Do personalities co-vary with metabolic expenditure and glucocorticoid stress response in adult lizards?

Stable differences in physiology among individuals may facilitate the evolution of consistent individual differences in behavior. In particular, according to the pace-of-life syndrome (POLS) hypothesis, individual variation in metabolic expenditure and stress physiology should be linked with exploration, aggression, or risk-taking behaviors. Previous studies have uncovered stable individual differences in metabolic expenditure and circulating glucorticoids in common lizards (Zootoca vivipara). We tested the correlations between standard metabolic rates (SMR), glucorticoid stress response, and behavioral traits (activity, aggressiveness, risk taking, and sociability) in males. In ectotherms, the thermal dependence of SMR should be included in the POLS hypothesis; we therefore measured SMR at three temperatures from rest to preferred body temperature. Activity, aggressiveness, and risk taking, but not sociability, exhibited significant, short-term repeatability, and little correlation was found between behavioral traits. The SMR of lizards with a low metabolism at rest increased faster with body temperature. The SMR at rest was negatively correlated with behavioral variation in sociability and activity but not with risk-taking behavior. In addition, the plasma corticosterone level after an acute handling stress increased slightly but not significantly with aggressiveness. We discuss alternative interpretations for these relationships and conclude that the link between inter-individual variation in physiology and behavior is trait-dependent in the common lizard. Selection better promotes the evolution of consistent differences in behavior, or personalities, when they are coupled with differences in physiology. In adult common lizards, inter-individual differences in metabolic expenditure and glucocorticoid stress response are consistent and could play a crucial role in the maintenance of personalities. This study supported partially this hypothesis. We found that more sociable and active personality types had a lower metabolic expenditure, while more aggressive personality types tended to have a higher physiological stress response. At the same time, physiology was not correlated with individual differences in risk-taking behavior and drove little part of behavioral variation. The coupling between personalities and physiology appears to be trait-dependent, suggesting that behaviors may be relatively free to evolve independently from physiology.