Individual variation in spontaneous problem-solving performance among wild great tits

Behavioural traits generally and cognitive traits in particular are relatively understudied in an evolutionary ecological context. One reason for this is that such traits are often difficult to characterize among large numbers of individuals, without the influence of diverse environmental effects swamping intrinsic individual differences. We conducted standardized assays on a natural population of great tits, Parus major, to quantify and characterize individual variation in problem-solving performance, a simple cognitive trait often linked to innovative foraging ability. Forty-four per cent of 570 birds solved a food-motivated, lever-pulling problem and this proportion was consistent across three seasons. Individual performance was consistent within and across captivity sessions, across seasons, and between two different problem-solving tasks (lever and string pulling). Problem-solving performance was not explained by differences in latency to approach the empty task, nor latency to feed after human disturbance. Variation was unrelated to body condition, while age and natal origin explained significant but minimal amounts of variation, the importance of which varied between seasons. Problem-solving performance did not covary with exploration behaviour of a novel environment, suggesting that individual differences in problem solving represent an independent source of behavioural variation in our population. Rather than simply reflecting covariance with state or with other behavioural traits, our results suggest that variation in problem-solving performance represents inherent individual differences in the propensity to forage innovatively. We suggest that standardized problem-solving assays may prove ideal for studying the evolutionary ecology of simple cognitive traits.