<p>Basking in groups with conspecifics is extensively documented in reptiles, and in some clades, notably Scincidae, basking aggregations can be representative of repeatable and stable social kin groups. Evidence for social basking in non-squamate reptiles is less reported, but mounting evidence suggests that social behaviour in turtles may be common. To elucidate the social drivers of basking aggregations, we recorded basking behaviour in a population of Midland Painted Turtles (<i>Chrysemys picta marginata</i>) over two active seasons. We explored potential factors influencing associations during basking aggregations, and tested whether social basking is impacted by individual traits that may reflect dominance hierarchies. We used network analysis to construct matrices of social association between basking individuals, and explored spatial overlap, genetic relatedness, body size, and sex as possible predictors of these population-level associations. We also quantified individual social behaviour by calculating network metrics and examined whether body size and a proxy of boldness influenced individual-level social behaviour. We found no support for spatial overlap, kinship, body size, or sex similarity as drivers of social associations while basking. Turtles did not generally associate with one another, but when they did, close to the majority of association were preferred (i.e., non-random). Further, we found that body size and a proxy of boldness affected the social behaviour of female turtles. These individual-level findings suggest that a dominance hierarchy may influence social structure in basking aggregations of Painted Turtles. Our findings have implications for understanding the evolution of cryptic sociality, and call for more thorough examination of social organization in a wider range of non-avian reptiles.</p>