The effects of pharmacologically manipulated cortisol levels on memory for emotional and neutral information

The effects of elevations in cortisol levels on memory were assessed. Based on the actions of cortisol in the hippocampus and amygdala, it was predicted that a low dose of hydrocortisone (which is identical to cortisol) would facilitate explicit memory, that a high dose would impair explicit memory, but that neither dose would affect implicit memory or vigilance. Differential effects of hydrocortisone on memory for neutral vs. negative information were also tested. Ninety male participants were given placebo, a 20 mg dose, or 40 mg dose of hydrocortisone in the evening when endogenous cortisol levels are minimal. Memory for stimuli presented 45 minutes after drug administration was tested on the same evening and two evenings later when cortisol levels were no longer manipulated. Explicit memory for negative and neutral words and pictures was tested using recognition memory and free recall tasks. Implicit memory for words was tested using a word fragment completion task, and vigilance was tested using a continuous performance task. Hydrocortisone had no effect on implicit memory or vigilance. On the evening of drug administration, participants who received hydrocortisone made fewer errors on free recall tasks for both words and pictures than did the placebo group. Two evenings later, participants in the 20 mg group performed better than participants in both the placebo group and the 40 mg group on word and picture recognition memory. Tests of correlations between memory performance and observed salivary cortisol levels revealed that, within the 20 mg group, higher cortisol levels predicted worse memory performance. These correlations were typically accounted for by memory for negative rather than neutral stimuli. Thus, the 20 mg group as a whole performed better than the 40 mg and placebo groups, but within the 20 mg group, higher cortisol levels predicted worse performance, especially for the negative stimuli. This implies that the peak of dose-response curve for the relation between cortisol and memory performance occurred in the lower range of cortisol elevations within the 20 mg group, and that the decline in memory associated with higher levels of cortisol was more pronounced for negative than neutral stimuli.