ARTICLE IN PRESS Animal Behaviour xxx (2010) 1–5

Contents lists available at ScienceDirect

Animal Behaviour journal homepage: www.elsevier.com/locate/anbehav

Turbidity as an ecological constraint on learned predator recognition and generalization in a prey fish Maud C.O. Ferrari*, Kenton R. Lysak, Douglas P. Chivers Department of Biology, University of Saskatchewan

a r t i c l e i n f o Article history: Received 27 August 2009 Initial acceptance 22 October 2009 Final acceptance 16 November 2009 Available online xxx MS. number: A09-00568 Keywords: antipredator behaviour ecological constraint fathead minnow generalization learning Pimephales promelas predator recognition turbidity

The way in which prey animals respond to predators is crucial in shaping direct and indirect interactions in ecosystems. Here, we investigated how a change in turbidity would affect the ability of fathead minnows to recognize potential predators. Minnows were taught to recognize the sight of predatory brown trout and were subsequently tested for their responses to brown trout, rainbow trout or yellow perch in either clear or turbid conditions. In clear water, minnows exhibited an antipredator response towards the brown trout, but they also generalized their recognition to a novel rainbow trout. In turbid water, however, the responses of minnows towards brown trout were lessened and they did not exhibit an antipredator response to the sight of a rainbow trout. None of the minnows displayed a response upon presentation of yellow perch. These results indicate that turbidity alters the quality and quantity of visual information received by the minnows, eliminating their ability to generalize learned recognition of some predators. Ó 2009 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.

The ability of prey to recognize predators is fundamental in allowing prey to maximize energy gain while minimizing costly antipredator responses (Lima & Dill 1990). Nevertheless, behavioural and evolutionary ecologists know surprisingly little about the specific characteristics that prey use to recognize predators from nonpredators (Ferrari et al. 2007, 2008). Recent work with mammals, amphibians and fishes has shown that prey can quickly learn to recognize unknown predators and can use this information to generalize their recognition to other predators that share characteristics with the reference predator. In a pioneering study, Griffin et al. (2001) showed that naı¨ve tammar wallabies, Macropus eugenii, do not show antipredator responses to models of foxes, cats or goats. However, when the wallabies were trained to recognize foxes as predators, they also generalized their recognition to cats, but not to goats. Similarly, Stankowich & Coss (2007) showed that deer could generalize their recognition of a cougar to a novel tiger, but not to a jaguar, a species with similar morphology but which is camouflaged. Ferrari et al. (2008, 2009) similarly showed that fishes and amphibians have the ability to generalize the recognition

* Correspondence and present address: M. C. O. Ferrari, Department of Environmental Science and Policy, University of California, Davis, CA 95616, U.S.A. E-mail address: [email protected] (M.C.O. Ferrari).

of predator odour. Although studies on generalization of predator recognition are at their infancy, they have revolutionized our view of the role of learning in mediating predator recognition. To date, no studies have examined potential ecological constraints that could influence the ability of prey to generalize their visual recognition of predators. In this study, we examined whether suspended solids (turbidity) act as a constraint on the ability of a common freshwater fish (fathead minnows, Pimephales promelas) to acquire recognition of the sight of a predator and subsequently generalize their recognition to other predators. Increasing levels of turbidity caused by eutrophication and other anthropogenic activities are a serious ecological concern (DaviesColley & Smith 2001; Schwartz et al. 2008). Turbidity causes benthic smothering, alters rates of photosynthesis (Davies-Colley & Smith 2001; Bilotta & Brazier 2008) and leads to significant changes in community structure, including a decline in fisheries (LiljendahlNurminen et al. 2008). Much of the observed effects of turbidity have been attributed to differences in cascading effects of fish on prey populations mediated by changes in macrophyte density (Van de Meutter et al. 2005). There is a substantive literature showing that predator–prey interactions are significantly altered by turbidity (Gregory 1993; Bonner & Wilde 2002; Lehtiniemi et al. 2005; Zamor & Grossman 2007). Turbidity degrades transmission of visual information, and consequently interferes with visually

0003-3472/$38.00 Ó 2009 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.anbehav.2009.12.006

Please cite this article in press as: Ferrari, M.C., et al., Turbidity as an ecological constraint on learned predator recognition and generalization in a prey fish, Animal Behaviour (2010), doi:10.1016/j.anbehav.2009.12.006

ARTICLE IN PRESS 2

M.C.O. Ferrari et al. / Animal Behaviour xxx (2010) 1–5

mediated responses to predators. For example, several authors have demonstrated that prey reaction distances to predators are reduced in high turbidity conditions (Vogel & Beauchamp 1999; Quesenberry et al. 2007). To date, no studies have specifically addressed whether recognition of predators is influenced by turbidity. A common method for fish to learn to recognize unknown predators is through conditioning with chemical alarm cues released from injured conspecifics; after being simultaneously exposed to both stimuli, the prey learns to show a fright response to the predator. Prey fish can learn both visual and odour characteristics of predators in this manner (Chivers & Smith 1994a, b). Interestingly, a single pairing of alarm cues with the unknown predator is enough to result in learning, and in the case of fathead minnows, recognition is retained for at least 2 months. In this study, we trained trout-naı¨ve fathead minnows to recognize a predatory brown trout, Salmo trutta, as a threat in clear water conditions. We then tested the antipredator response of the minnows to the sight of a brown trout in both clear water and turbid water conditions. We also tested whether minnows conditioned to recognize brown trout as a threat would generalize their recognition to the sight of a closely related and similar looking rainbow trout, Oncorhynchus mykiss, but not to yellow perch, Perca flavescens, which are morphologically distinct fish predators. The testing was done in both clear water and turbid water conditions to test whether turbidity constrains the recognition and generalization abilities of the minnows. We predicted that turbidity would reduce the quantity and/or quality of visual information and hence decrease the ability of fathead minnows to recognize and generalize their predators. METHODS Test Species and Predators We collected fathead minnows in September 2008 from a pond on the University of Saskatchewan campus using Gee’s Improved minnow traps (Tackle Factory, Fillmore, NY, U.S.A.). Minnows were housed in a 6000-litre flow-through pool and fed ad libitum with commercial fish flakes (Nutrafin basix, Rolf C. Hagen, Inc., Montreal, Quebec, Canada). We obtained brown trout and rainbow trout from the Fort Qu’Appelle fish hatchery, Saskatchewan, in July 2008. The two species were held separately in 6000-litre flow-through pools and fed daily with commercial trout pellets. We collected yellow perch from Blackstrap Lake, Saskatchewan, in July 2005 using seine nets. They were housed in a 6000-litre flow-through pool and fed live minnows. All fishes were maintained in dechlorinated tap water at 12  C under a 14:10 h light:dark cycle. Stimulus Preparation We used five donor minnows (fork length, FL: mean  SD ¼ 4.66  0.46 cm) to prepare the alarm cue solution. Minnows were killed with a blow to the head, in accordance with the Canadian Council on Animal Care. Fillets were removed from both sides of the fish, placed in chilled distilled water and homogenized using a Polytron homogenizer. We then filtered the solution through glass wool. A total of 23.2 cm2 of skin was collected and diluted with distilled water to make a final solution of 1 cm2 of skin per 30 litres. Such a concentration was shown previously to elicit overt antipredator behaviour in minnows (Ferrari et al. 2005). Skin extracts were then frozen at 20  C in 30 ml aliquots until used.

Experimental Set-up The experimental set-ups used for the conditioning and the testing phases of the experiment were similar. A 37-litre tank filled with dechlorinated tap water housed a single minnow. The tank contained a gravel substrate, an air stone, a 2 m long injection tube attached to the air stone, and a shelter that consisted of a 10  20 cm ceramic tile mounted on three 3.5 cm long cylindrical glass legs. We placed another 37-litre tank, housing a predator, directly beside the prey tank. The predator tank was divided into thirds by Plexiglas dividers placed along the long axis of the tank. We placed the predators in the section of the tank closest to the prey tank to prevent it from moving towards or away from the prey. This ensured that all minnows were presented with comparable visual information across trials (i.e. had a lateral view of the predator). The tanks were covered with black plastic on the outer and back sides to provide visual isolation from other tanks. We placed a black removable divider between the two adjacent tanks to ensure visual isolation between prey and predators prior to the start of the trial. The experiment consisted of a conditioning phase, during which minnows were taught to recognize a brown trout as a predator in clear water, followed by a testing phase, where minnows were exposed to brown trout, rainbow trout or perch in clear or turbid water. Conditioning phase Twenty-four hours prior to the start of the trial, we placed a single minnow in the prey tank for acclimation and fed it. One hour prior to the start of the conditioning phase, we fed the minnows again and placed a brown trout (mean standard length  SD ¼ 20.2  2.6 cm) in the predator tank filled with clear water. The conditioning phase started when we removed the visual divider separating the minnow and the brown trout tanks, and we immediately injected either 5 ml of conspecific alarm cues or 5 ml of dechlorinated water (control) in the prey tank. The divider was placed back between the tanks after the minnow was given 15 s to see the predator. No behavioural observations were performed during the conditioning phase. Testing phase One hour following the end of the conditioning phase, we transferred the minnow into a similar 37-litre tank filled with clean tap water and fed it. The only difference was that the prey testing tanks were not equipped with an injection hose. The testing phase occurred 24 h after the conditioning phase. One hour prior to the testing, we fed the minnow and placed a brown trout, a rainbow trout (mean SL: 23.6  2.1 cm) or a perch (mean SL: 21.4  2.4 cm) in the predator tank, and we either left the water clear or made it turbid by adding 4.5 g of bentonite in the predator tank (0.12 g/litre of water, w27 NTU, equivalent to w20 cm secchi depth, Shoup & Wahl 2009). This turbidity level was lower than that often experienced by minnows under natural conditions (Hartman & Abrahams 2000). We observed the minnow for 8 min prior to, and 8 min after the removal of the divider. Two well-documented antipredator behaviours of single fathead minnows are increased shelter use and decreased activity level (Chivers & Smith 1994a). Hence, we measured the number of seconds spent under shelter and the number of seconds spent swimming during the prestimulus and poststimulus presentation periods. We conducted 15 replicates for each of the six control treatments (minnows conditioned with water and exposed to 3 predators in 2 turbidity conditions) and 20 replicates for each of the six experimental groups (minnows conditioned with alarm cues and exposed to 3 predators in 2 turbidity conditions), for a total of 210

Please cite this article in press as: Ferrari, M.C., et al., Turbidity as an ecological constraint on learned predator recognition and generalization in a prey fish, Animal Behaviour (2010), doi:10.1016/j.anbehav.2009.12.006

ARTICLE IN PRESS M.C.O. Ferrari et al. / Animal Behaviour xxx (2010) 1–5

trials. Each minnow was used only once. The order of the trials was randomized across treatments and the observer was blind to the conditioning treatments.

3

Turbidity did not affect the responses of minnows to perch (F2,37 ¼ 0.3, P > 0.8). DISCUSSION

Statistical Analysis We calculated the change in shelter use and time moving from the prestimulus baseline. Because the two variables were not independent from each other, we analysed them simultaneously using a MANOVA procedure. Because of the obvious interaction between conditioning cues (water versus alarm cues), responses to predators (brown trout versus rainbow trout versus perch) and turbidity (clear versus turbid environment), we split the analysis to test both the effect of predator and the effect of turbidity on minnows conditioned with water and with conspecific alarm cues separately. Behavioural data from the control minnows followed parametric assumptions and were analysed with a two-way MANOVA. Behavioural data from the alarm cue minnows did not meet homoscedasticity assumptions (nonhomogeneity of variances). Hence, the data were rank-transformed prior to performing a nonparametric MANOVA using the Sheirer–Ray–Hare extension of the Kruskal–Wallis test (Sokal & Rohlf 2003). Subsequent analyses were performed to further investigate the nature of interactions. RESULTS Minnows Conditioned with Water The two-way MANOVA revealed no effect of predator (Pillai’s trace: F4,168 ¼ 1.0, P > 0.3), no effect of turbidity (Pillai’s trace: F2,83 ¼ 0.5, P > 0.5) and no interaction between the two factors (Pillai’s trace: F4,168 ¼ 1.0, P > 0.9) on the behaviour of minnows (Fig. 1). Minnows Conditioned with Conspecific Alarm Cues Overall, minnows in a turbid environment showed a lower intensity of antipredator response than minnows in a clear environment, and minnows responded the strongest to the brown trout and did not respond to the perch. The two-way nonparametric MANOVA revealed a significant interaction between predator and turbidity (Pillai’s trace: F4,228 ¼ 3.2, P ¼ 0.014; Fig. 1). Effect of Predators In both clear and turbid water, predators had a significant effect on the responses of minnows (Pillai’s trace: clear: F4,114 ¼ 15.1, P < 0.001; turbid: F4,114 ¼ 9.9, P < 0.001). When in clear water, minnows showed the strongest antipredator response when exposed to brown trout, showed an intermediate intensity when exposed to rainbow trout, and showed no response to perch (Tukey pairwise comparisons: shelter use: all P < 0.001, time moving: all P < 0.05). When in turbid water, minnows showed the strongest antipredator response to brown trout (Tukey pairwise comparisons: shelter use: P < 0.001; time moving: P < 0.001). However, they did not distinguish between rainbow trout and perch (shelter use: P > 0.8, time moving: P > 0.9). Effect of Turbidity When exposed to brown trout or rainbow trout, minnows in clear water responded more strongly to the predators than did minnows in turbid environment (Pillai’s trace: brown trout: F2,37 ¼ 15.5, P < 0.01; rainbow trout: F2,37 ¼ 10.7, P < 0.001).

The results of our study demonstrate that fathead minnows are able to learn to respond to the sight of a predatory brown trout based on conditioning with alarm cues and use this information to generalize their recognition to rainbow trout. The minnows did not generalize their recognition to the sight of the perch, which is morphologically distinct from the trout. Besides the two studies on mammals (Griffin et al. 2001; Stankowich & Coss 2007), ours is the only other study demonstrating visual generalization of predator recognition by prey animals. We found that minnows significantly reduced the intensity of their antipredator response to the rainbow trout compared with their response to the brown trout. This diminished response illustrates that predator generalization does not result simply from a lack of differentiation between the brown trout and rainbow trout. Rather, these results indicate that the trout species share enough similarity in morphology to lead to the generalization, but that they are distinct enough that the minnows reduce their response to the rainbow trout. The more characteristics that the prey share in common, the greater the degree of generalization there should be. In a study on chemosensory generalization of predator recognition, Ferrari et al. (2007) showed that minnows conditioned to the odour of lake trout, Salvelinus namaycush, showed antipredator responses to the odour of lake trout, brook trout and rainbow trout, but the intensity of antipredator response reflected the phylogenetic relatedness of the three trout species. Minnows generalized more to salmonid predators that were in the same genus as the reference predator than they did to those that were in the same family as the reference predator. Our results provide evidence that predator recognition is influenced by turbidity. Minnows showed a weaker intensity of antipredator response when exposed to the sight of brown trout in turbid water than in clear water. The weaker response in the turbid conditions could result because the appearance of the predator in the turbid conditions was not a complete match to the predator template that the prey had learned in the clear conditions. Alternatively, the prey may perceive that they are less vulnerable to predation in the turbid water, and hence reduce their antipredator response accordingly (Hartman & Abrahams 2000). Some studies have demonstrated that prey are less vulnerable to capture in turbid water conditions (Gregory 1993). Our results provide evidence that generalization of predator recognition was impaired under turbid conditions, as minnows did not respond to the sight of rainbow trout in turbid conditions, while minnows in clear conditions did respond. Turbidity acts to decreases the amount and/or quality of information that prey perceive when looking at the predator. In the psychology literature, a number of conditioning experiments with arbitrary stimuli have investigated the effect of stimulus alteration on learned responses, referred to as ‘stimulus generalization’ (Ghirlanda & Enquist 2003). Stimulus generalization has been reported using different sensory modalities (sounds and light), in a variety of taxa (from insects to mammals), in several biological contexts (foraging and courtship) and seems independent of experience with the stimulus (reviewed by Ghirlanda & Enquist 2003). A response can be evoked upon exposure to a particular stimulus (e.g. a sound or a light of a given frequency) but also to stimuli that are ‘similar’ to that stimulus (sound or light of a slightly different frequency). The intensity of the response to the test stimulus decreases as the frequency deviates from the learned frequency. Put back in the context of predator recognition, the ability of prey to respond to a variety of stimuli that

Please cite this article in press as: Ferrari, M.C., et al., Turbidity as an ecological constraint on learned predator recognition and generalization in a prey fish, Animal Behaviour (2010), doi:10.1016/j.anbehav.2009.12.006

ARTICLE IN PRESS 4

M.C.O. Ferrari et al. / Animal Behaviour xxx (2010) 1–5

Cue Water + brown trout

Alarm cues + brown trout

Mean change in shelter use (s)

200

150

100

50

0

−50

Mean change in time spent moving (s)

50

0

−50

−100

−150 Brown trout

Rainbow trout

Perch

Brown trout

Rainbow trout

Perch

Figure 1. Mean  SE change from the prestimulus baseline in shelter use (top panel) and time spent moving (bottom panel) for minnows exposed to one of three fish species (brown trout, the known predator, rainbow trout or yellow perch) maintained in either clear (,) or turbid ( ) conditions. Minnows were previously exposed to a brown trout (in clear water) paired with either water (control, left panel) or conspecific alarm cues (right panel).

are ‘similar’ to the ones the prey already recognize as dangerous can have important survival consequences. Turbidity influences the appearance of the rainbow trout enough to eliminate the recognition by the minnows. Several studies have examined how anthropogenic activities that influence water chemistry could alter recognition of predators. Acidic environments (pH < 6.4) occurring from acid rains or industrial effluents deactivate the ‘alarm’ function of alarm cues in salmonid fishes, resulting in the absence of learned predator recognition under these conditions (Leduc et al. 2004, 2006, 2007). In addition, the pH of predator odours also affects learning. Smith et al. (2008) showed that juvenile rainbow trout conditioned to recognize the odour of a predator at pH 6 or 7 subsequently responded to the odour at the same pH as the one used in

the conditioning phase. This indicates that pH affects the ability of prey to recognize the chemistry of the predator odours. Pollutants, such as heavy metals (Scott et al. 2003) or herbicides (Mandrillon & Saglio 2007), also impair learned predator recognition. While our study is the first to test directly how environmental constraints influence the ability of prey to generalize their predators, other factors, such as water chemistry, will probably have similar effects. Moreover, changes in the visual environment caused by alterations in the density of aquatic vegetation could likewise restrict visual information and influence predator generalization. Examining the effects of anthropogenic activities from a behavioural ecology perspective has great potential to elucidate mechanisms by which negative effects manifest themselves.

Please cite this article in press as: Ferrari, M.C., et al., Turbidity as an ecological constraint on learned predator recognition and generalization in a prey fish, Animal Behaviour (2010), doi:10.1016/j.anbehav.2009.12.006

ARTICLE IN PRESS M.C.O. Ferrari et al. / Animal Behaviour xxx (2010) 1–5

Acknowledgments The Natural Sciences and Engineering Research Council of Canada and the University of Saskatchewan provided financial support to D.P. Chivers. All work reported herein was in accordance with the Guidelines to the Care and Use of Experimental Animals published by the Canadian Council on Animal Care and was conducted under the University of Saskatchewan Committee of Animal Care and Supply protocol no. 20070083. References Bilotta, G. S. & Brazier, R. E. 2008. Understanding the influence of suspended solids on water quality and aquatic biota. Water Research, 42, 2849–2861. Bonner, T. H. & Wilde, G. R. 2002. Effects of turbidity on prey consumption by prairie stream fishes. Transactions of the American Fisheries Society, 131, 1203–1208. Chivers, D. P. & Smith, R. J. F. 1994a. Fathead minnows, Pimephales promelas, acquire predator recognition when alarm substance is associated with the sight of unfamiliar fish. Animal Behaviour, 48, 597–605. Chivers, D. P. & Smith, R. J. F. 1994b. The role of experience and chemical alarm signalling in predator recognition by fathead minnows, Pimephales promelas. Journal of Fish Biology, 44, 273–285. Davies-Colley, R. J. & Smith, D. G. 2001. Turbidity, suspended sediment, and water clarity: a review. Journal of the American Water Resources Association, 37, 1085–1101. Ferrari, M. C. O., Trowell, J. J., Brown, G. E. & Chivers, D. P. 2005. The role of learning in the development of threat-sensitive predator avoidance by fathead minnows. Animal Behaviour, 70, 777–784. Ferrari, M. C. O., Gonzalo, A., Messier, F. & Chivers, D. P. 2007. Generalization of learned predator recognition: an experimental test and framework for future studies. Proceedings of the Royal Society B, 274, 1853–1859. Ferrari, M. C. O., Messier, F. & Chivers, D. P. 2008. Can prey exhibit threat-sensitive generalization of predator recognition? Extending the Predator Recognition Continuum Hypothesis. Proceedings of the Royal Society B, 275, 1811–1816. Ferrari, M. C. O., Brown, G. E., Messier, F. & Chivers, D. P. 2009. Threat-sensitive generalization of predator recognition by amphibians. Behavioral Ecology and Sociobiology, 63, 1369–1375. Ghirlanda, S. & Enquist, M. 2003. A century of generalization. Animal Behaviour, 66, 15–36. Gregory, R. S. 1993. Effect of turbidity on the predator avoidance behaviour of juvenile chinook salmon (Oncorhynchus tshawytscha). Canadian Journal of Fisheries and Aquatic Sciences, 50, 241–246. Griffin, A. S., Evans, C. S. & Blumstein, D. T. 2001. Learning specificity in acquired predator recognition. Animal Behaviour, 62, 577–589. Hartman, E. J. & Abrahams, M. V. 2000. Sensory compensation and the detection of predators: the interaction between chemical and visual information. Proceedings of the Royal Society B, 267, 571–575.

5

Leduc, A. O. H. C., Ferrari, M. C. O., Kelly, J. M. & Brown, G. E. 2004. Learning to recognize novel predators under weakly acidic conditions: the effects of reduced pH on acquired predator recognition by juvenile rainbow trout. Chemoecology, 14, 107–112. Leduc, A. O. H. C., Roh, E., Harvey, M. C. & Brown, G. E. 2006. Impaired detection of chemical alarm cues by juvenile wild Atlantic salmon (Salmo salar) in a weakly acidic environment. Canadian Journal of Fisheries and Aquatic Sciences, 63, 2356–2363. Leduc, A. O. H. C., Roh, E., Breau, C. & Brown, G. E. 2007. Effects of ambient acidity on chemosensory learning: an example of an environmental constraint on acquired predator recognition in wild juvenile Atlantic salmon (Salmo salar). Ecology of Freshwater Fish, 16, 385–394. ¨ m-Ost, J. & Viitasalo, M. 2005. Turbidity decreases antiLehtiniemi, M., Engstro predator behaviour in pike larvae, Esox lucius. Environmental Biology of Fishes, 73, 1–8. ¨ , J. 2008. Spatial Liljendahl-Nurminen, A., Horppila, J., Uusitalo, L. & Niemisto variability in the abundance of pelagic invertebrate predators in relation to depth and turbidity. Aquatic Ecology, 42, 25–33. Lima, S. L. & Dill, L. M. 1990. Behavioral decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology, 68, 619–640. Mandrillon, A. L. & Saglio, P. 2007. Herbicide exposure affects the chemical recognition of a non native predator in common toad tadpoles (Bufo bufo). Chemoecology, 17, 31–36. Quesenberry, N. J., Allen, P. J. & Cech, J. J., Jr. 2007. The influence of turbidity on three-spined stickleback foraging. Journal of Fish Biology, 70, 965–972. Schwartz, J. S., Dahle, M. & Robinson, B. R. 2008. Concentration-durationfrequency curves for stream turbidity: possibilities for assessing biological impairment. Journal of the American Water Resources Association, 44, 879–886. Scott, G. R., Sloman, K. A., Rouleau, C. & Wood, C. M. 2003. Cadmium disrupts behavioural and physiological responses to alarm substance in juvenile rainbow trout (Oncorhynchus mykiss). Journal of Experimental Biology, 206, 1779–1790. Shoup, D. E. & Wahl, D. H. 2009. The effects of turbidity on prey selection by piscivorous largemouth bass. Transactions of the American Fisheries Society, 138, 1018–1027. Smith, J. J., Leduc, A. O. H. C. & Brown, G. E. 2008. Chemically mediated learning in juvenile rainbow trout. Does predator odour pH influence intensity and retention of acquired predator recognition? Journal of Fish Biology, 72, 1750–1760. Sokal, R. R. & Rohlf, F. J. 2003. Biometry: the Principles and Practice of Statistics in Biological Research, 3rd edn. New-York: W.H. Freeman. Stankowich, T. & Coss, R. G. 2007. The re-emergence of felid camouflage with the decay of predator recognition in deer under relaxed selection. Proceedings of the Royal Society B, 274, 175–182. Van de Meutter, F., De Meester, L. & Robby, S. 2005. Water turbidity affects predator–prey interactions in a fish–damselfly system. Oecologia, 144, 327–336. Vogel, J. L. & Beauchamp, D. A. 1999. Effects of light, prey size, and turbidity on reaction distances of lake trout (Salvelinus namaycush) to salmonid prey. Canadian Journal of Fisheries and Aquatic Sciences, 56, 1293–1297. Zamor, R. M. & Grossman, G. D. 2007. Turbidity affects foraging success of drift-feeding rosyside dace. Transactions of the American Fisheries Society, 136, 167–176.

Please cite this article in press as: Ferrari, M.C., et al., Turbidity as an ecological constraint on learned predator recognition and generalization in a prey fish, Animal Behaviour (2010), doi:10.1016/j.anbehav.2009.12.006