Teleosts exhibit inter- and intra-specific variation in the size and shape of their brains. Interpopulation differences in gross brain morphology among numerous teleost fish species have been observed and have been partially attributed to plastic changes in response to their environment, including predation. These differences manifest themselves macroscopically, potentially because teleosts retain the capacity for active neuroproliferation into adulthood. Building on previous work, showing chronic exposure to predation can affect brain morphology, we sought to determine whether these differences manifest themselves on a time scale shown to induce phenotypically plastic behavioural changes. In separate trials, we held northern redbelly dace (Phoxinus eos (Cope, 1861) = Chrosomus eos Cope, 1861) and juvenile Atlantic salmon (Salmo salar Linnaeus, 1758) in semi-natural conditions and exposed them to conspecific skin extract as a proxy for predation risk over 2 weeks. After exposure, their brains were excised, photographed, and analyzed for size (multivariate ANOVA) and shape (Procrustes ANOVA). Despite their brief exposure to simulated predation pressure, subjects from both species developed significantly different brain morphologies. Compared with controls, the Atlantic salmon exhibited a different brain shape and smaller optic tecta, whereas the northern redbelly dace had larger brains with more developed olfactory bulbs and optic tecta. Our results highlight the rapidity with which external environment can alter patterns of growth in the brain.

Teleost fish are neuroplastic and are known to alter their brain morphology and behaviour in response to environmental change such as an increase in predation pressure. The hypothalamus plays a key role in regulating behavioural responses to predation risk. In this study, wild-caught northern red bellied dace (Chrosomus eos) developed smaller and less symmetric hypothalami when held in captivity for 14 days; both measures correlated with boldness in a latency to emerge test. This study's results highlight the potential impact of short-term holding conditions on brains and behaviour.

Recent studies have documented that exposure to high levels of background risk can induce neophobic predator avoidance in prey animals, whereby they respond to any novel cue with an anti-predator response. Such phenotypically plastic predator avoidance may allow prey to maximize anti-predator benefits in variable risk environments. It remains poorly understood whether risk assessment information from different sensory modalities can be integrated to induce generalized, cross-sensory system neophobic responses. Here, we directly test this hypothesis by exposing juvenile convict cichlids (Amatitlania nigrofasciata) to high- versus low-risk environments using either conspecific alarm cue (chemosensory risk) or a model avian predator (visual/mechanical risk) and testing their response to a novel chemosensory cue (Experiment 1) or visual cue (Experiment 2). Our results suggest that regardless of the sensory modality used to increased perceived risk, cichlids pre-exposed to high-risk conditions exhibited increased predator avoidance in response to any novel visual or chemical cue. As expected, cichlids pre-exposed to low-risk conditions did not display any neophobic responses. Our results suggest that induced neophobia is not cue specific; rather, it may function as a generalized response to perceived predation risk.

Exposure to conditions of elevated predation risk, even for relatively short periods, has been shown to induce neophobic responses to novel predators. Such phenotypically plastic responses should allow prey to exhibit costly anti-predator behaviour to novel cues only in situations where the risk of predation is high. While there is evidence that the level of background risk shapes the strength of induced neophobia, we know little about how long neophobic responses are retained. Here we exposed juvenile convict cichlids (Amatitlania nigrofasciata) to three background levels of short-term background risk and then tested their responses to novel predator odours. Cichlids exposed to low risk did not show neophobic responses, while those exposed to intermediate and high risk did. Using extinction trials, we demonstrate that the retention of neophobic responses is greater among cichlids exposed to high versus intermediate predation risk conditions. Moreover, we found much longer retention of the neophobic responses when cichlids were tested a single time compared to when they were tested repeatedly in the extinction trials. This work supports the prediction that neophobic responses to specific odours are relatively long lasting but can quickly wane if the cues are experienced repeatedly without them being associated with risk. It is clear that background level of risk and the frequency of exposure to novel cues are crucial factors in determining the retention of risk-related information among prey.

No abstract