The effect of cortisol on emotional responses depends on order of cortisol and placebo administration in a within-subject design

Abstract

Cortisol does not exhibit a straightforward relationship with mood states; administration of glucocorticoids to human subjects has produced mixed effects on mood and emotional processing. In this study, participants (N=46) received intravenous hydrocortisone (synthetic cortisol; 0.1mg/kg body weight) and placebo in randomized order over two sessions 48h apart. Following the infusion, participants rated neutral and unpleasant pictures. In Session 1, participants reported elevated negative affect (NA) following the picture-rating task, regardless of treatment. In Session 2, however, only participants who received cortisol (and thus who had received placebo in Session 1) reported elevated NA. Arousal ratings for unpleasant pictures followed a similar pattern. These findings suggest that the effects of cortisol on emotion vary based on situational factors, such as drug administration order or familiarity with the tasks and setting. Such factors can influence cortisol's effects on emotion in two ways: (A) cortisol may only potentiate NA and arousal ratings in the absence of other, overwhelming influences on affect, such as the novelty of the setting and tasks in Session 1; and (B) cortisol in Session 1 may facilitate learning processes (e.g., habituation to the stimuli and setting; extinction of aversive responses) such that emotional responses to the pictures are lessened in Session 2. This interpretation is compatible with a body of literature on the effects of glucocorticoids on learning and memory processes.

Figure 1

Timeline of events in each study session.

Figure 2

Effect on plasma cortisol levels of IV hydrocortisone infusion, 0.1 mg/kg body weight, compared to 0.9% saline infusion. Plasma sample #1 (−40 min) was collected immediately after placement of intravenous lines at ~ 1700h. Infusion (shown by black bar) began at ~ 1740h (0 min) and lasted 30 minutes. Error bars represent standard error of the mean (SEM). * indicates significant difference (p < 0.05) in post-hoc test.

Figure 3

PANAS-NOW negative affect (NA; Figure 3A) and positive affect (PA; Figure 3B) scores at each of the seven measurement points throughout the study sessions, by drug and session (S1 = Session 1; S2 = Session 2). Error bars represent SEM. White bars indicate placement of intravenous lines; black bars indicate infusion. The Picture Task occurred during the latter half of the infusion (black bar). “Cortisol-first” group, N=24; “placebo-first” group, N=23. Figure 3C shows PANAS NA scores at PANAS administration #4, immediately after the Picture Task, by drug and session. Error bars represent SEM. Left, participants who received cortisol in Session 1 and placebo in Session 2; right, participants who received placebo in Session 1 and cortisol in Session 2. * indicates significant difference in post-hoc test, p < 0.05; ** indicates p < 0.01. There is no significant difference between Session 1 and 2 NA in the group who received placebo first (right two bars).

Figure 3

PANAS-NOW negative affect (NA; Figure 3A) and positive affect (PA; Figure 3B) scores at each of the seven measurement points throughout the study sessions, by drug and session (S1 = Session 1; S2 = Session 2). Error bars represent SEM. White bars indicate placement of intravenous lines; black bars indicate infusion. The Picture Task occurred during the latter half of the infusion (black bar). “Cortisol-first” group, N=24; “placebo-first” group, N=23. Figure 3C shows PANAS NA scores at PANAS administration #4, immediately after the Picture Task, by drug and session. Error bars represent SEM. Left, participants who received cortisol in Session 1 and placebo in Session 2; right, participants who received placebo in Session 1 and cortisol in Session 2. * indicates significant difference in post-hoc test, p < 0.05; ** indicates p < 0.01. There is no significant difference between Session 1 and 2 NA in the group who received placebo first (right two bars).

Figure 4

Average arousal ratings for unpleasant (Figure 4A) and neutral (Figure 4B) IAPS pictures in the Picture Task, by drug and session. Error bars represent SEM. Left, participants who received cortisol in Session 1 and placebo in Session 2, N=23. Right, participants who received placebo in Session 1 and cortisol in Session 2, N=22. * indicates significant difference in post-hoc test, p < 0.05. In Figure 4A (ratings for unpleasant pictures), post-hoc test indicates a trend level difference between Session 1 and Session 2 arousal ratings for those who received placebo first, p < 0.06.