Neurophysiological signature of working memory updating during encoding

Working memory (WM) has been extensively studied in cognitive psychology, with numerous tasks developed to measure its capacity. Among these, binding and updating tasks that assess the ability to build, maintain, and manipulate temporary relational representations have been shown to be particularly suitable for assessing WM capacity (WMC). In the present study, we investigated the specific processes involved in updating during WM encoding by comparing the electrophysiological signals in binding and updating tasks. In addition, we compared specific forms of updating demands by contrasting substitution and transformation processes. Finally, we examined whether individual differences in updating activity are related to WM performance. We recruited a heterogeneous sample of 151 participants (female = 89, M_age = 34.77, SD = 12.78) who completed a binding and two updating tasks while an EEG was recorded. Mass univariate cluster-based permutation analyses of event-related potentials consistently showed a temporally and spatially widespread cluster when comparing updating and binding. This cluster was characterized by an increased positivity associated with updating, appearing predominantly in the frontal regions during early time windows (approximately 130-350 msec after stimulus onset) and later shifting to the parietal regions (from approximately 350 msec onward). Additionally, when comparing transformation and substitution, we found an increase in positivity associated with transformation, peaking within a typical P3 time window (approximately 400 msec). Finally, we found that only individual differences in transformation-related activity were associated with individual differences in WMC.