Uncertainty matters under high predation risk: the impact of ambient predation and recreational disturbance history in Trinidadian guppies.

Abstract Within aquatic ecosystems, localized anthropogenic activities may disrupt the availability and reliability of publicly available information, leading to increased uncertainty among prey. In turn, prey may reduce their level of uncertainty by actively increasing behavioural activities that allow for greater information gathering. The impact of uncertainty on prey behaviour is hypothesized to be greater under conditions of higher compared to lower ambient predation risk, potentially leading to a greater need to gather information regarding local conditions. Here, we test this hypothesis using in situ and laboratory trials. Initially, we quantified inspection behaviour following the presentation of an ambiguous stimulus (rock drop) in anthropogenically disturbed (vs. upstream undisturbed) sites in a high and a low predation population. Guppies at disturbed sites within the high predation population had shorter inspection latencies, inspected at higher rates, and were more likely to remain following the stimulus presentation compared to the paired undisturbed site. Disturbance had no effect on the inspection pattern of guppies within the low predation population. For our laboratory trials, we collected guppies from disturbed and undisturbed sites within the same rivers used for the in situ trials. Under laboratory conditions, we quantified space use and exploration within a novel arena. We found that guppies from disturbed (compared to paired undisturbed) sites within the high predation risk population were more active in their exploration of the novel arena (increased line crosses, less time freezing, more even space use). We also found that this pattern was reversed for guppies collected from the low predation river. Combined, our results support the hypothesis that the impact of uncertainty among prey may be increased by predation pressure.