The repeatability of cognitive performance: a meta-analysis

Selection acts on heritable individual variation in behaviours. Both behavioural and cognitive processes play important roles in mediating an individual's interactions with their environment. Yet, while there is a vast literature on repeatable individual differences in behaviour, relatively little is known about the repeatability of cognitive performance. To further our understanding of the evolution of cognition we gathered 44 datasets on individual performances of 25 species and used meta-analysis to evaluate whether cognitive performance is repeatable across six animal classes. We assessed repeatability (R) in performance (1) on the same task presented at different time intervals (temporal repeatability), and (2) on different tasks that measure the same putative cognitive ability (contextual repeatability). We also addressed whether R estimates are influenced by seven extrinsic factors (moderators): type of cognitive task, type of measurement, delay between tasks, origin of the subjects, experimental context, taxonomic class and if the R value was published or unpublished. We found support for both temporal and contextual repeatability of individual variation in cognitive performance, with significant mean R estimates ranging between 0.15 and 0.28. R estimates were mostly influenced by the type of cognitive performance measures and the fact that R values was published, none of the other moderators showed consistent and significant impacts on repeatability estimates. Our overall findings highlight the widespread occurrence of consistent inter-individual variation in cognition which, like behaviour, may have fitness implications.