table H.2 Number and percentage of cats in each treatment vocalizing at each time point
Appendix I: Cat and Client History Form Study Three
163
Appendix I: Cat and Client History Form Study Three
164
Owner name____________Cat’s name______________ Date_____________
Contact Information: Phone #_______________ E-mail
Breed
______
_Date of Birth __________________Weight
Sex: (circle one) FI FS
Owned How Long?
____
_lb/kg
MI MN Declawed? No_ Yes_ If yes, Front __All___
__ __years _____ months
Total Cats_______ Total Dogs_____ Other Pets ________Other people_______
Housing: Apartment: studio, 1-2 bedrooms, 3 or more bedrooms, attached
house/twin duplex, attached house, 3 or more units, single house, other________
Previous Illnesses or Surgeries
In order to care for your cat in the best possible way we would like to know a few of
his/her likes and dislikes. Please complete to the best of your knowledge.
My cat’s favorite toy is:_____________________________________________
My cat’s favorite treat is:_____________________________________________
My cat dislikes:___________________________________________________
My cat is afraid of:_________________________________________________
Other:___________________________________________________________
Has your cat been hospitalized in the past year?__________________________
If so when and for how long?________________________________________
Has your cat stayed at a boarding facility in the past year?_________________
If so when and for how long?________________________________________
If you have additional comments on any of the questions, please write them
below, including the question #.
165
Directions: For items below, please use the following choices to describe how many
times you have seen your pet experience the symptom, adding comments/explanation
– as appropriate
Score = 0 = I have NEVER seen it
3 = I see it at least ONCE per MONTH
1 = I have seen it at least ONCE
4 = I see it at least ONCE per WEEK
2 = I see it at least ONCE per YEAR
5 = I see it DAILY
Score
How often does your cat:
Comments/explanation
Cough
Sneeze
Have excessive appetite
Have little appetite
Have difficulty breathing
Vomit (food, hair, bile, other)
Have hairballs
Have diarrhea
Have constipation
Defecate outside the litter box
Strain to urinate
Have frequent attempts to urinate
Urinate outside the litter box
Have blood in the urine
Spray urine
Grooms excessively
Have excessive hair loss
Scratch excessively
Have discharge from eyes
Seem nervous (anxious)
Seem fearful
Seem Aggressive
Seem “needy” of contact or attention
166
Please check the box that best applies to your cat
Diet – wet food
(name_______________________)
Diet – dry food
(name_______________________)
None
25%
50%
75%
100%
None
25%
50%
75%
100%
How many hours each day, on average, does
0-6
your cat spend indoors? (check one)
Only
6-12
12-18
18-24
Indoor
If you have more than one cat, what is their relationship?
Not Related
Littermate
 Sibling
Parent-Offspring
Single Cat Household
Other
Where did you obtain your cat (source)?
Shelter
Offspring from a pet I already own(ed)
Purchased from a friend
Purchased from a breeder Purchased from a pet shop
Appendix J: Cage Condition Study Three
167
Stray/orphan
Gift
Other _____________
Appendix J: Cage Condition Study Three
168
Label
Estimate
SE
DF
t Value
Pr > |t|
Alpha
Lower
Upper
Exponentia
Exponentiated
Exponentiated
ted
Lower
Upper
Estimate
Macro+ vs - @
170
1.8922
1.5266
69
1.24
0.2194
0.05
-1.1532
4.9376
6.634
0.316
139.439
3.6773
1.3772
69
2.67
0.0094
0.05
0.9299
6.4247
39.541
2.534
616.916
1.8922
1.5266
69
1.24
0.2194
0.05
-1.1532
4.9376
6.634
0.316
139.439
3.2773
1.3491
69
2.43
0.0177
0.05
0.5860
5.9687
26.505
1.797
390.989
day 1, year 1
Macro+ vs - @
day 2, year 1
Macro+ vs - @
day 1, year 2
Macro+ vs - @
day 2, year 2
Table J.1 Simple effect comparisons, Food Intake Study Three, Comparisons of the treatment groups at each day over year one
and year two
169
Year one
Year two
Tx
M+
MM+
M-
No U x 2 days
1
3
0
3
U out day 1
1
0
1
1
U out day 2
0
1
0
1
Table J.2 Number of cats that did not urinate during study period or urinated out of the
litter pan
Year one
Year two
Tx
M+
MM+
M-
No BM x 2 days
4
8
5
9
BM out day 1
0
0
0
0
BM out day 2
1
0
0
0
Table J.3 Number of cats that did not have a BM during the study period or had BM out
of the litter pan
Year one
Year two
Tx
M+
MM+
M-
Day 1
1 (8%)
1 (7.6%)
0 (0%)
0 (0%)
Day 2
1 (8%)
1 (7.6%)
0 (0%)
2 (15%)
Total # in Tx group
12
13
12
13
Table J.4 Summary of additional sickness behaviors
170
Appendix K: Scan Sample Data Study Three
171
1
Tx
M+m+
173
M+m-
M-m+
M-m-
Position
Front half
Rear half
LB
Elevated
Hide
Front half
Rear half
LB
Elevated
Hide
Front half
Rear half
LB
Elevated
Hide
Front half
Rear half
LB
Elevated
Hide
2
n
%
n
%
3
0
0
0
3
4
1
1
0
0
4
0
0
2
2
3
2
0
0
0
50%
0%
0%
0%
50%
67%
17%
17%
0%
0%
50%
0%
0%
25%
25%
60%
40%
0%
0%
0%
3
0
0
0
2
3
1
1
0
0
0
0
0
4
4
4
1
0
0
0
50%
0%
0%
0%
33%
50%
17%
17%
0%
0%
0%
0%
0%
50%
50%
80%
20%
0%
0%
0%
Table K.1. Position in cage Year 1 Scan data Study Three
Day 1
Day 2
3
4
5
6
7
n
%
n
%
n
%
n
%
n
%
n
1
0
0
3
2
5
1
0
0
0
0
0
0
5
3
3
2
0
0
0
17%
0%
0%
50%
33%
83%
17%
0%
0%
0%
0%
0%
0%
63%
38%
60%
40%
0%
0%
0%
1
0
0
2
3
5
1
0
0
0
0
0
0
4
4
3
2
0
0
0
17%
0%
0%
33%
50%
83%
17%
0%
0%
0%
0%
0%
0%
50%
50%
60%
40%
0%
0%
0%
2
0
0
1
3
5
1
0
0
0
0
0
0
3
5
1
3
1
0
0
33%
0%
0%
17%
50%
83%
17%
0%
0%
0%
0%
0%
0%
38%
63%
20%
60%
20%
0%
0%
4
0
0
0
1
4
1
0
0
0
2
0
0
3
3
1
3
1
0
0
67%
0%
0%
0%
17%
67%
17%
0%
0%
0%
25%
0%
0%
38%
38%
20%
60%
20%
0%
0%
4
0
0
0
2
5
1
0
0
0
0
0
0
4
4
1
3
1
0
0
67%
0%
0%
0%
33%
83%
17%
0%
0%
0%
0%
0%
0%
50%
50%
20%
60%
20%
0%
0%
Table K.1 Position in cage Year 1 Scan data Study Three
172
3
0
0
2
1
5
1
0
0
0
1
0
0
3
4
0
4
1
0
0
8
9
%
n
%
50%
0%
0%
33%
17%
83%
17%
0%
0%
0%
13%
0%
0%
38%
50%
0%
80%
20%
0%
0%
4
0
0
0
2
5
1
0
0
0
1
0
0
2
5
0
4
1
0
0
67%
0%
0%
0%
33%
83%
17%
0%
0%
0%
13%
0%
0%
25%
63%
0%
80%
20%
0%
0%
Total
6
6
8
5
1
Tx
M+m+
M+m-
174
M-m+
M-m-
Position
Front
half
Rear half
Elevated
Hide
Front
half
Rear half
Elevated
Hide
Front
half
Rear half
Elevated
Hide
Front
half
Rear half
Elevated
Hide
2
Day 1
3
n
%
Day 2
n
%
n
%
n
%
n
%
n
%
n
%
4
5
6
7
8
9
n
%
n
%
4
0
0
2
67%
0%
0%
33%
4
0
0
2
67%
0%
0%
33%
4
0
0
2
67%
0%
0%
33%
3
0
1
2
50%
0%
17%
33%
2
0
1
3
33%
0%
17%
50%
5
0
0
1
83%
0%
0%
17%
3
0
2
1
50%
0%
33%
17%
2
0
2
2
33%
0%
33%
33%
3
0
1
2
50%
0%
17%
33%
5
1
0
0
83%
17%
0%
0%
4
2
0
0
67%
33%
0%
0%
4
2
0
0
67%
33%
0%
0%
4
2
0
0
67%
33%
0%
0%
3
3
0
0
50%
50%
0%
0%
5
1
0
0
83%
17%
0%
0%
4
2
0
0
67%
33%
0%
0%
5
1
0
0
83%
17%
0%
0%
1
5
0
0
17%
83%
0%
0%
0
0
2
6
0%
0%
25%
75%
0
0
2
6
0%
0%
25%
75%
0
0
2
6
0%
0%
25%
75%
1
0
1
6
13%
0%
13%
75%
0
0
1
7
0%
0%
13%
88%
2
0
3
3
25%
0%
38%
38%
0
0
1
7
0%
0%
13%
88%
2
0
0
6
25%
0%
0%
75%
0
0
1
7
0%
0%
13%
88%
3
2
0
0
60%
40%
0%
0%
1
4
0
0
20%
80%
0%
0%
1
4
0
0
20%
80%
0%
0%
0
5
0
0
0%
100%
0%
0%
1
4
0
0
20%
80%
0%
0%
3
2
0
0
60%
40%
0%
0%
0
5
0
0
0%
100%
0%
0%
0
5
0
0
0%
100%
0%
0%
1
4
0
0
20%
80%
0%
0%
Table K.2 Position in cage Year 2 Scan data Study Three
173
Total
6
6
8
5
Time
Treatment
Vocalization
1
Day 1
3
10
(83%)
0 (0%)
2 (17%)
13
(100%)
0 (0%)
0 (0%)
4
10
(83%)
0 (0%)
2 (17%)
13
(100%)
0 (0%)
0 (0%)
9 (75%)
0 (0%)
3 (25%)
M- Y 1
None
Growl, hiss
Meow
12
(92%)
0 (0%)
1 (8%)
M+ Y2
None
Growl, hiss
Meow
3 (25%)
0 (0%)
9 (75%)
6 (50%)
0 (0%)
6 (50%)
8 (67%)
0 (0%)
4 (33%)
8 (67%)
0 (0%)
4 (33%)
None
12
(92%)
13
(100%)
13
(100%)
13
(100%)
13
(100%)
10
(78%)
13
(100%)
13
(100%)
13
(100%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
1 (8%)
0 (0%)
0 (0%)
0 (0%)
0 (0%)
3 (22%)
0 (0%)
0 (0%)
0 (0%)
175
None
Growl, hiss
Meow
Growl, hiss
Meow
6
9 (75%)
0 (0%)
3 (25%)
6 (50%)
0 (0%)
6 (50%)
13
(100%)
0 (0%)
0 (0%)
12
(100%)
0 (0%)
0 (0%)
10
(78%)
0 (0%)
3 (22%)
9
11
(92%)
0 (0%)
1 (8%)
13
(100%)
0 (0%)
0 (0%)
12
(100%)
0 (0%)
0 (0%)
M+ Y 1
M- Y 2
5
Day 2
7
8
10
10
(83%)
(83%)
0 (0%)
0 (0%)
2 (17%) 2 (17%)
12
13
(92%)
(100%)
0 (0%)
0 (0%)
1 (8%)
0 (0%)
10
8 (67%)
(83%)
0 (0%)
0 (0%)
4 (33%)
2 (17%)
2
10
(83%)
0 (0%)
2 (17%)
13
(100%)
0 (0%)
0 (0%)
7 (58%)
0 (0%)
5 (42%)
Table K.3 Number and percentage of cats in each treatment vocalizing at each time point.
174
N
12
13
12
13