Differential effects of stress-induced cortisol responses on recollection and familiarity-based recognition memory

Abstract

Stress-induced changes in cortisol can impact memory in various ways. However, the precise relationship between cortisol and recognition memory is still poorly understood. For instance, there is reason to believe that stress could differentially affect recollection-based memory, which depends on the hippocampus, and familiarity-based recognition, which can be supported by neocortical areas alone. Accordingly, in the current study we examined the effects of stress-related changes in cortisol on the processes underlying recognition memory. Stress was induced with a cold-pressor test after incidental encoding of emotional and neutral pictures, and recollection and familiarity-based recognition memory were measured one day later. The relationship between stress-induced cortisol responses and recollection was non-monotonic, such that subjects with moderate stress-related increases in cortisol had the highest levels of recollection. In contrast, stress-related cortisol responses were linearly related to increases in familiarity. In addition, measures of cortisol taken at the onset of the experiment showed that individuals with higher levels of pre-learning cortisol had lower levels of both recollection and familiarity. The results are consistent with the proposition that hippocampal-dependent memory processes such as recollection function optimally under moderate levels of stress, whereas more cortically-based processes such as familiarity are enhanced even with higher levels of stress. These results indicate that whether post-encoding stress improves or disrupts recognition memory depends on the specific memory process examined as well as the magnitude of the stress-induced cortisol response.

Keywords: Cortisol; Familiarity; Memory; Recollection; Stress.

Figure 1

Schematic depiction of the protocol. Time values represent the mean latency to begin each task (measured from the start of each session).

Figure 2

Mean salivary cortisol for the stress (blue) and control (red) groups at each sample. Error bars represent SEs of the means.

Figure 3

Observed memory parameters plotted against observed cortisol responses, for subjects in the stress group (circles) and control group (x’s). Recollection (top row) and Familiarity (bottom row) are plotted separately for neutral (left column) and negative images (right column) and are fit to quadratic and linear functions respectively.

Figure 4

Observed memory parameters plotted against observed baseline cortisol, for subjects in the stress group (circles) and control group (x’s). Recollection (top row) and Familiarity (bottom row) are plotted separately for neutral (left column) and negative images (right column), and are fit to linear functions.

Figure 5

Schematic depiction of the effects of post-encoding stress-related increases in cortisol on hippocampal and cortical dependent memory processes.