Figure 2. Aversive learning enhanced memory for CS+ items during conditioning, but had highly variable effects on memory biases for Phase 1 items.

(a) Shock expectancy ratings for CS+ (red bar) and CS− (gray bar) trials verified that participants were able to learn which visual category was paired with shock. Values reflect the number of trials that participants indicated they expected to be shocked, broken down by CS Type (out of 30 trials each). Delayed recognition memory test results for (b) false alarm rates and (c) hit rates broken down by CS Type. (c) Corrected recognition scores were computed by subtracting participants’ false alarm rates from their hit rates for each CS Type, separately. After a 24-hr delay, participants showed an aversive learning-related memory enhancement for CS+ (red bars) compared to CS− items (gray bars) from the conditioning phase of the experiment (Phase 2). However, aversive learning did not retroactively bias memory in favor of conceptually-related items from the preconditioning phase (Phase 1). For plots a-d, colored boxplots represent 25th–75th percentiles of the data, the center line the median, and the error bars the s.e.m. Overlaid dots represent individual participants. (c) Across-participant variability in the aversive learning-related selective and retroactive memory effect (RME) for Phase 1 items. A subtraction score between corrected recognition rates for CS+ minus CS− items from Phase 1 is plotted on the y-axis. Values below zero represent a memory bias towards remembering items from the CS− category (gray box). Values above zero represent a memory bias towards remembering items from the CS+ category (red box), an effect that is termed “RME”. ~p = .050; *p < .05; ***p < .001.