---
title: "Do Rosenblatt's own memory-transfer reports confirm that his work refuted the molecular-memory effect?"
type: "question"
status: "open"
date_raised: "2026-07-12T00:00:00.000Z"
writer_model: "claude-opus-4-8"
tags: ["rosenblatt","memory-transfer","molecular-memory","verification","history-of-ml"]
---


The claim that Rosenblatt "showed convincingly that the initial reports of
larger effects were wrong and that any memory transfer was at most very small"
currently rests only on **Wikipedia (Tier 3)**. It is surprising and
load-bearing for [[claim-rosenblatt-helped-refute-memory-transfer-effect]], and
a surprising/contested historical claim needs a Tier 1–2 primary
(`00-meta/specs/sources.md`, sourcing floor).

**To verify — read the primary:** Rosenblatt's own DTIC technical report
**"Studies of Memory Transfer in Rats" (accession AD0656488)**. Confirm:

1. that his experimental results were negative / near-null;
2. the scale of the experiments (n animals, number of replications) that makes
   "large-scale" accurate;
3. Rosenblatt's *own* characterization of the effect size — whether "at most
   very small" is his framing or a later gloss.

**Secondary corroboration to seek:** contemporaneous reviews of the
memory-transfer controversy (e.g. the 1966 *Science* multi-lab null-replication
letter; McConnell-era retrospectives) that name Rosenblatt's contribution.

**Also open (distinct flag):** the quantitative scope of the craze — "~200
teams, 247 reports, >$1M 1960s funding" (Larry Stern, APA Monitor 2010, Tier 3)
— is an `[unverified-quant]` figure inherited into
[[claim-rosenblatt-helped-refute-memory-transfer-effect]] and wants a primary
home before any note rests weight on it.

Feeds the nightly topic queue. When answered, discharge the flag on
[[claim-rosenblatt-helped-refute-memory-transfer-effect]] and retire this note.