Extending the recovery window: Effects of trait rumination on subsequent evening cortisol following a laboratory performance stressor

• We extended the post-stressor recovery window into the subsequent evening.
• Trait rumination predicted greater total cortisol into the evening of the stressor.
• Rumination–cortisol patterns are complex, differing for individuals and by context.
• Rumination–cortisol links are more pronounced for social-evaluative stressors.
• Different cortisol patterns emerged for men and women who tend to ruminate.

SummaryMental rehearsal of past stressors through rumination may extend the physiological stress response and exposure to stress-related physiological mediators, such as cortisol. If repeated over time, this prolonged activation may contribute to a number of chronic health conditions. Findings from the emerging literature on the tendency to ruminate and its association with cortisol have been somewhat mixed. In the present study, we tested whether trait rumination predicted elevated cortisol concentrations in response to a performance stressor, and whether this association varied by the social-evaluative context of the stressor and gender. We also examined whether associations persisted into the evening of the stressor. Participants (50% female; mean age = 19.83, SD = 1.62) were randomly assigned to complete a laboratory speech stressor either in a social-evaluative (SET; n = 86) or non-evaluative context (non-SET; n = 58). Salivary cortisol concentrations were measured throughout the laboratory visit and later that evening. There was a main effect of trait rumination on greater total cortisol exposure into the evening of the stressor. In addition, trait rumination interacted with stressor context to predict cortisol declines: on the night of the SET stressor, high trait ruminators did not exhibit typical declines in cortisol. Different cortisol patterns emerged for men and women with tendencies to ruminate: women with higher rumination scores had flatter cortisol slopes with greater evening cortisol, whereas men with higher trait rumination scores had greater initial cortisol reactivity to the stressor. Together, these findings suggest that the relationship between the tendency to ruminate and cortisol concentrations is qualified by individual differences (gender) and stressor characteristics (social-evaluative threat).