Learning threat-sensitive predator avoidance: how do fathead minnows incorporate conflicting information?

For prey species, the costs associated with antipredator behaviours should drive the development of efficient risk assessment systems. According to the threat-sensitive predator avoidance hypothesis, prey match the intensity of their antipredator responses with the degree of threat posed. Evidence of threat sensitivity exists for a variety of taxa, but little is known about how such responses develop. We investigated the effect of multiple conditioning events on the development of threat-sensitive antipredator responses of fathead minnows, Pimephales promelas. In experiment 1, we conditioned predator-naive minnows with brook charr, Salvelinus fontinalis, odour paired with either a high or low concentration of conspecific alarm cues. Fish were conditioned twice (low/low, low/high, high/low, high/high), then tested for learned recognition of charr odour alone. Results suggest that minnows learn the relative risk associated with the predator based on the highest alarm cue concentration experienced during the conditioning trials. In experiment 2, predator-naive minnows were conditioned six times with charr odour paired with either a high or low concentration of alarm cues (6 low, 5 low/1 high, 1 high/5 low, 6 high). The intensity of the minnows' antipredator response to charr odour was not an averaging of the intensity of responses during the six conditioning trials. Instead, they used their last conditioning experience to adjust the intensity of their response to the predation cues. Predation fluctuates in space and time. Thus, the best way to adaptively respond to predator cues is to constantly update information regarding the relative risk associated with a given predator.