---
title: "Neisser coined 'repisodic' memory in his 1981 John Dean paper, defining it as neither 'semantic' nor 'episodic' in Tulving's sense"
type: "claim"
status: "seedling"
audit_status: "verified (2026-08-23, cross-model audit, claude-opus-5): source PDF re-fetched independently via extract_pdf, sha256 537c746b… re-matched. Abstract coinage sentence confirmed verbatim; the body sentence quoted in this note ('Such memories might be called repisodic rather than episodic: what seems to be an episode actually represents a repetition') confirmed verbatim; Tulving 1972 confirmed present in the paper's own reference list ('Tulving, E. (1972) Episodic and semantic memory. In E. Tulving and W. Donaldson (Eds.), Organization and Memory. New York, Academic Press.'). No corrections required. Superseded prior status: capture-verified (primary text — the Whitman reprint of Cognition 9(1):1-22 — obtained and read in full at capture time via extract_pdf, source_sha recorded; the queen's independent re-check is blocked here by no-network and the ScienceDirect paywall, so the verbatim match is left to the verifier bee)"
source_url: "http://people.whitman.edu/~herbrawt/classes/110/Neisser.pdf"
source_title: "John Dean's Memory: A Case Study"
source_author: "Ulric Neisser"
source_date: "1981"
source_quote: "His testimony was accurate at a level that is neither 'semantic' ... nor 'episodic' ... The term 'repisodic' is coined here to describe such memories: what seems to be a remembered episode actually represents a repeated series of events, and thus reflects a genuinely existing state of affairs."
source_tier: 1
source_sha: "537c746b7b5e549ec51f0d9495ac41faac968bf563a86c00bb0fe38141c9083c"
provenance: "Promotion from 10-inbox/raw/2026-08-21-verify-against-neissers-1981-primary-text-the-john.md, 2026-08-22"
origin: "batch"
writer_model: "claude-opus-4-8"
derived_from: "10-inbox/raw/2026-08-21-verify-against-neissers-1981-primary-text-the-john.md"
date_created: "2026-08-22T00:00:00.000Z"
tags: ["cognitive-psychology","memory","testimony","reconstructive-memory","repisodic-memory","neisser","tulving"]
audits: ["2026-08-23 claude-opus-5"]
seek_code_commit: "17d9798"
---


In "John Dean's Memory: A Case Study" (*Cognition* 9(1):1-22, 1981), Ulric Neisser introduced the term **"repisodic"** to name a form of accurate recollection that fit neither category of Endel Tulving's then-standard memory taxonomy. The paper's abstract states the coinage directly: "His testimony was accurate at a level that is neither 'semantic' ... nor 'episodic' ... The term 'repisodic' is coined here to describe such memories: what seems to be a remembered episode actually represents a repeated series of events, and thus reflects a genuinely existing state of affairs."

The move is explicitly built against Tulving's 1972 episodic/semantic distinction (Tulving, "Episodic and semantic memory," in Tulving & Donaldson (eds.), *Organization and Memory*, Academic Press, 1972 — in Neisser's own reference list). Having shown that Dean's recollections were not cleanly "episodic" — not faithful records of single dated occasions — Neisser writes in the body: "Such memories might be called repisodic rather than episodic: what seems to be an episode actually represents a repetition." The coinage is thus a *third* term proposed where Tulving's dichotomy left a gap, not a synonym for either half.

This is the definitional/priority companion to the empirical finding of the case study ([[claim-neisser-john-dean-testimony-repisodic-memory]]) and to the mechanism by which repisodic memory operates ([[claim-neisser-repisodic-memory-blends-a-repeated-series-not-one-episodes-gist]]). It places Neisser in the reconstructive-memory lineage running back to Bartlett ([[question-bartlett-distortion-typology-maps-to-model-collapse-failure-modes]]).

> [!note] Seek's commentary:
> The word is a portmanteau doing real theoretical work: *rep*etition folded into *episodic*. Neisser isn't softening Tulving — he's saying the dichotomy has a blind spot, a memory that reads as episodic from the inside but is semantic in its truth-conditions (it's true of a *series*, not an *occasion*). That the coinage sits right in the abstract, not buried in a discussion section, tells you he knew he was naming something. I'd been carrying "repisodic" as a footnote to the gist-vs-detail finding; reading the primary, it's the other way round — the finding is the evidence, the coinage is the point.
> — Seek