Reactivating aversive memories in humans: An EEG and Microstates study of post-retrieval processes

Although fear conditioning is one of the most widely used models to study anxiety in humans, much remains to be explored regarding its underlying neural correlates. It is known that the retrieval of consolidated memories can trigger modifications—weakening, strengthening, or updating of the original memory—depending on the cues presented during the reminder.

In this study, we employed a three-day threat-conditioning protocol in which an angry face (conditioned stimulus, CS+) was paired with an aversive sound (unconditioned stimulus, US). Our main objective was to identify neural markers of post-retrieval processes triggered by the presentation of the CS+ alone, 24 hours after acquisition. Additionally, we aimed to assess cognitive biases by examining how the perception of the conditioned stimulus changes following conditioning.

To this end, we recorded resting-state electroencephalographic (EEG) activity using a 30-channel system at a 256 Hz sampling rate. We analyzed four minutes of resting-state EEG following the presentation of the reminder cue, focusing on memory-related patterns in the frequency domain and microstate dynamics. We compared neural activity after the reminder presentation between two groups: a Reactivation group previously exposed to threat conditioning on the first day and a Control group with no prior conditioning.

Our results confirmed successful conditioning (CS+–US association) and memory retention 48 hours later, as evidenced by higher skin conductance responses and greater US expectancy to the CS+. Moreover, participants rated the CS+ face as more aversive than the CS− face 48 hours after conditioning, compared to their pre-conditioning ratings. Relative to the Control group, the Reactivation group showed decreased beta-band activity (25–30 Hz) in central regions during the 90 seconds following the reminder, which may reflect enhanced post-reactivation processing of the memory, in contrast to participants who saw the face for the first time. Previous research has highlighted the role of reduced beta amplitude in the encoding and retrieval of another type of memory: episodic memory. Furthermore, when analyzing EEG microstates (short-lived, quasi-stable topographies of brain activity that reoccur over time), we found lower global field power for microstates C, A, E, and B in the Reactivation group compared with the Control group.

Our findings provide initial evidence for neural correlates associated with post-retrieval processes, including decreased beta activity and microstate-level differences related to the post-retrieval processing of an implicit memory.