Cue overlap supports preretrieval selection in episodic memory: ERP evidence

Abstract

People often want to recall events of a particular kind, but this selective remembering is not always possible. We contrasted two candidate mechanisms: the overlap between retrieval cues and stored memory traces, and the ease of recollection. In two preregistered experiments (Ns = 28), we used event-related potentials (ERPs) to quantify selection occurring before retrieval and the goal states — retrieval orientations — thought to achieve this selection. Participants viewed object pictures or heard object names, and one of these sources was designated as targets in each memory test. We manipulated cue overlap by probing memory with visual names (Experiment 1) or line drawings (Experiment 2). Results revealed that regardless of which source was targeted, the left parietal ERP effect indexing recollection was selective when test cues overlapped more with the targeted than non-targeted information, despite consistently better memory for pictures. ERPs for unstudied items also were more positive-going when cue overlap was high, suggesting that engagement of retrieval orientations reflected availability of external cues matching the targeted source. The data support the view that selection can act before recollection if there is sufficient overlap between retrieval cues and targeted versus competing memory traces.

Selective memory retrieval revealed by electroencephalographic (EEG) neural pattern reinstatement

“…Neural reinstatement emerged by 500 ms post-stimulus, as did the established left parietal event-related potential (ERP) signature of recollection. Reinstatement was also target-selective (greater for targets than non-targets) when test cues overlapped more with targets, as had previously been shown for the left parietal ERP (Moccia and Morcom, 2021). In contrast, when cues overlapped more with non-targets, neural reinstatement was non-selective or reversed, unlike the left parietal ERP.…”

“…However, the separate decoding of target and non-target patterns did not allow us to test whether reinstatement was target-selective in the other two conditions, which had shown target-selective left parietal ERPs. In the Target-Audio block of Experiment 1 neither target nor non-target reinstatement was separately significant (p = .197 and .329), while in the Target-Picture block of Experiment 2, reinstatement was only significant for targets (p < .001; .259 for non-targets) (see Moccia and Morcom, 2021). The next two analysis steps allowed us to directly compare the amount of target and non-target reinstatement (see Reinstated neural patterns are format-specific and Target-selective reinstatement tracks cue overlap in main manuscript).…”

“…In both Experiment 1, when test cues were words, and in Experiment 2, when test cues were line drawings, memory tended to be better for items studied as pictures, whether tested as targets or non-targets. Response times (RTs) were also faster for targets studied as pictures than audios in Experiment 1, and for all trial types in Experiment 2 (see supplementary Table S1; for statistical analyses see Moccia and Morcom, 2021).…”

“…Here, we examined two existing datasets for which the pattern of left parietal ERP target selectivity was known. In the previous report, the left parietal ERP was larger for targets than non-targets when the retrieval cues had a high degree of overlap with the targeted material (auditory words in Experiment 1, and pictures in Experiment 2; Moccia & Morcom, 2021). To assess the locus of selection, we investigated whether neural reinstatement is target-selective in the same task conditions in which the left parietal ERP is selective.…”

“…All reported having normal or corrected to normal vision and hearing, being in good health, not taking medication that might affect cognition, and being very fluent in English. Sample sizes were determined a priori for comparisons of left parietal ERPs (see Moccia & Morcom, 2021). The Psychology Research Ethics Committee at the University of Edinburgh approved the experiments, ref.…”