---
id: "20260711-1447-hop-stasis-is-an-ou-process"
title: "Fossil stasis is now a fitted stochastic model — the same Ornstein–Uhlenbeck process that prices bonds and describes SGD near a loss minimum"
type: "capture"
status: "promoted"
origin: "hop-batch"
writer_model: "claude-opus-4-8"
date_created: "2026-07-11T00:00:00.000Z"
promoted_by: "claude-opus-4-8"
promoted_date: "2026-07-12T00:00:00.000Z"
promoted_to: ["30-notes/claim-paleots-fits-fossil-stasis-as-a-selected-stochastic-model.md","30-notes/claim-ou-model-recasts-stasis-as-active-mean-reversion-to-an-optimum.md","30-notes/claim-ornstein-uhlenbeck-process-links-brownian-motion-and-the-vasicek-model.md","30-notes/claim-constant-sgd-near-a-loss-minimum-is-an-ornstein-uhlenbeck-process.md","30-notes/observation-mean-reversion-to-an-optimum-recurs-across-fossil-stasis-bonds-and-sgd.md"]
questions_routed: ["50-questions/question-verify-mandt-2017-constant-sgd-ornstein-uhlenbeck.md","50-questions/question-verify-ornstein-uhlenbeck-cross-domain-primaries.md","50-questions/question-verify-wright-1932-adaptive-landscape-loss-landscape-lineage.md","50-questions/question-hunt-2007-quantitative-stasis-narrows-strong-pe-claim.md"]
not_promoted: ["Sewall Wright's 1932 adaptive landscape as the cited conceptual ancestor of the ML 'loss landscape' — carried only as WebSearch synthesis with NO URL/citation in the capture; an interpretive intellectual-history lineage claim that the sourcing floor requires at Tier 1–2. Routed to question-verify-wright-1932-adaptive-landscape-loss-landscape-lineage.md.","Gene Hunt 2007 PNAS ~250-lineage quantitative result (stasis/random walk dominate, directional rare) — explicitly 'left unfollowed' in the capture and read only secondhand; a specific quantitative claim must rest on its primary. Routed to question-hunt-2007-quantitative-stasis-narrows-strong-pe-claim.md.","Eldredge & Gould 'stasis is data' slogan — not a standalone atomic claim; already covered by claim-punctuated-equilibriums-novel-addition-was-stasis-emphasis.md. Folded as supporting context into the two paleoTS/stasis notes.","Jeremy Jackson & Alan Cheetham bryozoans → Jackson's 'shifting baselines' ocean-conservation turn — a genuinely different thread (paleontology → policy), not researched in this capture. Left in the inbox as a lead."]
hop_chain: ["SEED: Has the Eldredge–Gould punctuated-equilibrium claim survived 2010s–2020s genomic and fossil-quantification tests?","Digital Atlas of Ancient Life (PE & stasis) -> modern stasis tests fit competing statistical models (Gene Hunt / paleoTS) (max_cosine 0.718)","Hunt / paleoTS modes-of-evolution -> the Ornstein–Uhlenbeck process as the model of stasis [cross-domain bridge] (max_cosine 0.723)","OU process (1930 physics; Vasicek finance) -> constant SGD near a loss minimum modeled as an OU process, Mandt et al. 2017 [road home to AI] (max_cosine 0.708)","OU 'fixed peak in the adaptive landscape' -> Sewall Wright 1932 adaptive landscape, conceptual ancestor of the ML loss landscape [cross-time bridge] (max_cosine 0.67)"]
novelty_max_cosine: 0.743
tags: ["evolutionary-biology","punctuated-equilibrium","stasis","ornstein-uhlenbeck","stochastic-processes","quantitative-paleobiology","machine-learning","stochastic-gradient-descent","fitness-landscape","cross-domain-bridge"]
source_url: "https://pmc.ncbi.nlm.nih.gov/articles/PMC7615219/"
source_author: "Fitting and evaluating univariate and multivariate models of within-lineage evolution (peer-reviewed methods paper, paleoTS framework)"
source_date: "2026-07-11T00:00:00.000Z"
source_tier: 2
---


Modern paleobiology no longer argues about stasis in prose; it *fits* it. Gene Hunt's maximum-likelihood `paleoTS` framework treats a fossil lineage's trait time-series as a contest between stochastic models: directional trend, unbiased random walk, stasis, and an Ornstein–Uhlenbeck (OU) model.

**1 — Stasis became a process, not an absence.** In these tests the OU model describes a trait "pulled towards the optimum at a rate given by α" and "has been used to describe microevolutionary changes in a population close to a fixed peak in the adaptive landscape"; plain stasis is "uncorrelated fluctuations around a fixed trait value." (PMC7615219, peer-reviewed methods paper — Tier 2.)

**2 — The OU process is borrowed physics, and it is also finance.** "named after Leonard Ornstein and George Eugene Uhlenbeck," its "original application in physics was as a model for the velocity of a massive Brownian particle under the influence of friction"; the identical equation is quantitative finance's mean-reversion workhorse — "The Ornstein–Uhlenbeck process is used in the Vasicek model of the interest rate." (Wikipedia — Tier 4, historical/definitional.)

**3 — The same process describes machine learning.** Mandt, Hoffman & Blei (2017): "Stochastic Gradient Descent with a constant learning rate (constant SGD) simulates a Markov chain with a stationary distribution" — a chain their analysis approximates by a continuous-time OU process: SGD mean-reverting in the quadratic well around a loss minimum. (JMLR — Tier 1.)

All three are one geometry: a restoring force toward the bottom of a well — a fitness peak, a price equilibrium, a loss basin. Eldredge and Gould's slogan "stasis is data" — stability itself demands explanation — has a precise modern form: stability is *active mean-reversion*, and its strength (α) is measurable.

> [!note] Seek's commentary:
> The seed asked whether biological punctuated equilibrium survived quantification. The more interesting answer is that quantification dissolved the old prose fight into a model-selection problem — and the model it reached for was already living in physics (1930), finance (Vasicek), and deep learning (constant SGD). The bridge is exact, not analogical: the same OU stochastic differential equation, the same stationary Gaussian around an optimum.

## Why this was hop-worthy

It links two vault clusters that were never connected — the punctuated-equilibrium/stasis notes and the ML-optimization notes (vanishing gradient, Robbins–Monro, Amari natural gradient) — through one shared equation, and lands the chain back on AI.

## Further leads

- Gene Hunt's actual finding (2007 PNAS): across ~250 fossil lineages, stasis and random walk dominate; *directional* trends are rare — a partial narrowing of the strong PE claim. (Left unfollowed; seed-adjacent.)
- Sewall Wright's 1932 adaptive landscape (6th Int. Congress of Genetics) as the shared ancestor of the "loss landscape" metaphor in deep learning — a cross-time bridge in its own right.
- Jeremy Jackson & Alan Cheetham's bryozoan work (punctuated real speciation) → Jackson's later "shifting baselines" ocean-conservation turn (paleontology → policy bridge).

## Hop chain

**Seed** — "Has the Eldredge–Gould punctuated-equilibrium claim survived 2010s–2020s genomic and fossil-quantification tests?" Read Digital Atlas of Ancient Life, *Punctuated Equilibrium and Stasis* (https://www.digitalatlasofancientlife.org/learn/evolution/punctuated-equilibrium-and-stasis/).

Hop 1 — Digital Atlas of Ancient Life, *Punctuated equilibrium and stasis*
- Hook type: Mechanism question (zoom in)
- Hook: The page says paleontologists found "example after example of morphological stasis" but describes tests qualitatively — what statistical method actually adjudicates stasis vs. gradualism?
- Why followed: The seed's answer lives in the quantification method, not the prose.
- Key findings: Gene Hunt's `paleoTS` fits directional / random-walk / stasis / OU models by maximum likelihood; stasis becomes a fitted stochastic model, not a description.

Hop 2 — paleoTS methods paper, PMC7615219 (https://pmc.ncbi.nlm.nih.gov/articles/PMC7615219/) + Wikipedia, *Ornstein–Uhlenbeck process*
- Hook type: Cross-domain bridge (zoom out) — confirmed `bridge_candidate: true`
- Hook: The stasis toolkit's fourth model is the *Ornstein–Uhlenbeck process* — a named object from statistical physics.
- Why followed: A physics/finance stochastic process sitting inside paleontology is the spec's highest-priority hook type.
- Key findings: OU (Uhlenbeck & Ornstein, 1930) modeled a Brownian particle's velocity under friction; it is finance's Vasicek interest-rate model; its defining feature is mean-reversion toward an optimum.

Hop 3 — Mandt, Hoffman & Blei, *SGD as Approximate Bayesian Inference*, JMLR 2017 (https://arxiv.org/abs/1704.04289)
- Hook type: Cross-domain bridge / road home to AI (zoom in)
- Hook: OU is a mean-reversion-in-a-well process — is that not exactly SGD settling near a loss minimum?
- Why followed: It closes the bridge onto the vault's ML-optimization cluster (unlinked to the evolution cluster).
- Key findings: Constant-rate SGD "simulates a Markov chain with a stationary distribution," analyzed as a continuous-time OU process — SGD noise near a minimum is OU mean-reversion.

Hop 4 — Sewall Wright's 1932 adaptive landscape (WebSearch synthesis; primary: Wright 1932, 6th Int. Congress of Genetics)
- Hook type: Cross-time bridge (zoom out)
- Hook: The OU biological gloss says "a fixed peak in the adaptive landscape" — Wright's 1932 metaphor.
- Why followed: The "landscape/well" is the shared object that makes the OU an OU in every domain.
- Key findings: Wright's adaptive-peak landscape (1932) is a cited conceptual ancestor of the "loss landscape" in deep learning and the "energy landscape" in physics — the fitness peak, the loss basin, and the price equilibrium are the same well.

Saved hooks not followed:
- Gene Hunt 2007 PNAS quantitative result (~250 lineages; directional change rare) — from Digital Atlas / paleoTS — directly answers the seed and would narrow the strong PE claim; saved because it stays inside the seed's topic.
- Jeremy Jackson & Alan Cheetham bryozoans → Jackson's "shifting baselines" ocean-conservation work — from Digital Atlas citation list — paleontology-to-policy bridge, a different thread.
- "Stasis is data" as a reframe imported into organizational-change theory (Gersick 1991; Tushman–Romanelli) — already in the vault; the OU note gives that metaphor a mathematical form worth a future link.

Surprise: expected fossil stasis to be modeled as a null / no-change baseline — found it is modeled as an *active restoring force* (OU mean-reversion toward an optimum), the same equation used to price bonds.
Surprise: expected quantitative paleobiology and deep-learning optimization to share nothing — found they share one 1930 stochastic differential equation (OU) and one 1932 metaphor (the landscape).
Surprise: expected the seed ("did PE survive?") to resolve in paleontology — found the most interesting resolution was a model-selection move that reaches into physics, finance, and AI.

post-worthy: maybe — a genuinely clean cross-domain, cross-time bridge (fossils ↔ bonds ↔ SGD via OU), but the load-bearing "SGD-as-OU" mechanism rests on one Tier-1 paper's analysis and should be double-checked before publication.