Retrieval-dependent vulnerability of conditioned responding

Citation: Karim Nader, Glenn E. Schafe, and Joseph E. Le Doux. Fear memories require protein synthesis in the amygdala for reconsolidation after retrieval. Nature 406, 722–726, 17 August 2000. Peer-reviewed Letter; received 3 April, accepted 18 May 2000. Publisher, DOI: 10.1038/35021052; PubMed.

Access and reading scope: Read the complete target article's main text, Methods, all five figures and legends, and references in a public third-party copy of the publisher-layout PDF. Visually inspected all five pages, including drug units, timelines, and the inconsistent degrees of freedom. The first and last pages also contain neighboring articles; those are not included in this reading claim. Title, authors, pagination, dates, abstract, and figure identities match the publisher record. The publisher PDF endpoint returned subscription-preview HTML; the LeDoux institutional list links to that publisher, and the legacy Nader laboratory link is erroneous. No author/repository full text was located. Public access to the retained copy was unobstructed, but authorization for third-party posting and redistribution was not independently established; the PDF bears © 2000 Macmillan Magazines Ltd. No supplement was linked by the inspected publisher page or identified in the article. Provenance records four retained artifacts, hashes, access routes, and these limits. No raw data or later reconsolidation experiments were analyzed.

Evidence, concise paraphrase: Adult male rats learned one tone–shock association. A tone-only reminder one day later, followed immediately by high-dose anisomycin in lateral/basal amygdala, reduced freezing at the next day's test. Lower-dose treatment and vehicle did not show that reduction. Omitting the reminder prevented the reported impairment; delaying infusion six hours produced a smaller, nonsignificant difference. Immediate treatment also reduced subsequent freezing when the reminder occurred fourteen days after training. Another experiment found comparable responding four hours after treatment but reduced responding at twenty-four hours. These temporal and retrieval controls support renewed vulnerability of an established conditioned response, consistent with reconsolidation. The measured outcome was freezing, not decoded event content or a directly observed erased trace. This article did not test later spontaneous recovery, reinstatement, or renewed relearning after its reconsolidation intervention. Its discussion acknowledged that susceptibility might depend on experience, memory system, and motivational conditions rather than accompany every retrieval.

Methods and reporting audit:

Training paired a thirty-second tone with a coterminating one-second foot shock. Reminder and testing occurred in a different chamber from conditioning, reducing contextual generalization. The target was lateral amygdala, with expected spread into basal amygdala. High and low doses were 62.5 and 6.2 μg per 0.5 μl per side, respectively. The high-dose protein-synthesis suppression rationale came from a prior cortical study, not a new molecular measurement in these animals. The published plots show mean freezing and standard errors; this audit did not digitize them.

Methods' group allocations sum to 76, but a total enrollment/exclusion flow is not reported. Randomization procedures, blinded scoring, and explicit exclusion rules/counts are not described. Histology checked cannula placement and reported no amygdala damage. Experiment 4 contains an unresolved discrepancy: its Methods specify sixteen animals, but the reminder's period effect is printed as F(1,10)=147; the later long-term group effect uses F(1,14)=14, consistent with sixteen. The PDF visually confirms both denominators. Do not silently infer exclusions, repair the denominator, or treat the initial comparison as a verified sixteen-animal analysis.

Critical interpretation:

• The strongest causal claim is about a retrieval- and timing-dependent effect of the intervention on later conditioned responding. Reminder freezing occurred before infusion, and the four-hour test demonstrates preserved ability to perceive and respond to the tone at that time. Those controls challenge immediate global incapacity and some simple toxicity accounts. They do not, by themselves, exclude every delayed change in expression/access or uniquely identify loss of stored information.
• The six-hour group difference was nonsignificant, with F(1,13)=4; the authors themselves describe a smaller nonsignificant impairment. That is not an equivalence result or proof of a sharp six-hour cutoff. Immediate versus delayed infusion also differs in drug-to-test interval. Separate significant and nonsignificant comparisons are not a reported drug-by-delay interaction. Likewise, the reminder/no-reminder pattern is informative without substituting for an explicitly reported factorial interaction.
• Experiment 4's early test is another retrieval exposure. Its twenty-four-hour result therefore follows an additional three-tone test, not an unobserved retention interval. Experiment 1A demonstrates impairment without that extra early test, but the short/long comparison should retain its intervention history.
• Persistence and relearning are limited here. The article cites normal relearning after the same drug dose was given following initial learning. It does not present a new post-reconsolidation relearning curve, savings analysis, spontaneous-recovery interval, reinstating shock, or context-renewal test. An assertion of permanent erasure would exceed these data. Later papers addressing those alternatives require separate reading and cannot be imported into this study's results.
• A tone without shock provides both a retrieval cue and a changed contingency. The authors argue that simply blocking acquisition of extinction should increase, rather than decrease, subsequent freezing. This challenges that simple extinction-blockade explanation, but does not isolate retrieval from all mismatch/new-learning processes or directly test every facilitated-extinction/access account. The paper's structural and synaptic-tagging discussion is a proposed explanation, not a measurement of the affected trace.

Original extraction-as-intervention implication: A procedure intended to read a memory can also be an event affecting its future accessibility or expression. This study supplies a controlled animal example in which the consequences of a drug depend on a preceding reminder; it does not show that ordinary interviewing erases memories or that every human recall opens the same molecular window. For extraction research, retain cue, report, timing, and later-experience histories. Compare paired recipients receiving or not receiving a measurement after the same acquisition history to estimate information released; separately compare acquisition policies to estimate their total effects on the source. Track original-event accuracy, the source's later report, and later behavior separately. A changed report after elicitation cannot automatically be labeled improved readout, destroyed storage, or faithful transfer to a recipient.

Proposed discrimination, not a result: In an eventual authorized human extension, use prospectively recorded benign episodes and separate reminder/no-reminder or probe/no-probe schedules, with delayed source and behavioral tests. Establish which details were retained without feeding verification answers to the recipient. Any claim that acquisition changes the source needs its own comparison; any molecular reconsolidation claim needs evidence beyond changed recall scores. No pharmacological intervention or participant study is proposed for execution by this reading task.