SCR responses to conditioned stimuli in two groups of study 1 indicating short-term memory amnesia (A) Experimental design and timeline. (B) SCRs of fear conditioned stimuli (CS+) and the control stimulus (CS−) across fear acquisition, extinction and test phases for two groups (reminder and no-reminder). (C) Mean differential SCRs (CS+ minus CS−) in the acquisition phase (trials in the latter half). (D) Differential SCRs (CS+ minus CS−) in the extinction phase (last trial). (E) Differential fear recovery index between CS+ and CS− in the test phase. ***P < 0.001, *P < 0.05. NS: Non-significant. Error bars represent standard errors.

SCR responses to conditioned stimuli in three groups of study 2 indicating amnesia at different timescales. (A) Experimental design and timeline. (B) SCRs of fear conditioned stimuli CS1+ (reminder) and CS2+ (No-reminder), and the control stimulus (CS−) across the fear acquisition, extinction and test phases for each group (30min, 6h and 24h groups). (C) Mean differential SCRs (CS1+ minus CS− and CS2+ minus CS−) in the acquisition phase (trials in the latter half). (D) Differential SCRs in the extinction phase (last trial). (E) Differential fear recovery index between CS+ and CS− in the test phase. ***P < 0.001. **P < 0.01. *P < 0.05. NS: Non-significant. Error bars represent standard errors.

Fear recovery as a function of thought-control abilities. (A)Thought control ability was significantly correlated with fear recovery index in the 30min group (P = 0.003, Bonferroni correction), but not in the 6h (B) or 24h group (C, Ps > 0.7). The violin graphs indicate the distribution of fear recovery index across subjects (Figure 2E). **P < 0.01. NS: Non-significant.

SCR responses to conditioned stimuli in the cTBS study (study 3) indicating that both intact dlPFC and cue reminder are required for short-term memory amnesia. (A) Experimental design and timeline. (B) SCRs of fear conditioned stimuli CS1+ (reminder) and CS2+ (No-reminder), and the control stimulus (CS−) across the fear acquisition, extinction and test phases for each group (R-PFC, R-VER, NR-PFC and NR-VER groups). (C) Mean differential SCRs (CS1+ minus CS− and CS2+ minus CS−) in the acquisition phase (trials in the latter half). (D) Differential SCRs in the extinction phase (last trial). (E) Differential fear recovery index between CS+ and CS− in the test phase. ***P < 0.001. *P < 0.05. NS: Non-significant. Error bars represent standard errors.

The correlation between differential fear recovery index and thought-control ability in four cTBS groups. (A) There was no significant correlation between differential fear recovery index and thought-control abilities in the R-PFC group (P > 0.4). (B) However, in the R-VER group, the correlation between thought control ability scores and the differential fear recovery index was significant (P = 0.008, Bonferroni correction), with high thought-control ability participants showing less fear recovery for both CSs+. Such correlation was not observed in the (C) NR-PFC or the (D) NR-VER group (Ps >0.4). ** P < 0.01, NS: Non-significant.

Time courses of short- and long-term amnesia. (A) At the short interval (30min to 1h), fear recovery of the reminded CS (CS1+) is inhibited (green). As time progresses (from 6h to 24h), amnesia effect is likely due to the fear memory reconsolidation effect emerged later (orange). Actual SCR data in black solid line. (B) Fear amnesia of the non-reminded CS (CS2+) is only evident at the 30-min interval and such effect starts to decay as test interval increases (green). However, long-term amnesia does not generalize to CS2+ (orange, cue-dependence). The observed SCR data (black solid line) paralleled the prediction from both the short-term and the long-term effects as the interval length increased (from 30min to 24h). (C) Schematics depicting the effect of cue-reminder on fear memory retention. After both CS1+ (black) and CS2+ (grey) successfully elicit fear responses in the acquisition phase, CS1+ is reminded (black) before both CS+ go through the extinction training. The lack of both CS1+ and CS2+ fear responses in the short-term memory (STM) test (30 min) might be explained by the dlPFC dependent direct suppression effect (dotted circle of US representation). On the other hand, the cue-specific fear amnesia effect in the long-term memory (LTM) test (24h) of CS1+ but not CS2+ could be attributed to the reconsolidation effect specific to CS1+. * P < 0.05, Error bars represent standard errors.