Specifying the neural basis of the spacing effect with multivariate ERP

The spacing effect refers to the finding that, given a fixed amount of study time, a longer interval between study repetitions improves long-term retention (e.g., Cepeda et al., 2006; Ebbinghaus, 1885/1967; Melton, 1970). Although the spacing effect is a robust and reliable finding in the memory literature, its cognitive and neural mechanisms remain unclear. We used event-related potentials (ERPs) to investigate the neural correlates of the spacing effect in the context of the study-phase retrieval hypothesis, which posits that repeated exposure of an item serves as a reminder of one's previous experience with the item, thereby promoting long-term retention. ERPs were recorded from 30 healthy young adults as they studied pairs of words under three levels of lag, corresponding to 0, 4, or 12 intervening pairs between the first and second occurrences of a target pair. We used two study-phase tasks that differed in the degree of retrieval that was required. During the test phase, participants were tested on paired-associate recall. The results demonstrated a significant effect of spacing on memory performance. However, the effect of encoding task and the interaction between encoding task and spacing were not significant. The results of the partial least squares analyses, which are not constrained by time window or electrode selection, revealed a spacing effect on the ERP data for both study-phase tasks; this effect occurred late in the epoch and was most salient over the centro-parietal scalp region. The results add to the literature on the neural correlates of the spacing effect by providing a more comprehensive account compared to past ERP findings that were focused on testing specific ERP components. They also call for further investigation on the various theoretical accounts of the spacing effect.