Integrative and Comparative Biology
Integrative and Comparative Biology, volume 53, number 5, pp. 780–786
doi:10.1093/icb/ict031
Society for Integrative and Comparative Biology
SYMPOSIUM
Foraging Behavior Delays Mechanically-Stimulated Escape
Responses in Fish
Jimena Bohórquez-Herrera,1,* Sandy M. Kawano† and Paolo Domenici‡
*Fundación Colombiana para la Investigación y Conservación de Tiburones y Rayas (SQUALUS), Carrera 79 No 6-37,
Cali, Colombia; †Department of Biological Sciences, Clemson University, Clemson, SC 29634, USA; ‡CNR-IAMC,
Località Sa Mardini, 09072 Torregrande, Oristano, Italy
From the symposium ‘‘When Predators Attack: Sensing and Motion in Predator-Prey Interactions’’ presented at the
annual meeting of the Society for Integrative and Comparative Biology, January 3–7, 2013 at San Francisco, California.
1
E-mail: [email protected]
Synopsis Foraging and the evasion of predators are fundamental for the survival of organisms, but they impose contrasting demands that can influence performance in each behavior. Previous studies suggested that foraging organisms
may experience decreased vigilance to attacks by predators; however, little is known about the effect of foraging on escape
performance with respect to the kinematics and the timing of the response. This study tested the hypothesis that engaging
in foraging activities affected escape performance by comparing fast-start escape responses of silver-spotted sculpins
Blepsias cirrhosus under three conditions: (1) control (no foraging involved), (2) while targeting prey, and (3) immediately after capture of prey. Escape response variables (non-locomotor and locomotor) were analyzed from high-speed
videos. Responsiveness was lower immediately after capturing a prey item compared with the other two treatments, and
latency of performance was higher in the control treatment than in the other two. Locomotor variables such as maximum
speed, maximum acceleration, and turning rates did not show statistical differences among the three groups. Our results
demonstrate that foraging can negatively affect two fundamental components of the escape response: (1) responsiveness
and (2) latency of escape, suggesting that engaging in foraging may decrease an individual’s ability to successfully evade
predators.
Introduction
In fishes, escape responses usually consist of a C-start
maneuver, which involves the unilateral contraction
of the body musculature into a ‘‘C’’ shape (stage 1),
followed by a propulsive stroke of the tail in the
opposite direction of the initial contraction
(stage 2), allowing fish to avoid a sudden danger in
the environment (Domenici and Blake 1997). C-start
performance can be evaluated on the basis of non-locomotor (e.g., responsiveness, directionality, and
latency of escape) and locomotor (e.g., turning
rate, speed, and acceleration) components
(Domenici 2010a, 2010b). C-start escape responses
are not stereotypic and can differ in aspects that
may significantly influence an individual’s success
in evading predators (Eaton et al. 2001; Scharf
et al. 2003; Marras et al. 2011). Typically, C-start
escape responses are triggered by the Mauthner
cells, which ensure short latencies (of the order of
10–20 ms) and fast turning rates (Eaton and Hackett
1984). However, responses not mediated by
Mauthner cells can also occur, although with
longer latencies (Eaton et al. 2001; Kohashi and
Oda 2008).
Although reaction distance is one of the most fundamental variables that affect preys’ vulnerability
(Godin 1997; Scharf et al. 2003; Walker et al. 2005;
Domenici 2010a, 2010b), little is known about the
context-dependency of other components of escape
behavior, such as the swimming kinematics and latency in response to a mechanical stimulation.
Foraging is one of the potential factors affecting
escape behavior, along with the presence of conspecifics or predators, and proximity to a refuge, among
others (Ydenberg and Dill 1986; Godin 1997;
Langerhans et al. 2004; Domenici et al. 2008;
Advanced Access publication April 26, 2013
ß The Author 2013. Published by Oxford University Press on behalf of the Society for Integrative and Comparative Biology. All rights reserved.
For permissions please email: [email protected].
781
Foraging behavior and escape responses
Domenici 2010b). Foraging and the evasion of predators impose different challenges (Ydenberg and Dill
1986), so that individuals that are attacked while
feeding face the ‘‘feed-or-flee’’ dilemma (i.e., feed
and risk being eaten or flee and lose resources).
Previous work found that foraging was associated
with prey allowing predators to approach more closely, and being less responsive to them (Krause and
Godin 1996), resulting in greater vulnerability to
predators. However, C-start escape latencies and kinematics of fish attacked while foraging have not
been investigated thus far.
Understanding how feeding may modulate escape
performance will provide insight into the dynamics
of predator–prey interactions and the ecological significance of locomotor and non-locomotor aspects of
escaping (Scharf et al. 2003; Domenici 2010b). Here,
we test whether engaging in behaviors related to foraging can influence C-start escape performance of
the silver-spotted sculpin, Blepsias cirrhosus (Pallas
1814), by comparing escape responses exhibited in
the absence of foraging behavior with those performed while focusing on a prey item, prior to and
after capturing it.
Materials and methods
Collection and maintenance of specimens
Forty-five adult silver-spotted sculpins, B. cirrhosus
(family Cottidae), were collected using a beach
seine between June and July 2009 at Jackson Beach,
Friday Harbor (WA). Fish were transported to Friday
Harbor Laboratories (FHL), a research facility of the
University of Washington, Seattle, and were acclimated to aquaria with recirculating seawater
(12–168C) for at least 4 days prior to testing.
Blepsias cirrhosus can be found from the intertidal
zone to depths of 37 m (Miller and Lea 1972), and
feeds primarily on nekton or benthic crustaceans
(Kolpakov and Dolganova 2006), so a diet of dock
shrimp Pandalus danae was used. All animal care and
experimental protocols were in accordance with
guidelines approved by the Institutional Animal
Care and Use Committee at the University of
Washington (IACUC protocol # 3018-09) for the
use and care of research animals at FHL.
Description of treatments
Individuals were randomly divided into three groups
(C: control, n ¼ 15; BF: before feeding, n ¼ 12; and
AF: after feeding, n ¼ 15); and tested according to
each group’s protocol. For the two experimental
treatments that involved foraging behaviors; a freshly
euthanized P. danae shrimp was attached to a
transparent wire for 30 s to simulate prey. The control treatment (C) analyzed escape responses triggered in the absence of prey.
The protocol for the experimental treatments was
as follows: a single prey was placed approximately
one fish-body-length from the place at which the
stimulus would be dropped, allowing the fish to
approach. In the BF treatment, the stimulus was
dropped before the fish captured the prey (i.e.,
when the individual’s mouth was less than onebody-length from the prey). In the AF treatment,
fish were stimulated immediately after they had captured prey.
For all treatments, the mechanical stimulus was
released when the fish was approximately 1–2
body-lengths away (range: 0.083–0.248 m), and at
least 0.5 body-length from the sides of the tank, so
as to eliminate wall effects (Eaton and Emberley
1991). Stimulus distance (Dst) was measured as the
distance between the center of the stimulus and the
center of mass (COM) of the fish.
Comparisons between the individuals in C and BF
allowed us to assess the effects of the fish focusing
their attention on the prey on escape performance
while comparisons between BF and AF analyzed the
effects of having successfully captured a prey on
escape performance. Individuals were starved for at
least 2 days prior to testing to ensure that satiation
did not influence the analysis (Sass and Motta 2002).
Recording escape responses
Prior to testing, both sides of the fish were marked
dorsal to the COM between the two dorsal fins with
6 6 mm2 Spectrolite reflective tape to allow
tracking with an automated data analysis program
(DLTdataViewer3) in MATLAB (v2009a). The
COM was determined on a subsample of euthanized
fish (n ¼ 2) using procedures described by Domenici
and Blake (1991). Marked individuals were then
placed in the filming tank and allowed 30 min to
acclimate. Preliminary observations showed that individuals resumed normal behaviors in about 5 min
after being introduced into the tank. Of the 45
B. cirrhosus that were collected, 42 fish were successfully filmed and analyzed. The sculpins that were
analyzed ranged from 0.087 to 0.163 m total length,
with an average of 0.131 0.002 m. Total length was
not significantly different between the C, BF, and AF
groups (ANOVA: F ¼ 0.941, P ¼ 0.399).
The stimulus consisted of a 50-ml plastic centrifuge tube containing a 100-g brass weight, and attached to a thin rope. The stimulus was dropped
down a black tube, which was suspended into the
782
filming tank and 0.5 cm above the water’s surface, in
order to avoid visual stimulation (Dadda et al. 2010).
A small mirror was placed near the water’s surface by
the black tube in order to identify the frame in
which the stimulus hit the water; that frame was
used as a reference for calculating latency of
escape. The filming tank was lined with a grid of
5 5 cm2 to facilitate calibrating and digitizing
the videos. A screened hole between the filming
and outer tanks allowed for fresh seawater to be
circulated throughout the apparatus. All videos
were recorded using a high-speed Fastec Imaging
Sportscam camera running at 250 Hz and subdivided
based on whether or not fish responded to a startling
stimulation.
The COM, tip of the snout, and midpoint of the
stimulus tube were digitized from high-speed videos
using DLTdataViewer3 routines (Hedrick 2008) in
MATLAB (vR2009a).
Description of the variables analyzed
For videos in which an escape response was
observed, escape performance was assessed on the
basis of both non-locomotor and locomotor variables (Domenici 2010b). Non-locomotor variables
for each treatment included: responsiveness (proportion of individuals responding), directionality (proportion of C-bends directed away from the stimulus
[Domenici and Blake 1993]) and latency of escape
(time between the onset of the stimulus touching the
water’s surface and the first visible response by the
fish). Responsiveness was determined for all individuals, but only those that escaped were analyzed for
the other variables. Stimulus distance (Dst) was also
measured in order to control for potential effects of
differences in that variable among the three
treatments.
The variables related to maneuverability included
average turning rate (ATR; average angular velocity
of the anterior part of body [head to COM] while
turning during stage 1) and maximum turning rate
(MTR; maximum angular velocity of the anterior
part of the body while turning during stage 1).
Variables for performance over distance and through
time were based on the displacement of the COM
and included average cumulative distance (CD) traveled following stimulation, maximum speed (MS),
and maximum acceleration (MA) during a fixed
time (0.06 s; determined to be the average duration
of the sum of stage 1 and stage 2 for our data). Data
on speed and acceleration were derived from calculation of CD by using five-point smoothing techniques on the raw data (Lanczos 1956). Speed
J. Bohórquez-Herrera et al.
prior to the escape response (Uprior) was measured
in all trials, using data from the COM that was digitized from the high-speed videos, in order to control
for potential differences in activity before stimulation. The average speed prior to the escape responses
was not significantly different between treatments
(F(2,24) ¼ 0.295, P ¼ 0.747), suggesting that activity
prior to escaping was not affected by the presence
of food. ATR was measured in 15 trials of routine
swimming on fish that were analyzed for escape responses, in order to provide a baseline for comparing
escape responses with spontaneous behavior.
Statistical analysis
Differences between groups for locomotor variables
were analyzed with parametric ANCOVA’s with
Tukey’s post-hoc tests, using the standard length of
fish and the speed prior to escape (Uprior) as covariates. Group differences for latency were also tested
with an ANCOVA using the previous covariates as
well as the stimulus distance (Dst: distance between
the stimulus to the COM) as another covariate.
Categorical data (directionality and responsiveness)
were compared using chi-squared tests. Statistical
analyses were conducted using ‘‘R,’’ ‘‘SigmaStat,’’
and ‘‘Statistica.’’
Results
Fish exhibited significant differences between treatments for non-locomotor components of the escape
response. Although the majority of the fish escaped
in response to mechanical stimulation in C (73%)
and in BF (83%), only 43% of AF individuals
exhibited this behavior (Fig. 1A). Responsiveness
was significantly different among treatments
(2 ¼ 6.28, P50.05). Subsequent chi-squared analysis
among treatments shows that AF differs significantly
from both C (2 ¼ 5.09; P50.05) and BF (2 ¼ 7.49,
P50.05), while C and BF did not differ from each
other (2 ¼ 0.56, P40.05). Directionality did not
differ among treatments (2 ¼ 2.24, P40.05).
Latency showed significant differences among
treatments after controlling for the covariates
(F(2,21) ¼ 4.146, P ¼ 0.030). The post-hoc Tukey test
indicated that C was significantly different from the
feeding treatments (C-AF: P50.001, C-BF:
P50.001), but the feeding treatments were not different from each other (P ¼ 0.845). Thus, BF and AF
had longer latencies than did C (Fig. 1B).
Locomotor aspects of the escape responses did not
differ among the three treatments (Fig. 2): ATR
(F(2,22) ¼ 2.004, P ¼ 0.159), MTR (F(2,22) ¼ 1.091,
P ¼ 0.353), CD (F(2,22) ¼ 3.256, P ¼ 0.058), MS
783
Foraging behavior and escape responses
(F(2,22) ¼ 0.733, P ¼ 0.492), and MA (F(2,22) ¼ 0.151,
P ¼ 0.861) (Table 1).
The ATR for routine turns and escape responses
were significantly different (two-tailed t-test:
t ¼ 16.534, df ¼ 37.998, P50.001), and there was
Fig. 1 The effect of foraging behavior on responsiveness (A) and
latency of escape (B) (means SE).
no overlap between turns during routine swimming
(32–6968s1) and escape responses (1065–27908s1)
(Fig. 3).
Discussion
Our findings suggest that foraging activities affect
two important components of the escape response:
responsiveness and latency (Fig. 1). Individuals in AF
exhibited lower responsiveness (Fig. 1A), which may
result in reduced success in evading predators, since
the lack of a response would most likely allow the
predator to successfully capture the prey. Previous
work shows that responsiveness was the most important factor affecting the success of fish larvae to
escape, accounting for 86% of the variation in potential for escape (Fuiman et al. 2006). Interestingly,
the lowest responsiveness was found in the AF treatment, suggesting that fish that have ingested a prey
may be physically constrained or inhibited from
escaping due to limitations in the rate of neurological processing of multiple sources of information
(Dukas 2004). Occasionally, fish (14%) in the AF
treatment released the shrimp after completing an
escape response, supporting the idea that the ability
of fish to perform multiple tasks (i.e., feeding and
escaping) is limited.
Foraging decreases latency of performance by increasing the delay in escaping following stimulation
Fig. 2 Results of the locomotor variables among treatments (means SE).
784
J. Bohórquez-Herrera et al.
Table 1 Means (SE) of variables analyzed for C (n ¼ 11), BF (n ¼ 10), and AF (n ¼ 6)
Variable
Dst (m)
Latency (ms)
Control
Before feeding
After feeding
0.122 0.009
0.182 0.013
0.157 0.016
24.727 4.849
61.600 6.483
57.33 8.620
ATR (8s1)
1632.100 122.698
2017.435 120.719
2026.272 236.917
MTR (8s1)
2221.706 136.174
2534.568 127.876
2515.692 308.997
MS (ms1)
1.273 0.076
1.184 0.054
1.096 0.145
MA (ms2)
47.027 3.919
42.754 4.483
41.911 2.948
0.050 0.002
0.049 0.002
0.042 0.003
CD (m)
Fig. 3 Frequency distribution of turning rates (ATR) during routine turns (white bars) and during escape responses (black bars).
to about 61.60 6.48 ms (BF) and 57.33 ms 8.62
(AF), compared with 24.73 4.85 ms (C) (Table 1).
This suggests that focusing on, and capturing, prey
may alter the timing of an individual’s response to
an attack. Escape latencies of C-starts can be
extremely fast (10–50 ms) (Domenici 2010a) and
can determine whether or not a fish is captured in
a predator–prey gambit in which high speeds and
fast reactions are at a premium, and life or death
may be a matter of a few milliseconds (Catania
2009). Therefore, an increase of the order of about
30 ms, such as the one we found, may have important consequences for survival. This increase is likely
to be due to the negative effect of multitasking on an
individual’s ability to perform certain behaviors
(Krause and Godin 1996; Kaby and Lind 2003;
Dadda and Bisazza 2006). C-start escape responses
may vary in latency depending on the environmental
context (Domenici 2010b). At the neural level, variability in latency largely depends on whether the
responses are triggered by the Mauthner cells or
not (Eaton et al. 2001; Kohashi and Oda 2008).
When measured at the intercellular level in larvae,
responses by Mauthner cells show latencies of
about 4 ms compared with 10 ms in non-Mauthner
cells (Kohashi and Oda 2008). This difference is
comparable with the one observed between the foraging and control groups, suggesting that performing
another task, such as foraging, may increase the
threshold for Mauthner-mediated responses. In addition, the locomotor responses mediated by nonMauthner cells does not differ from those mediated
by Mauthner-cells in larval zebrafish (Kohashi and
Oda 2008), a finding in line with our results that
show no effect of foraging on turning rate or on
performance over distance or time.
The negative effects of foraging on non-locomotor
aspects of escape responses have been documented in
various systems (e.g., Krause and Godin 1996; Kaby
and Lind 2003), suggesting a general pattern of decreased performance during multitasking such as that
due to simultaneous feeding and fleeing. For instance, Krause and Godin (1996) found that guppies
(Poecilia reticulata) responded with a longer reaction
distance when foraging, possibly as a result of the
fish’s attention being taken up by searching for
food and by feeding (Krause and Godin 1996).
Based on the speed of their model predator, the differences in timing of the responses observed in control versus foraging fish can be estimated to be of the
order of 1–2 s. Such differences in time are likely to
be due to different degrees of attention being paid to
the surroundings and/or limits in the visual field
caused by a head-down foraging posture (Krause
and Godin 1996). Our results are based on mechanical stimulation and show differences that are of the
order of 30 ms, which are more likely to be due to
different neural controls triggering the escape
response, such as Mauthner cells versus nonMauthner cells, although the possibility that other
mechanisms related to delays in the sensory processing cannot be excluded.
Our work shows that the ability to perform an
anti-predator response can be limited when another
task, such as pursuing or capturing a prey, is under
Foraging behavior and escape responses
way. Because multitasking can be more easily accomplished by lateralized fish, e.g., when foraging while
in the presence of a predator (Dadda and Bisazza
2006), it would be interesting to test the possibility
that the escape response of lateralized fish may be
less affected by foraging activities than those of nonlateralized ones. Lateralized fish have indeed been
found to exhibit shorter latencies than non-lateralized ones, possibly because the latter tend to possess
greater complexity in the neural control of their
responses that could result in a higher threshold
for triggering Mauthner cells (Dadda et al. 2010).
Further work on the effect of multitasking on sensory and locomotor performance of fish is necessary
in order to further unravel the complex relationship
between the need to forage and the need to respond
to attacks by predators. Furthermore, it would be
interesting to include the duration of the starvation
period as another variable to be analyzed in future
studies, since it has been reported to be an aspect
that influences the decision of whether to flee or feed
(Godin and Sproul 1988).
Acknowledgments
We thank John Steffensen, Jon Svendsen, the 2009
FHL Fish Swimming Course and FHL staff. We also
thank anonymous reviewers for helpful comments
provided on previous drafts of the article.
Funding
Partial funding and support was provided by the
project SIP-IPN 20120949 (Mexico; to J.B.H.) and
the Stephen and Ruth Wainwright Fellowship
(USA; to J.B.H. and S.M.K.) and scholarships from
‘‘Consejo Nacional de Ciencia y Tecnologı́a’’
(CONACYT, Mexico) and ‘‘Programa Institucional
de Formación de Investigadores’’ from ‘‘Instituto
Politécnico Nacional’’ (PIFI-IPN, Mexico) (to
J.B.H.) and the Clemson University Dean’s
Fellowship (to S.M.K.).
References
Catania KC. 2009. Tentacled snakes turn C-starts to their
advantage and predict future prey behaviour. Proc Natl
Acad Sci USA 106:11183–7.
Dadda M, Bisazza A. 2006. Does brain asymmetry allow efficient performance of simultaneous tasks? Anim Behav
72:523–9.
Dadda M, Koolhaas WH, Domenici P. 2010. Behavioural
asymmetry affects escape performance in a teleost fish.
Biol Lett 6:414–7.
Domenici P. 2010a. Escape responses in fish: kinematics, performance and behaviour. In: Domenici P, Kapoor BG,
785
editors. Fish locomotion: an eco-ethological perspective.
Enfield (CT): Science Publishers. p. 123–70.
Domenici P. 2010b. Context-dependent variability in the
components of fish escape response: integrating locomotor
performance and behaviour. J Exp Zool 313A:59–79.
Domenici P, Blake RW. 1991. The kinematics and performance of the escape response in the angelfish
(Pterophyllum eimekei). J Exp Biol 156:187–205.
Domenici P, Blake RW. 1993. Escape trajectories in angelfish
(Pterophyllum eimekei). J Exp Biol 177:253–72.
Domenici P, Blake RW. 1997. The kinematics and
performance of fish fast-start swimming. J Exp Biol
200:1165–78.
Domenici P, Turesson H, Brodersen J, Brönmark C. 2008.
Predator-induced morphology enhances escape locomotion
in crucian carp. Proc R Soc B 275:195–201.
Dukas R. 2004. Causes and consequences of limited attention.
Brain Behav Evol 63:197–210.
Eaton RC, Hackett JT. 1984. The role of Mauthner cells in
fast-starts involving escape in teleost fish. In: Eaton RC,
editor. Neural mechanisms of startle behaviour. New
York: Plenum Press. p. 213–66.
Eaton R, Emberley DS. 1991. How stimulus direction determines the trajectory of the Mauthner-initiated escape response in a teleost fish. J Exp Biol 161:469–87.
Eaton RC, Lee RKK, Foreman MB. 2001. The Mauthner cell
and other identified neurons of the brainstem escape network of fish. Prog Neurobiol 63:467–85.
Fuiman LA, Rose KA, Cowan JH, Smith EP. 2006. Survival
skills required for predator evasion by fish larvae and their
relation to laboratory measures of performance. Anim
Behav 71:1389–99.
Godin JGJ. 1997. Evading predators. In: Godin JGJ, editor.
Behavioural ecology of teleost fishes. Oxford, UK: Oxford
University Press. p. 191–236.
Godin JGJ, Sproul CD. 1988. Risk taking in parasitized sticklebacks under threat of predation: effects of energetic need
and food availability. Can J Zool 66:2360–7.
Hedrick TL. 2008. Software techniques for two- and threedimensional kinematic measurements of biological and biomimetic systems. Bioinspir Biomim 3:1–6.
Kaby U, Lind J. 2003. What limits predator detection in blue
tits (Parus caeruleus): posture, task or orientation? Behav
Ecol Sociobiol 54:534–8.
Kohashi T, Oda Y. 2008. Initiation of Mauthner- or nonMauthner-mediated fast escape evoked by different modes
of sensory input. J Neurosci 28:10641–53.
Kolpakov NV, Dolganova NT. 2006. On the biology of
Blepsias cirrhosus (Hemitripteridae) from coastal waters of
northern Primorye. J Ichthyol 46:519–24.
Krause J, Godin JGJ. 1996. Influence of prey
foraging posture on flight behaviour and predation risk:
predators take advantage of unwary prey. Behav Ecol
7:264–71.
Lanczos C. 1956. Applied analysis. Englewood Cliffs (NJ):
Prentice Hall.
Langerhans RB, Layman CA, Shokrollahi AM, DeWitt TJ.
2004. Predator-driven phenotypic diversification in
Gambusia affinis. Evolution 58:2305–18.
Marras S, Killen SS, Claireaux G, Domenici P, McKenzie DJ.
2011. Behavioural and kinematic components of the
786
fast-start escape response in fish: individual variation and
temporal repeatability. J Exp Biol 214:3102–10.
Miller DJ, Lea RN. 1972. Guide to the coastal marine fishes of
California, Vol. 157. California Department of Fish and
Game, Fish Bulletin.
Sass GG, Motta PJ. 2002. The effects of satiation on
strike mode and prey capture kinematics in the largemouth bass, Micropterus salmoides. Environ Biol Fishes
65:441–54.
J. Bohórquez-Herrera et al.
Scharf FS, Buckel JA, McGinn PA, Juanes F. 2003.
Vulnerability of marine forage fishes to piscivory: effects
of prey behavior on susceptibility to attack and capture.
J Exp Mar Biol Ecol 294:41–59.
Walker JA, Ghalambor CK, Griset OL, McKenney D,
Reznick DN. 2005. Do faster starts increase the probability
of evading predators? Funct Ecol 19:808–15.
Ydenberg RC, Dill LM. 1986. The economics of fleeing from
predators. Adv Stud Behav 16:229–49.