Behavioral
Ecology
The official journal of the
ISBE
International Society for Behavioral Ecology
Behavioral Ecology (2014), 25(6), 1374–1379. doi:10.1093/beheco/aru131
Original Article
Personality predicts behavioral flexibility in a
fluctuating, natural environment
Katherine A. Herborn,a Britt J. Heidinger,b Lucille Alexander,c and Kathryn E. Arnolda,d
aInstitute of Biodiversity, Animal Health and Comparative Medicine, School of Life Sciences, College of
Medical, Veterinary and Life Sciences, University of Glasgow, University Avenue, Glasgow G12 8QQ,
UK, bDepartment of Biological Sciences, North Dakota State University, 1340 Bolley Drive, Fargo, ND
58102, USA, cWALTHAM® Centre for Pet Nutrition, Freeby Lane, Waltham-on-the-Wolds, LE14 4RT,
UK, and dEnvironment Department, University of York, Heslington, York YO10 5DD, UK
Received 31 October 2013; revised 29 April 2014; accepted 15 June 2014; Advance Access publication 21 August 2014.
In captivity, personality types differ in behavioral flexibility: “fast” (neophilic, exploratory, and aggressive) types quickly form routines,
whereas “slow” types continually adjust their behavior with environmental changes. If these differences extend to pertinent natural
environmental changes, which indicate changing predation or starvation threat, then personality may reflect important variation in how
animals manage risk. We examined whether personality traits classified in captivity would predict the behavioral flexibility of blue tits
Cyanistes caeruleus as they foraged in the wild. Behavioral flexibility was defined as the tendency to adjust feeder use with changing
winter air temperature, an indicator of starvation risk. At the population level, birds reduced their feeder use on warm days. Individuals,
though, varied widely, with some instead using feeders at a relatively fixed, temperature-independent rate. Sex and size, correlates of
dominance in blue tits, did not predict behavioral flexibility. Instead, behavioral flexibility was predicted by age and 2 personality traits:
old, neophobic, and exploratory birds were more behaviorally flexible than young, neophilic, and nonexploratory birds. Our findings
suggest that personality types differ in how they use information about their environments and hence cope with environmental change.
Key words: behavioral plasticity, Cyanistes caeruleus, environmental change, exploration, neophobia, starvation risk.
Introduction
The term “personality” describes behavioral differences among
individuals that are consistent over time or contexts, for example,
in the tendency to approach novel objects or predators, explore
new environments, or interact with conspecifics (Gosling 2001).
Such variation may reflect underlying differences in responsiveness
of hypothalamic–pituitary–adrenal axis that regulates hormonal
response to environmental stimuli (Groothuis and Carere 2005;
Cockrem 2013). Within the same environment, individuals differing in personality often differ also in their fitness and condition
(Dingemanse and Réale 2005; Herborn et al. 2011). In a few studies comparing fitness of personality types over multiple years, there
is evidence that “slow” personality types, that is, less exploratory,
neophobic, less active, or passive animals are less successful than
“fast” (exploratory, neophilic, active, or aggressive) types in relatively stable environments but that this is reversed in changeable
environments (Dingemanse and Réale 2005). Two observations in
captive studies suggest a mechanism for this environment-specific
Address correspondence to K. Herborn. E-mail: katherine.herborn@
glasgow.ac.uk.
© The Author 2014. Published by Oxford University Press on behalf of
the International Society for Behavioral Ecology. All rights reserved. For
permissions, please e-mail: [email protected]
advantage: 1) fast types learn novel associations or habituate to
novel or risky stimuli more quickly than slow types (Marchetti and
Drent 2000; Dugatkin and Alfieri 2003; Rodriguez-Prieto et al.
2010) and 2) once formed, fast types tend to stick to routines, using
old information to predict the environment, whereas slow types
continually gather and use new information (Benus et al. 1987,
1988; Guillette et al. 2010). These attributes may allow fast types to
quickly monopolize resources in stable environments. In contrast,
slow types may be relatively “behaviorally flexible,” that is, able to
adjust their behavior in response to changing environmental stimuli
(Coppens et al. 2010; Sih and Del Giudice 2012). Extra time or
energy spent on information gathering may allow slow, behaviorally flexible types to “keep pace” in changeable environments (Dall
and Johnstone 2002).
However, these predictions on personality and behavioral flexibility come from studies of captive animals. In captivity, fastexploring great tits (Parus major) continue to visit feeders for a period
after they are emptied, whereas slow explorers extend their search
to new sites (Marchetti and Drent 2000). Comparing range location after emptying artificial feeding stations in the wild, however,
van Overveld and Matthysen (2009) found the opposite. Reliance
on old information may have limited fast explorers to the emptied
Herborn et al. • Personality and behavioral flexibility in the wild
feeders in the simple captive environment but, equally, channelled
them to recalled, distant natural sites in the wild. A few other recent
studies have compared the behavior of personality types between
captivity and the wild (Dingemanse and de Goede 2004; Herborn
et al. 2010; Minderman et al. 2010; Schuett et al. 2012) or the
behavior of captive-bred to wild-caught animals (Mason et al.
2013). These studies raise an important point: to understand the
mechanisms underlying fitness variation among personality types,
predictions must be explored in the greater complexity of the natural environment (Archard and Braithwaite 2010). Therefore, we
aimed to test the prediction that personality types differ in behavioral flexibility in free-living birds.
Moreover, we aimed to quantify behavioral flexibility in relation
to a pertinent, natural environmental cue. To this end, behavioral
flexibility was defined as the change in foraging behavior, specifically artificial feeder use, with changing air temperature in winter.
In winter, when food is scarce and thermoregulatory costs high,
birds use feeders more often (Chamberlain et al. 2005) and gain
mass earlier in the day (Macleod et al. 2005) than under more
benign conditions. These activities not only increase body reserves,
providing a buffer against starvation, but also incur costs in terms
of increased exposure to predators while foraging and increased
mass-specific predation risk (Metcalfe and Ure 1995; Macleod et al.
2005). Thus, Parids may use air temperature as a cue to predict
starvation risk and, hence, optimize foraging rate (Fitzpatrick 1997)
or fat storage (Gosler 2002).
Our study species was the blue tit (Cyanistes caeruleus). The personality traits were exploratory tendency and neophobia, which describe
responses to novel environments and to resources near novel objects,
respectively (Verbeek et al. 1994; Greenberg 1995). Although in
many species these traits are positively correlated, hence considered
aspects of a single proactive–reactive trait (Groothuis and Carere
2005; le Vin et al. 2011), they are independent in blue tits (Herborn
et al. 2010). Previously, in a larger sample of individuals of which
this study uses a subset, we demonstrated that a blue tit’s novel environment activity and neophobia in captivity predicted analogous
behaviors in the wild and that these behaviors were repeatable in
both contexts (Herborn et al. 2010). In addition, we compared 2
aspects of exploration behavior: activity specifically in novel parts of
a cage and total activity in a cage. Only novel environment activity
predicted wild behavior (Herborn et al. 2010).
Here, we provide a novel test of whether an individual’s personality predicts their response to a fluctuating environmental cue. At
the population level, we anticipated that individuals would spend
more time foraging at feeders when the air temperature was colder,
to maintain body reserves. However, we predicted that this response
would vary among individuals and that birds that were more behaviorally flexible (i.e., more neophobic and less exploratory) would
adjust their behavior in response to fluctuating air temperature,
whereas less behaviorally flexible birds (i.e., less neophobic and more
exploratory) would not alter their behavior in response to temperature. Additionally, we examined whether sex and size, correlates of
dominance in Parids (Braillet et al. 2002; Krams et al. 2010), and
age influenced behavioral flexibility and/or average feeder use.
Methods
Overview of study system
Over 2 winters, we collected feeder use data at 8 feeding stations in
woodland around the Scottish Centre for Ecology and the Natural
Environment (SCENE; 56°08′N; 4°37′W). For detailed description
1375
of the study system and captive methods, see Herborn et al. (2010).
Briefly, each feeding station had 2 opaque feeders, baited with peanut granules. One bird could access each feeder at a time, from
a small perch fitted with an antenna. Feeders were baited for
1 month prior to the study, so that all birds were familiar with using
the feeders prior to collecting behavioral data. Each winter, we then
mist-netted on 2 or 3 mornings, spaced through the winter, at each
feeding station to capture feeder-using birds. On capture, each bird
was fitted with a leg-ring mounted passively integrated transponder
(“PIT” tag; 11.5 mm × 2.1 mm, <0.1 g, Trovan Unique™). When
on a feeder perch, within the electromagnetic field of the antenna,
this PIT tag allowed identification of the bird by a connected electronic reader (Trovan™ LID665). On capture, we also measured
wing length and determined age (juvenile/adult) from plumage
traits. Birds were then taken to SCENE for personality testing.
They were returned after 1 or 2 nights in captivity. PIT tag records,
from which we derived feeder use, were collected thereafter. We
used these records to determine also which feeding stations each
bird used. All birds were detected at their feeding station of capture
on the first day of PIT tag recording after release; thus, capture did
not cause the birds to change feeding site.
Feeder use data
We were interested in feeder use relative to air temperature. Where
readily available, artificial food makes up on average 15% of the
diet of blue tit nestlings and 21% of the adult winter diet of the
closely related black-capped chickadees (Poecile atricapillus) and scales
with total food intake that varies with energetic demands (Cowie
and Hinsley 1988; Brittingham and Temple 1992). Changes in
feeder use are, therefore, likely to represent changes in total food
intake.
Maximum daily air temperature (henceforth “air temperature”)
was collated from Met Office records for Gartocharn (56°03′N;
4°34′W), which ranged between 1.2 and 12.1 °C. Maximum daily
air temperature in the previous day (“previous day air temperature”) may alter birds’ body reserves on entering trials and changing day length their total foraging opportunity. Both variables may,
therefore, alter perceived starvation risk (Gosler 2002), hence feeder
use, so were collated to control when comparing among individuals sampled on different days. At the end of the study, we used the
average air temperature across study days (7.8 °C) to identify 68
focal individuals that had feeder use data for at least one above and
one below average air temperature.
Feeder use was measured in the 4 h following sunrise in winter
(22 December 2007 and 25 February 2008, and 12 January 2009
and 26 February 2009). To collect the feeder use data, we rotated 4
PIT tag readers around the 8 feeding stations throughout the winter,
obtaining between 8 and 14 mornings of readings per feeding station
per year. PIT tag readers recorded the identity of feeder users at 2-s
intervals. Thus, the number of detections per bird was our score of
their total feeder use on a given morning (mean number of mornings = 6.5 per bird, range = 2–14). We summed feeder use over the
whole morning rather than the most competitive early feeds because
we wanted to compare total food intake among birds rather than
intake under competition. We scored birds a 0 whenever their PIT
tag was not recorded on the focal feeding station during the morning but was recorded later that day; thus, they were in the area. Of
68 birds, 51 used 1 feeding station and 17 used 2 feeding stations.
In those 17 birds, feeder use data were excluded when they were
detected at both within the same week, in case they had divided their
time between the focal feeding station and another not currently
Behavioral Ecology
1376
connected to the PIT tag readers. On the 68 birds, we collected 441
feeder use measurements in total, of which 20 were 0 scores.
Personality trials
Personality tests were conducted more than 2 days in captivity (see
also Herborn et al. 2010). Each bird was tested only once in one of
the study years. On arrival, the focal bird was enclosed within one
half of a 150 cm × 50 cm × 50 cm cage. Food and water were provided ad libitum and the bird undisturbed for at least 2 h. The first
trial was the exploration trial, after which the bird had access to the
entire cage. Neophobia trials were conducted during the afternoon
of day 1 and repeated on the morning of day 2. In 2007–2008,
birds were then blood sampled for genetic sexing and released at
the capture site. In 2008–2009, birds were released after a second
exploration trial on the morning of day 3.
Exploratory tendency
By initially enclosing birds within one cage half, we aimed to create
a “familiar” and, behind the divider, a “novel” environment. We
removed food 1 h and water 30 min prior to the trial to motivate
birds to forage. The trial started when the observer removed the
cage divider. The focal bird was observed for 10 min, during which
we recorded the number of movements (hops or flights between
perches or covering >50 cm), and the endpoint of each movement:
novel or familiar.
To assay exploration independently of neophobia, the arrangement of 3 plastic vine-covered perches was the same in each cage
half, that is, the novel half was novel only in that it was unexplored.
To determine whether activity per se or specifically activity directed
toward the novel environment predicted behavioral flexibility, we
extracted 2 measures from the trial. First, we regressed day 1 activity in the novel environment against day 1 activity in the familiar
environment, after log-transforming both values. The residuals
of this model were our measure of “novel environment activity,”
with positive values indicating highly exploratory birds that were
relatively active in the novel environment, and vice versa. Novel
environment activity was, therefore, independent of the summed
activity in the novel and familiar environments: “total activity.”
The repeatability of novel environment activity was 0.27 and
total activity was 0.42 in a previous study of individuals from this
population, of which these 68 are a subset (Herborn et al. 2010).
In that study and here, novel environment activity was independent
of sex, mass, wing length, and age (analysis of variance [Anova]:
F4,63 = 0.46, P = 0.77).
Neophobia
Neophobia is measured using the extent to which approaching a
resource, here food, is delayed by the presence of a novel object
(Greenberg 1995). Food and water were removed for 30 min prior
to the trial. To start, the observer returned the food bowl with 1
of 2 similarly sized novel objects placed inside: a pink plastic frog
and a half of a purple rubber ball. Latency to approach the bowl
was recorded. After 10 min, the object was removed and water
returned. Birds underwent 2 trials, one per day, with the order of
objects randomized per bird.
Independent of differences in neophobia, individuals may also
differ in their motivation to feed or their response to disturbance
by the observer. To isolate neophobia, we performed the same procedure but without a novel object, returning the food bowl only.
This disturbance control phase was performed 1 h either before
or after each novel object phase, with the order randomized on
each day. After log-transformation, mean novel object latency was
regressed against mean disturbance latency. The residuals of this
model were our measure of neophobia, with positive values indicating birds that were more neophobic (Herborn et al. 2011).
The repeatability of neophobia was 0.28 in a previous study of
individuals from the population, of which these 68 are a subset
(Herborn et al. 2010). In that study and here, novel environment
activity was independent of sex, mass, wing length, age, novel environment activity, and total activity (Anova: F6,61 = 0.78, P = 0.59).
Statistical methods
Analyses were conducted using R 2.15.1 (R Development Core
Team 2012), with glmmADMB_0.7.2.12 (Skuag et al. 2006). We
had 3 objectives:
1) To test whether, at the population level, feeder use declined
with increasing air temperature. We employed a generalized
linear mixed model (GLMM) with feeder use per morning per
individual as the response variable and air temperature as a
fixed variable. A negative binomial error structure accounted
for overdispersion in the response variable and individual
identity was a random effect to account for repeated measures. Previous day air temperature and day length were fixed
effects. To reduce collinearity between the 2 air temperature
measurements, previous day air temperature was expressed as
high or low relative to air temperature, by taking the residual
of a regression with air temperature. As the feeding stations
were also the capture sites, in case birds altered feeder use
immediately postrelease from captivity, a factor “new release”
describing whether a given measurement was the first after
release (yes or no) was also a fixed effect.
2) We aimed to test whether individuals differed in their tendency
to reduce feeder use with increasing air temperature, that is, in
their “behavioral flexibility” with respect to air temperature.
We also tested whether individuals differed in “average feeder
use.” To the GLMM, we added the random slope of interaction: individual identity × air temperature, which then represented individuals’ behavioral flexibility. The random intercept
for individual identity represented individuals’ average feeder
use. We used likelihood ratio tests (LRTs) to compare between
models including and excluding those random effects.
3) We tested whether average feeder use or behavioral flexibility was related to novel environment activity, neophobia, total
activity, age, sex, or wing length. To identify correlates of average feeder use, these traits were added to the original GLMM
as fixed effects. We compared between consecutive nested
models using LRT. To identify correlates of behavioral flexibility, we followed the same procedure but starting with all 2-way
interactions between traits and air temperature. Significant
trait × air temperature interactions would indicate variation in
the relationship between feeder use and air temperature among
individuals differing in that trait, hence within-trait variation in
behavioral flexibility. Equally, we would expect “behaviorally
flexible” individuals to attend to current rather than old information and, hence, to temperature on the day of testing rather
than the temperature of the previous day. To test this possibility also, we include all 2-way interactions of traits × previous
day air temperature, with the expectation of no or opposite
relationships to those found in traits × air temperature.
Herborn et al. • Personality and behavioral flexibility in the wild
1377
Ethical note
Work was licensed by the UK Home Office, Scottish Natural
Heritage, and the British Trust for Ornithology. Studies met the
Association for the Study of Animal Behaviour ethical guidelines
on animal research and were subject to review by Waltham Centre
for Pet Nutrition and the University of Glasgow. No birds died,
were injured, or lost more than 10% body mass while in captivity.
Results
Does feeder use decline with increasing air
temperature at the population level?
As predicted, at the population level, feeder use declined with
increasing air temperature (Table 1i). Also, birds used feeders most
intensively when day length, and hence the opportunity to forage, was shortest. Finally, previous day air temperature correlated
negatively to feeder use, so birds used the feeders more when the
temperature on the previous day was relatively cold and vice versa.
The new release variable was not significant, so recent experience
of captivity did not alter feeder use. Day length, previous day air
temperature, and the new release factor were controlled as fixed
effects in subsequent analyses.
Do individuals differ in average feeder use or
behavioral flexibility?
Removal of the random intercept term significantly reduced
the model fit (LRT: χ2 = 108.92, P < 0.001), so average feeder
use differed between individuals. Removal of the random
slope term significantly reduced the model fit (LRT: χ2 = 7.38,
P = 0.007). Therefore, individuals differed in behavioral flexibility in terms of their change in feeder use with changing air
temperature.
Do personality and other traits predict average
feeder use or behavioral flexibility?
Adult birds (>1 year old) used feeders more than juveniles (<1 year
old) (Table 1ii). Behavioral flexibility was predicted by age, neophobia, and novel environment activity: adults, neophobic birds,
and those that were highly active in novel environments were most
responsive to temperature (Table 1iii). Juveniles, less neophobic
birds, and those that were less active in novel environments fed at a
constant rate irrespective of air temperature. There were no significant interactions between traits and previous air temperature. Sex,
wing length, and total activity did not predict average feeder use or
behavioral flexibility.
Discussion
We provide a rare demonstration that personality types differ in
behavioral flexibility in a seminatural and ecologically relevant context (see also Quinn et al. 2011; Betini and Norris 2012). At the
population level, feeder use declined with increasing air temperature. At the individual level, this response differed between birds:
some showed a steep decline, whereas others used feeders at a relatively fixed, temperature-independent level. Old, neophobic, and,
contrary to expectation from captive studies (Coppens et al. 2010),
exploratory individuals were more behaviorally flexible than young,
neophilic, and nonexploratory individuals. Feeder use also correlated negatively with temperature in the previous day, suggesting
some carryover of expectations and/or an energy deficit. However,
there were no interactions between traits and previous day air temperature, indicating that birds differed in their response to current
but not old temperature information. Behavioral flexibility was not
explained by sex or wing length: 2 correlates of dominance in blue
tits (Braillet et al. 2002; Krams et al. 2010).
To respond flexibly suggests an individual must first incur costs
in terms of time and energy expended on information gathering
Table 1
Predictors of feeder use, from GLMMs with feeder use as the dependent variable, a negative binomial error structure, and individual
identity as a random effect
Variable
(i) Population-level environmental predictors of average feeder use
Intercept
Air temperature (°C)
Previous air temperature (°C)
Day length
New release: yes
(ii) Individual-level predictors of average feeder use
Intercept
Age
(iii) Individual-level predictors of behavioral flexibility
Intercept
Day length
Air temperature (°C)
Previous air temperature (°C)
Age
Neophobia
Novel environment activity
Air temperature (°C) × age
Air temperature (°C) × neophobia
Air temperature (°C) × novel environment activity
Estimate
Standard error
z value
Pr(>|z|)
4.61
−0.09
−0.05
0.00
−0.01
0.36
0.02
0.02
0.00
0.13
12.66
−4.84
−2.12
−2.18
−0.05
<0.001
<0.001
0.034
0.029
0.96
1.30
−0.93
7.09
0.37
0.18
−2.5
0.86
0.012
5.04
0.00
−0.13
−0.05
−1.08
0.20
0.54
0.09
−0.03
−0.08
0.22
0.00
0.02
0.02
0.34
0.12
0.25
0.03
0.01
0.02
22.61
−2.82
−5.62
−2.11
−3.15
1.6
2.15
2.81
−2.05
−3.41
<0.001
0.005
<0.001
0.035
0.002
0.11
0.031
0.005
0.041
0.006
(i) Full model of population-level correlates. (ii) Minimum adequate model of individual-level correlates of average feeder use. (iii) Minimum adequate model of
individual-level correlates of behavioral flexibility, where a significant interaction indicates change in feeder use with air temperature, hence behavioral flexibility.
N feeder use measurements = 441, N birds = 68.
1378
Figure 1
Relationships between feeder use and air temperature in relation to (a–c)
novel environment activity, (d–f) neophobia, and (g and h) age. Novel
environment activity and neophobia were analyzed as continuous variables,
but for illustration are presented in thirds to show the relationship in
individuals with (a) low, (b) medium, and (c) high novel environment activity
and (d) long, (e) medium, and (f) short latency to approach novel objects.
Age is plotted as (g) birds < 1 year and (h) birds > 1 year. Individuals are
shown with gray lines and the average response per third with a black
line. A negative slope indicates a decline in feeder use with increasing
temperature, hence behavioral flexibility.
or learning (Coppens et al. 2010). The benefits of behavioral flexibility will, therefore, be context dependent. For example, personality variation appears to be maintained in one great tit population
by fluctuating selection on personality-specific responses to large,
interannual changes in food availability, with beech mast years
(Dingemanse and Réale 2005). However, our study shows that
personality differences are expressed even when food availability,
in artificial feeders, is constant. And in a great tit population lacking such beech mast years, there is no fluctuating selection and
evidence of directional selection away from “slow” traits in that
more constant foraging environment (Quinn et al. 2009).
Our findings with the neophobia trait fit the expectation from
captive studies that slow types should be behaviorally flexible
(Coppens et al. 2010). Previously, we have shown that this captive
measure of neophobia in blue tits also predicts their neophobia in
the wild (Herborn et al. 2010). A third study on blue tits from our
population showed that nestling diets associated with poor foraging environment were associated with development of neophilic
personality (Arnold et al. 2007). The authors suggested that this
may equip birds to “take risks” on fledging into an impoverished
environment. Taken together, these studies suggest neophobia measured in captivity may reflect a range of strategies, fixed and plastic, for coping with environmental change.
Contrary to several captive studies (Coppens et al. 2010), highly
exploratory blue tits proved more behaviorally flexible than less
exploratory individuals. Comparison of a captive and a wild study
of great tits, which found exploratory birds to be least (Marchetti
and Drent 2000) versus most (van Overveld and Matthysen 2009)
behaviorally flexible, respectively, in terms of their tendency to
move away from unprofitable situations, suggests that a naturalistic context may be important. Differences in how “exploratory
tendency” is defined from otherwise similar behavioral trials, however, as activity level (van Overveld and Matthysen 2009) or time
Behavioral Ecology
taken to explore (Marchetti and Drent 2000), or indeed space use
(Minderman et al. 2010), may also be important (van Dongen et al.
2010). If aspects of exploration behavior vary in state dependency,
the interval between captive and wild testing may also be important, with captive trials typically repeated within days but wild data
collected over months (Bell et al. 2009). Here, total activity did not
predict behavioral flexibility. Activity in a novel environment, however, which combined space use with activity, predicted both the
birds’ propensity to find new, short-lived feeding sites over months
in the wild (Herborn et al. 2010) and also, here, behavioral flexibility in the use of known feeding sites.
Total activity was also independent of average feeder use.
Although this is surprising, given that highly active individuals may
be expected to require more energy, as we did not test whether total
activity in the exploration trial correlated to activity levels in general, we cannot speculate on this possibility. The independence of
average feeder use and all 3 personality traits, however, is encouraging for studies using artificial feeding sites to assay behavior in the
wild (Echeverría and Vassallo 2008; van Overveld and Matthysen
2009; Herborn et al. 2010), where variation in feeder use per se
could bias results.
As expected, feeder use was highest on short and cold days, when
the opportunity to forage was brief and urgent, giving ecological
resonance to our results. However, we also expected, but did not
find, small and female birds to use feeders at a high rate to maintain body reserves as an “insurance policy” against starvation under
risk of competitive exclusion by male and large birds (Dall and
Johnstone 2002). In turn, we expected male and large birds to avoid
costs associated with carrying body reserves (Metcalfe and Ure
1995) by monopolizing feeders on cold days, hence to be behaviorally flexible. However, it is possible that dominance patterns were
lost by summing feeder use across the morning rather than comparing the exact times of feeding: Cole and Quinn (2011) found
that great tits defined as poor competitors employed alternative
strategies at feeders to gain similar total access despite competitive
exclusion. That subdominant groups did not have lower average
feeder use, nor show the opposite patterns in behavioral flexibility
by using feeders most under the low competition of warm days,
supports this possibility.
Interestingly, old birds were more behaviorally flexible than
young birds. As such, the predation-starvation risk trade-off may
differ with age, with young birds at greater predation risk than
older birds when it is warm, but old birds lacking the body reserves
for a sudden cold snap (Dall and Johnstone 2002). Age-specific patterns arise by selective mortality, but also when traits are plastic and
either learnt or have age-specific costs (Komers 1997). This raises
the intriguing possibility that, rather than a fixed characteristic of
particular personality types, personality differences in behavioral
flexibility may be plastic, arising within environments that would
promote behavioral flexibility (Wright et al. 2010). Indeed, there
is some evidence that the expression of personality traits themselves may be plastic, varying with context (Kluen and Brommer
2013) and experience (Bell and Sih 2007). It has been suggested
that personality, in particular behavioral and hormonal responsiveness to novel cues or opportunities, may play an important role
in population-level adaptation to new or shifting environments
(Greenberg 1995; Bell and Sih 2007; Echeverría and Vassallo 2008;
Cockrem 2013; Sih 2013). To understand the mechanisms, it will
be important to determine how personality differences develop and
are expressed as the environment changes (Visser 2008; Kluen and
Brommer 2013).
Herborn et al. • Personality and behavioral flexibility in the wild
Funding
K.H. was funded by a Biotechnology and Biological Sciences
Research Council Industrial Case studentship with WALTHAM®
(BBS/S/M/2006/13139) and K.E.A. by a Royal Society University
Research Fellowship.
We thank R. Macleod, A. Schofield, W. Miles, B. Zonfrillo, E. Leat,
L. Henderson, M. Gastañaga, S. Wilson, D. Fettes, and R. Brennan for field
support and N. Mirzai and T. Wallis for electronic support. Genetic sexing
was conducted by A. Adam and K. Stift.
Handling editor: Alison Bell
References
Archard GA, Braithwaite VA. 2010. The importance of wild populations in
studies of animal temperament. J Zool. 281:149–160.
Arnold KE, Ramsay SL, Donaldson C, Adam A. 2007. Parental prey selection affects risk-taking behaviour and spatial learning in avian offspring.
Proc R Soc. 274:2563–2569.
Bell AM, Hankison SJ, Laskowski KL. 2009. The repeatability of behaviour: a meta-analysis. Anim Behav. 77:771–783.
Bell AM, Sih A. 2007. Exposure to predation generates personality in threespined sticklebacks (Gasterosteus aculeatus). Ecol Lett. 10:828–834.
Benus RF, Koolhaas JM, Van Oortmerssen GA. 1987. Individual differences in behavioural reaction to a changing environment in mice and
rats. Behaviour. 100:105–121.
Benus RF, Koolhaas JM, Van Oortmerssen GA. 1988. Aggression and
adaptation to the light-dark cycle: role of intrinsic and extrinsic control.
Physiol Behav. 43:131–137.
Betini GS, Norris DR. 2012. The relationship between personality and
plasticity in tree swallow aggression and the consequences for reproductive success. Anim Behav. 83:137–143.
Braillet C, Charmantier A, Archaux F, Dos Santos A, Perret P, Lambrechts
MM. 2002. Two blue tit Parus caeruleus populations from Corsica differ in
social dominance. J Avian Biol. 33:446–450.
Brittingham MC, Temple SA. 1992. Use of winter bird feeders by blackcapped chickadees. J Wildl Manage. 56:103–110.
Chamberlain DE, Vickery JA, Glue DE, Robinson RA, Conway GJ,
Woodburn RJW, Cannon AR. 2005. Annual and seasonal trends in the
use of garden feeders by birds in winter. Ibis. 147:563–575.
Cockrem JF. 2013. Corticosterone responses and personality in birds: individual variation and the ability to cope with environmental changes due
to climate change. Gen Comp Endocrinol. 190:156–163.
Cole EF, Quinn JL. 2011. Personality and problem-solving performance
explain competitive ability in the wild. Proc R Soc B. 279:1168–1175.
Coppens CM, de Boer SF, Koolhaas JM. 2010. Coping styles and behavioural flexibility: towards underlying mechanisms. Philos Trans R Soc B.
365:4021–4028.
Cowie RJ, Hinsley SA. 1988. Feeding ecology of great tits (Parus major)
and blue tits (Parus caeruleus) breeding in suburban gardens. J Anim Ecol.
57:611–626.
Dall SRX, Johnstone RA. 2002. Managing uncertainty: information and insurance under the risk of starvation. Philos Trans R Soc B. 357:1519–1526.
Dingemanse NJ, de Goede P. 2004. The relation between dominance and
exploratory behavior is context-dependent in wild great tits. Behav Ecol.
15:1023–1030.
Dingemanse NJ, Réale D. 2005. Natural selection and animal personality.
Behaviour. 142:1159–1184.
van Dongen WFD, Maldonado K, Sabat P, Vasquez RA. 2010. Geographic
variation in the repeatability of a personality trait. Behav Ecol.
21:1243–1250.
Dugatkin LA, Alfieri MS. 2003. Boldness, behavioral inhibition and learning. Ethol Ecol Evol. 15:43–49.
Echeverría AI, Vassallo AI. 2008. Novelty responses in a bird assemblage
inhabiting an urban area. Ethology. 114:616–624.
Fitzpatrick S. 1997. The timing of early morning feeding by tits. Bird Study.
44:88–96.
Gosler AG. 2002. Strategy and constraint in the winter fattening response to
temperature in the great tit Parus major. J Anim Ecol. 71:771–779.
Gosling SD. 2001. From mice to men: what can we learn about personality
from animal research? Psychol Bull. 127:45–86.
1379
Greenberg R. 1995. Novelty responses: the bridge between psychology, behavioral ecology and community ecology? Trends Ecol Evol.
10:165–166.
Groothuis TGG, Carere C. 2005. Avian personalities: characterization and
epigenesis. Neurosci Biobehav Rev. 29:137–150.
Guillette LM, Reddon AR, Hoeschele M, Sturdy CB. 2010. Sometimes
slower is better: slow-exploring birds are more sensitive to changes in a
vocal discrimination task. Proc R Soc B. 278:767–773.
Herborn KA, Coffey J, Larcombe SD, Alexander L, Arnold KE. 2011.
Oxidative stress varies with personality in European greenfinches. J Exp
Biol. 214:1732–1739.
Herborn KA, Macleod R, Miles WTS, Schofield ANB, Alexander L,
Arnold KE. 2010. Personality in captivity reflects personality in the wild.
Anim Behav. 79:835–843.
Kluen E, Brommer JE. 2013. Context-specific repeatability of personality traits in a wild bird: a reaction-norm perspective. Behav Ecol.
24:650–658.
Komers PE. 1997. Behavioural plasticity in variable environments. Can J
Zool. 75:161–169.
Krams I, Cirule D, Suraka V, Krama T, Rantala MJ, Ramey G. 2010.
Fattening strategies of wintering great tits support the optimal body mass
hypothesis under conditions of extremely low ambient temperature.
Funct Ecol. 24:172–177.
Macleod R, Gosler AG, Cresswell W. 2005. Diurnal mass gain strategies
and perceived predation risk in the great tit Parus major. J Anim Ecol.
74:956–964.
Marchetti C, Drent PJ. 2000. Individual differences in the use of social
information in foraging by captive great tits. Anim Behav. 60:131–140.
Mason G, Burns CC, Ahloy Dallaire J, Kroshko J,MacDonald Kinkaid H,
Jeschke JM. 2013. Plastic animals in cages: behavioural flexibility and
responses to captivity. Anim Behav. 85:1113–1126.
Metcalfe NB, Ure SE. 1995. Diurnal variation in flight performance and
hence potential predation risk in small birds. Proc R Soc B. 261:395–400.
Minderman J, Reid JM, Hughes M, Denny MJH, Hogg S, Evans PGH,
Whittingham MJ. 2010. Novel environment exploration and home range
size in starlings Sturnus vulgaris. Behav Ecol. 21:1321–1329.
van Overveld T, Matthysen E. 2009. Personality predicts spatial responses
to food manipulations in free-ranging great tits (Parus major). Biol Lett.
6:187–190.
Quinn JL, Cole EF, Bates J, Payne RW, Cresswell W. 2011. Personality predicts individual responsiveness to the risks of starvation and predation.
Proc R Soc B. 279:1919–1926.
Quinn JL, Patrick SC, Bouwhuis S, Wilkin TA, Sheldon BC. 2009.
Heterogeneous selection on a heritable temperament trait in a variable
environment. J Anim Ecol. 78:1203–1215.
R Development Core Team. 2012. A language and environment for statistical computing. Available from: http://www.R-project.org (Accessed 01
August 2012).
Rodriguez-Prieto I, Martin J, Fernandez-Juricic E. 2010. Individual variation in behavioural plasticity: direct and indirect effects of boldness,
exploration and sociability on habituation to predators in lizards. Proc R
Soc B. 278:266–273.
Schuett W, Laaksonen J, Laaksonen T. 2012. Prospecting at conspecific
nests and exploration in a novel environment are associated with reproductive success in the jackdaw. Behav Ecol Sociobiol. 66:1341–1350.
Sih A. 2013. Understanding variation in behavioural responses to humaninduced rapid environmental change: a conceptual overview. Anim
Behav. 85:1077–1088.
Sih A, Del Giudice M. 2012. Linking behavioural syndromes and cognition: a behavioural ecology perspective. Philos Trans R Soc B.
367:2762–2772.
Skuag H, Fournier D, Nielsen A. 2006. glmmADMB: generalized linear
mixed models using AD Model Builder. R package version 0.3. Available
from: http://www.R-project.org (Accessed 01 May 2013).
Verbeek MEM, Drent PJ, Wiepkema PR. 1994. Consistent individual differences in early exploratory behaviour of male great tits. Anim Behav.
48:1113–1121.
le Vin AL, Mable BK, Taborsky M, Heg D, Arnold KE. 2011. Individual
variation in helping in a cooperative breeder: relatedness versus behavioural type. Anim Behav. 82:467–477.
Visser ME. 2008. Keeping up with a warming world; assessing the rate of
adaptation to climate change. Proc R Soc B. 275:649–659.
Wright TF, Eberhard JR, Hobson EA, Avery ML, Russello MA. 2010.
Behavioral flexibility and species invasions: the adaptive flexibility
hypothesis. Ethol Ecol Evol. 22:393–404.