---
title: "Was Amari's 1972 associative-memory model ever independently applied to developmental biology before the 2016 Taherian Fard Hopfield-landscape paper?"
type: "question"
status: "open"
progress_log: ["2026-07-14: Candidate [[claim-no-amari-application-to-developmental-biology-found-pre-2016]] promotes the bounded negative: four reference lists (Taherian Fard 2016, Lang 2012/2014, Yampolskaya & Mehta 2025, Maetschke & Ragan 2014) carry no Amari citation. Held as absence-of-evidence, not proof of nonexistence; obscure/uncited/non-English leads (Huang 2005, Wang 2011) keep it open."]
writer_model: "claude-opus-4-8"
date_raised: "2026-07-11T00:00:00.000Z"
tags: ["amari","hopfield-networks","developmental-biology","waddington","priority","history-of-ml"]
---


The vault holds that Amari's 1972 associative-memory model has substantive
priority over the Hopfield network and is uncited by Hopfield's 1982 paper
([[claim-amari-1972-associative-memory-precedes-hopfield]]). The 2016
Taherian Fard et al. paper that formalizes Waddington's epigenetic landscape
as a Hopfield energy surface
([[claim-hopfield-network-formalizes-waddington-epigenetic-landscape]]) cites
Hopfield 1982, not Amari.

**The question:** did anyone apply Amari's 1972 architecture — or Amari's own
later dynamical-systems / neural-field work — to cell differentiation or
developmental biology *before* 2016? If so, the Amari-priority thread would
extend into a second field, and the "Hopfield attractor" framing in systems
biology would inherit the same uncited-prior structure the vault already
documents in machine learning.

**What would answer it:** a citation search for pre-2016 developmental-biology
or gene-regulatory-network papers referencing Amari 1972 (or Amari's neural-field
work) as the associative-memory / attractor formalism; a check of whether
Taherian Fard et al.'s reference list mentions Amari at all. The capture that
raised this ("not checked") is
10-inbox/raw/2026-07-09-hop-waddington-hopfield-landscape.md.

**Update, 2026-07-12 (search done, negative — question stays open):** a
follow-up capture ran a moderately thorough citation-graph search across four
primary/near-primary sources — Taherian Fard et al. 2016 itself, Lang et al.
2012/2014 (58-entry reference list read in full), Yampolskaya & Mehta 2025
(69-entry review reference list read in full), and Maetschke & Ragan 2014 —
and found **no Amari citation in any of them**. The Taherian Fard reference list
contains no Amari entry. Result is promoted at
[[claim-no-amari-application-to-developmental-biology-found-pre-2016]] and held
`[unverified]` as absence-of-evidence, not proof of nonexistence. The question
stays `open` because the search cannot rule out an obscure, uncited, or
non-English-language application.

**Remaining leads not yet checked (would tighten the negative):**
- Huang, Eichler, Bar-Yam & Ingber (2005, *Phys. Rev. Lett.* 94:128701) — the
  earliest attractor-network cell-fate paper in this cluster; its own reference
  list was not independently checked.
- Wang, Zhang, Xu & Wang (2011, *PNAS* 108:8257–8262) — reference list not
  independently checked.
- Amari's *1977* neural-field work is a structurally distinct thread, tracked
  separately at [[question-amari-1977-neural-field-applied-to-morphogenesis]].