“Convergent-validity” of standardized assays and behavioural observations in North American red squirrels

Animal personality (i.e., consistent individual differences in behaviour) is measured through standardized behavioural assays, or, more rarely, using behavioural focals, with little overlap between these study designs existing to date. Due to this, it is unclear if animal personalities measured in standardized environments can accurately predict behaviour in the wild. Using long term data from a population of red squirrels in the Yukon, we tested the convergent validity between standardized behavioural assays and focal animal sampling using 88 individuals. If personality measured in a standardized assay represents true personality, then each individual is expected to have similar scores for aggression and activity independent of the method used to measure personality. Using both correlations and linear regressions, we found no link between the personality measures from the assays and the focals. We suggest that this lack of convergent validity is due to the influence of the environment on focal behaviours, and that focals are not accurate measures of personality. Our results indicate that these two methods of measuring personality are not interchangeable, and future studies should be wary of which method they choose. This research is the first empirical test of the validity of standardized measures to score personality traits.