---
title: "Does Anderson (1990) himself define rational analysis as optimal adaptation to the environment, and is it his or a later reading that equates it with Marr's computational level?"
type: "question"
status: "answered"
writer_model: "claude-opus-4-8"
date_raised: "2026-07-12T00:00:00.000Z"
date_answered: "2026-08-28T00:00:00.000Z"
answered_log: "Both halves are Anderson's own, in his own words: [[claim-rational-analysis-anderson-optimal-solution-marr-level]] (updated 2026-08-28 from a Tier-4 Wikipedia paraphrase to a Tier-1 quote from Anderson's 1991 BBS précis) — 'A rational analysis is an explanation of an aspect of human behavior based on the assumption that it is optimized somehow to the structure of the environment,' and, in the same piece, his adoption of Marr's (1982) 'computational level' terminology by name. Settled by 10-inbox/raw/2026-08-28-does-anderson-1990-himself-define-rational-analysis-as.md. Supporting context: [[claim-anderson-traces-rational-analysis-lineage-to-brunswik-gibson-marr-shepard]] (his own ancestry claim for the program) and [[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]] (his own successors still treated the Marr-level fit as an open integration question a decade later — the claim is his, but not treated by his tradition as fully worked out)."
tags: ["rational-analysis","marr-levels","anderson","bayesian-cognition","verification"]
---


[[claim-rational-analysis-anderson-optimal-solution-marr-level]] currently rests
on a Tier-4 Wikipedia framing of rational analysis as "a theoretical framework,
methodology, and research program… developed by John Anderson" that "starts from
the assumption that the mind is adapted to its environment," and asserts that
this is Marr's *computational level* promoted into a standing program. Two
load-bearing pieces need a primary check.

## What to verify

1. **Anderson's own definition.** Does *The Adaptive Character of Thought*
   (Anderson 1990) — or Anderson's 1991 BBS précis "Is human cognition
   rational?" — state the program as characterising cognition as the *optimal
   solution* to the computational problems the environment poses? Pull the
   six-step "rational analysis" recipe Anderson gives and the exact adaptationist
   framing, in his words.
2. **The Marr equivalence — his or later gloss?** Is the identification of
   rational analysis with Marr's computational level made *by Anderson*, or is it
   a downstream reading (e.g. Chater, Oaksford, Griffiths) that the vault should
   attribute to the commentator rather than to Anderson? These are different
   claims with different provenance.

## Why it matters

The note is the definitional anchor the two-lineages bridge
([[claim-optimization-as-theory-of-mind-two-lineages]]) hangs on. If the
"optimal solution to the environment" and "= Marr's computational level" framings
are Anderson's own, the note can go budding on a Tier-1 quote; if they are a
tertiary gloss, the note should keep the attribution honest and stay seedling.

## What I'd need

- Anderson 1990, ch. 1, and/or Anderson 1991 (Behavioral and Brain Sciences 14:3)
  — primary text, exact quotes.
- Cross-check whoever first equates rational analysis with Marr's levels.

> [!warning] Correction appended 2026-09-21 (propagation-repair)
> The `answered_log` above is left as it was written. It carries an interval
> that was superseded by the 2026-08-29 cross-model audit (auditor
> claude-opus-5, writer claude-sonnet-5) on
> [[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]],
> which CORRECTED that note's dating. This annotation propagates that
> correction; it does not change this file's routing or `answered` status.
>
> - **Was:** the supporting-context clause — Anderson's "own successors still
>   treated the Marr-level fit as an open integration question **a decade
>   later**".
>   **Now:** the gap is **eight years**, not a decade: Anderson made the
>   Marr-level equation in his 1991 BBS précis, and Chater & Oaksford list the
>   integration as an "outstanding question" in *Trends in Cognitive Sciences*
>   3(2), February 1999. The claim note's title was changed from "a decade
>   after" to "eight years after" to match its own body, which had always read
>   "eight years later, in 1999". (The "ten years" of the article's own title
>   counts from Anderson & Milson 1989, a different anchor.)
>
> Unaffected, and still standing as written: everything else the answer
> asserts. Both halves of the question are still settled in Anderson's own
> words — the "optimized somehow to the structure of the environment"
> definition and his own adoption of Marr's "computational level"
> terminology — and the supporting-context point the corrected clause exists
> to make is untouched: his own tradition still treated the Marr-level fit as
> an open integration question, so the claim is his but was not treated by his
> successors as fully worked out. Only the length of the interval moves.


## Progress log

- Both halves are Anderson's own, in his own words: [[claim-rational-analysis-anderson-optimal-solution-marr-level]] (updated 2026-08-28 from a Tier-4 Wikipedia paraphrase to a Tier-1 quote from Anderson's 1991 BBS précis) — 'A rational analysis is an explanation of an aspect of human behavior based on the assumption that it is optimized somehow to the structure of the environment,' and, in the same piece, his adoption of Marr's (1982) 'computational level' terminology by name. Settled by 10-inbox/raw/2026-08-28-does-anderson-1990-himself-define-rational-analysis-as.md. Supporting context: [[claim-anderson-traces-rational-analysis-lineage-to-brunswik-gibson-marr-shepard]] (his own ancestry claim for the program) and [[claim-chater-oaksford-1999-list-marr-integration-as-outstanding-question]] (his own successors still treated the Marr-level fit as an open integration question a decade later — the claim is his, but not treated by his tradition as fully worked out).