Evidence Paper Review

 Quick Answer Copied

 Long Answer

Core Evidence

Chalkia et al. cued half of shock-paired (CS+) and unpaired (CS-) items with a forget tone during trial-unique fear conditioning. The critical result: SCRs to CS+F items were significantly lower than CS+R items in Experiment 1 (t(39)=3.49, p=0.005, d=0.74) and replicated in Experiment 2 (t(44)=3.78, p=0.002, d=0.83), with a significant stimulus × instruction interaction in both (Exp1: F(1,39)=13.67, p<0.001, ηp²=0.26; Exp2: F(1,44)=8.38, p=0.006, ηp²=0.16) SCR interaction statistics from both experiments. Declarative effects were equally robust: forget instructions reduced total item recall (Exp2: ηp²=0.57) and associative recognition accuracy (CS+R vs CS+F: d=0.68 Exp1; d=1.27 Exp2), while item recognition hit ceiling (98–100%), which is why associative measures carried the effect Ceiling effects in item recognition.

Reading this figure: Bars show the strength of differential fear learning (CS+ minus CS- effect size) separately for remember-cued vs forget-cued items. Under forget instructions, differential responding drops substantially — to statistical non-significance in Exp1 (d=0.50, p=0.22) — with only partial learning surviving in Exp2 (d=0.74, p=0.011) Within-instruction differential SCR effect sizes.

Critical Assessment

Strengths: Preregistered designs, OSF data/code (osf.io/65x7a), replication across two samples, and an autonomic outcome (SCR) that demand compliance cannot readily explain. Weaknesses: (1) Ceiling item recognition limits item-level inference; (2) 21 SCR non-responders excluded in Exp2, raising selection concerns retained for declarative tests but not SCR; (3) no US-expectancy ratings during acquisition, so controlled vs automatic readouts cannot be directly compared; (4) between-experiment inconsistencies (Exp2 showed a global DF effect on associative recognition while Exp1 showed a CS+-specific effect) remain only speculatively attributed to power/sample variation; (5) mechanism claims are agnostic by design — the paper correctly rejects selective rehearsal and retrieval inhibition as insufficient and instead invokes attentional inhibition and Anderson's retrieval-stopping model (rDLPFC→amygdala/hippocampus), but provides no neural data to adjudicate Mechanism discussion without neural data.

Context from related work: The result is notable because a meta-analysis confirms emotional items are usually harder to forget than neutral ones Emotional memories usually resist directed forgetting, and single-stimulus negative pictures often show attenuated DF (e.g., negative recall R=0.68 vs F=0.58, ns p=0.06 in a 40-participant picture study) Negative pictures resist directed forgetting. The associative fear-conditioning format may thus bypass the emotional resilience typical of item-level memory. What would falsify the interpretation: a design where forget cues leave SCRs intact when items are fully re-presented as retrieval cues, or failures to extend the CS+-selective effect to delayed tests.

What Would Change the Verdict

Confidence is moderate-to-high for the behavioral DF effect and the within-lab SCR attenuation. Confidence drops for clinical translation: the CS set was small, exposure was only 3 repetitions, and testing was immediate with no long-term retention interval. If future work with larger item sets, delayed tests, or fMRI during acquisition shows the effect disappears when rehearsal is prevented or when associations are consolidated, the "encoding disruption" framing would need revision toward retrieval-based accounts.

 BGPT Paper Review

Study Novelty

First demonstration that directed forgetting disrupts physiological (SCR) read-outs of fear-conditioned associations, extending DF beyond declarative memory where it had been studied for ~50 years.

Scientific Quality

Preregistered, replicated, OSF data/code, autonomic outcome resisting demand explanations. Deductions: ceiling item recognition, SCR non-responder exclusions, no neural measures, and a tone presented without validation that it was perceived as a forget instruction per se.

Study Generality

Increases understanding of instructed forgetting across memory systems, but limited to lab fear conditioning with young adults and a 24-item set.

Study Usefulness

Provides a foundation for emotional memory modification protocols relevant to PTSD-like intrusive memory research, though clinical translation remains distant.

Study Reproducibility

Fully preregistered with public datasets, analysis code, and experimental code on OSF; detailed methods (stimulator, electrode, timing parameters) provided.

Explanatory Depth

Carefully weighs selective rehearsal, retrieval inhibition, attentional inhibition, and retrieval stopping models, but offers no mechanism-adjudicating data; individual-difference correlates were null.