Observing the suppression of individual aversive memories from conscious awareness

Abstract

When reminded of an unpleasant experience, people often try to exclude the unwanted memory from awareness in an effort to forget it, a process known as retrieval suppression. Yet, how fast can individual memories be targeted and controlled, and the neural dynamics in modulating cortical traces of individual memories, remain elusive. Here, using multivariate decoding analyses on time-domain and time-frequency-domain EEGs, we found that retrieval suppression of aversive memories was distinct from retrieval and passive viewing, when given a reminder. Specifically, early elevation of mid-frontal theta power during the first 500 ms distinguished retrieval suppression from passive viewing, suggesting that suppression recruited early active control processes. On an item-level, we could discern activities relating to individual memories during active retrieval-initially, based on perceptual responses to reminders (0-500 ms) and later, via the reinstatement and maintenance of the target aversive scenes (500-3000 ms). Critically, suppressing retrieval significantly weakened (during 420-600 ms) and eventually abolished these item-specific cortical patterns till cue disappeared (1200-3000 ms), suggesting the successful exclusion of the unwelcome memory from awareness. Suppression of item-specific cortical patterns bore behavioral consequences in predicting subsequent episodic forgetting. These findings provide unique insight into the neural dynamics underlying the control of unwelcome memories: upon perceiving an unwelcome reminder, people rapidly deploy inhibitory control to truncate retrieval within 500 ms, which likely terminate the reminder-to-memory conversion at around 500 ms that would ordinarily arise through hippocampal pattern completion. We concluded that both rapid and sustained control are critical in abolishing cortical patterns of individual memories, limiting unwelcome awareness, and precipitating later forgetting.

Attentional capture mediates the emergence and suppression of intrusive memories

“…These results corroborate observations that inhibitory control during memory suppression abolishes the item-specific EEG patterns associated with memory reinstatement. 50 Moreover, non-intrusive cues that remained stable in subsequent suppression attempts were associated with a weaker memory strength than those that relapsed in subsequent blocks. These differences also occurred in time windows that both preceded and followed the peak of decoding activity.…”

“…Previous studies have relied on event-related potentials (ERPs) 43 , 44 , 45 , 46 or time-frequency representations 47 , 48 , 49 to characterize the oscillatory signature of inhibitory control. 50 However, these studies did not track the subjective reports associated with memory intrusion, leaving the alleged role of an interference signal in adaptive forgetting untested. Moreover, including these behavioral reports is not sufficient, given that intrusive mental images are unpredictable and the precise timing of their appearance unknown, precluding inferences on the neural signature of individual intrusive mental images.…”

Suppression-induced forgetting of complex aversive scenes - A replication and meta-analysis

“…Those were identified by screening all articles that have cited the original publication until November 2023. We thus included effect sizes from three publications (8, 13, 15), one unpublished preprint (14), and our replication study (18 effect sizes in total). Individual effect sizes for the comparison of baseline and suppress conditions were entered as standard mean change using raw score standardization (SMCR).…”

“…To assess the overall effect size of suppression-induced forgetting, we conducted a three-level meta-analysis that included the 18 effect sizes from Küpper et al ( 8), Catarino et al ( 13), Lin et al (14), Meyer and Benoit (15), and our replication study. The robust metaanalysis revealed a significant moderate effect of memory suppression (g = 0.30, p = .02, SE = 0.09, 95% CI = [0.08 0.53]) (Fig.…”

“…They replicated suppression-induced forgetting of details and gist for the healthy controls and showed reduced memory suppression effects for patients. Lin et al (14) used the procedure to investigate how suppression of individual memories unfolds over time using multivariate decoding methods on EEG data. They modified several minor aspects of the procedure and found suppression-induced forgetting for the identification and details, but not for the gist of the scene.…”