Context-dependent extinction of threat memories: influences of healthy aging

Abstract

Although a substantial progress has been made in recent years on understanding the processes mediating extinction of learned threat, little is known about the context-dependent extinction of threat memories in elderly individuals. We used a 2-day differential threat conditioning and extinction procedure to determine whether young and older adults differed in the contextual recall of conditioned responses after extinction. On Day 1, conditioned stimuli were paired with an aversive electric shock in a 'danger' context and then extinguished in a different 'safe' context. On Day 2, the extinguished stimulus was presented to assess extinction recall (safe context), and threat renewal (danger context). Physiological and verbal report measures of threat conditioning were collected throughout the experiment. Skin conductance response (SCR data revealed no significant differences between age groups during acquisition and extinction of threat conditioning on Day 1. On Day 2, however, older adults showed impaired recall of extinction memory, with increased SCR to the extinguished stimulus in the 'safe' context, and reduced ability to process context properly. In addition, there were no age group differences in fear ratings and contingency awareness, thus revealing that aging selectively impairs extinction memories as indexed by autonomic responses. These results reveal that aging affects the capacity to use context to modulate learned responses to threat, possibly due to changes in brain structures that enable context-dependent behaviour and are preferentially vulnerable during aging.

Figure 1

Stimuli and experimental design. Threat acquisition and extinction were established on the first day (Day 1). Participants were threat conditioned in the danger context, in which the conditioned stimulus (CS+) was associated with a shock pulse on 60% of trials, while the CS− was not associated with any consequence. Extinction followed this phase, during which both CSs were presented within the safe context and none of them was associated with the shock pulse. Extinction recall and threat renewal were administered on the second day (Day 2). The recall of extinction was tested presenting the conditioned stimuli (CSs) within the safe context (in which extinction occurred on the first day). Subsequently, renewal of threat was tested presenting CSs within the danger context (in which the threat association was learned on the first day). On the second day, all CSs were presented in absence of the shock pulse.

Figure 2

Skin conductance responses. Graphs illustrate mean skin conductance responses (SCRs) to the conditioned (CS+) and neutral (CS−) stimuli during early and late blocks, in young (A) and older (B) participants on Day 1 (threat acquisition and extinction phase) and Day 2 (extinction recall and threat renewal phase). Data demonstrate no effect of aging on threat acquisition and extinction on Day 1. In contrast, only older participants failed to recall the previous extinction in the safe context on Day 2, while young participants specifically adapted their conditioned responses according to the context. Error bars represent standard error.

Figure 3

ΔSCR (calculated by subtracting SCR to CS− from SCR to CS+) during early extinction recall and threat renewal (Day 2). While young participants adjusted their psychophysiological response based on the context, old participants show a similar activation regardless of the contextual information. Error bars represent standard error.

Figure 4

Subjective fear ratings. Graphs illustrate the level of self-reported fear to the conditioned stimuli during the experimental phases in young (A) and older (B) participants. Error bars represent standard error.

Figure 5

Impact of neuropsychological variables. Regression analysis reported a significant influence of selective visual attention, as assessed by the Attentional Matrices Test, on ΔSCR measured during early extinction recall phase.