Malaria parasites have been shown to decrease host fitness in several species in the wild and their detrimental effects on host cognitive ability are well established in humans. However, experimental demonstrations of detrimental effects on non-human host behaviour are currently limited. In this study, we experimentally tested whether injections of an anti-malaria drug affected short-term behavioural responses to a problem-solving task during breeding in a wild population of great tits (Parus major) naturally infected with malaria. Adult females treated against malaria were more active than control females, even though they were not more likely to solve the task or learn how to do so, suggesting that energetic constraints could shape differences in some behaviours while changes in cognitive performances might require more time for the neural system to recover or may depend mainly on infection at the developmental stage. Alternatively, parasite load might be a consequence, rather than a cause, of inter-individual variation in cognitive performance. These results also suggest that inter-individual as well as inter-population differences in some behavioural traits may be linked to blood parasite load.