What Matters Most: Selective Memory Enhancement Surrounding Acute Stress

Abstract

When stress occurs, this changes our memory for events around the stressor. These stress-driven changes in memory are not universal, but depend on various factors. Nevertheless, the factors driving this selectivity remain elusive. The behavioural tagging hypothesis offers a framework for understanding the etiology of such selectivity. Based on this framework, we proposed two mechanisms that may tag encoded information, allowing for selective memory enhancement if a stressful encounter occurs in the temporal periphery. Study I investigated whether a cue to a stressful context yields memory improvements for subsequently encoded information. While we did not find selective memory improvement for information that cued a stressful experience, we found a general memory improvement for material encoded before stress, specifically when the stressful experience was well represented in memory. In studies IIa and IIb, we explored whether stress selectively enhances the memory for information associated with a prediction error (PE). In Study IIa, we observed selective memory enhancement for high-PE information encoded both before and after the stressor. In Study IIb, the effect was specific to information encoded after the stressful encounter. These findings highlight that the influence of stress on memory is selective, and PE stands out as potential "gating" mechanism through which stress amplifies specific memories. Such selective memory enhancement can be considered adaptive based on the predictive brain hypothesis – when stress occurs, this may encourage the prioritization of specific information that helps to make better predictions in the future. These findings may be leveraged in educational settings and help us understand the development of strong post-traumatic memories.