Fig. 2.

Experiment 1: Free-recall task with added context. a Participants (n = 15) studied lists of words in contexts distinguished by different pictures. b We probed how these contexts affect performance on a short-term recall task under three conditions: (1) when working memory was not disrupted, (2) briefly disrupted (break distraction), or (3) completely disrupted (full distraction). c Participants made more errors in the distraction conditions compared to the no distraction condition (p < .01 for all comparisons,  paired, two-sided t tests). *p < .05, **p < .01, ***p < .001. Black horizontal lines within boxes indicate median substitutions. Bottom and top edges of the box indicate the 25th and 75th percentiles. Whiskers extend to the most extreme data points not considered outliers. Black points outside boxes indicate outliers. d Within each interference condition, left bars reflect subject data and right bars reflect simulated data based on randomized substitutions from the experiment’s word set. In all three conditions, participants made errors that reflected the influence of reinstated context. Specifically, participants substituted words from the previous trial at a higher rate than would be expected if they were randomly substituting words previously learned in the experiment. As computed by bootstrap analysis, the number of previous trial substitutions was greater than chance on full-interference (p < .001), break-interference (p < .001), and no-interference trials (p < .001). e Participants also made substitution errors during recall that reflected the encoding context of the target set, or same-context errors, at a higher rate than would be expected if they were randomly substituting words previously learned in the experiment. As computed by bootstrap analysis, the amount of same-context errors made was greater than chance on full-interference (p = .001), break-interference (p = .001), and no-interference trials (p = .025). Box plots follow the same conventions as in d