There is growing evidence that individuals actively assess the match between their phenotype and their environment when making habitat choice decisions (so-called matching habitat choice). However, to our knowledge, no studies have considered how the social environment may interact with social phenotype in determining habitat choice, despite habitat choice being an inherently social process and growing evidence for individual variation in sociability. We conducted an experiment using wild great and blue tits to understand how birds integrate their social phenotype and social environment when choosing where and how to feed. We used programmable feeders to (i) record social interactions and estimate social phenotype and (ii) experimentally manipulate the local density experienced by birds of differing social phenotype. By tracking feeder usage, we estimated how social environment and social phenotype predicted feeder choice and feeding behaviour. Both social environment and social phenotype predicted feeder usage, but a bird’s decision to remain in a particular social environment did not depend on their social phenotype. In contrast, for feeding behaviour, responses to the social environment depended on social phenotype. Our results provide rare evidence of matching habitat choice and shed light on the dependence of habitat choice on between-individual differences in social phenotype.

We quantified the social phenotypes of wild RFID-tagged great and blue tits via social network analysis, with social networks estimated using records of visits to RFID feeders in the winter of 2020–2021. We then allocated less social and more social birds (i.e. birds with low and high weighted degrees) to either a high-density or a low-density feeder at their pre-experimental location. We subsequently followed their usage of each feeder in the pair to understand whether their choices over where and how to feed were predicted by their social phenotype, experimental social environment (high vs low density) or a combination of the two, in order to understand whether they displayed social phenotype-dependent habitat choice.