Recent research has found that individuals often vary in how consistently they express their behaviour over time (i.e. behavioural predictability) and suggested that these individual differences may be heritable. However, little is known about the intrinsic factors that drive variation in the predictability of behaviour. Indeed, whether variation in behavioural predictability is sex-specific is not clear. This is important, as behavioural predictability has been associated with vulnerability to predation, suggesting that the predictability of behavioural traits may have key fitness implications. We investigated whether male and female eastern mosquitofish (Gambusia holbrooki) differed in the predictability of their risk-taking behaviour. Specifically, over a total of 954 behavioural trials, we repeatedly measured risk-taking behaviour with three commonly used assays—refuge-use, thigmotaxis, and foraging latency. We predicted that there would be consistent sex differences in both mean-level risk-taking behaviour and behavioural predictability across the assays. We found that risk-taking behaviour was repeatable within each assay, and that some individuals were consistently bolder than others across all three assays. There were also consistent sex differences in mean-level risk-taking behaviour, with males being bolder across all three assays compared to females. In contrast, both the magnitude and direction of sex differences in behavioural predictability were assay specific. Taken together, these results highlight that behavioural predictability may be independent from underlying mean-level behavioural traits and suggest that males and females may differentially adjust the consistency of their risk-taking behaviour in response to subtle changes in environmental conditions.

See README.md file for methods description.