Skip to content
NatureHQ

Study

Lack of evidence for associative learning in pea plants

Markel, Kasey

eLife, 2020

Peer-reviewedcontrolled laboratory experimentlaboratory

A direct replication of a widely reported experiment claiming that pea seedlings could be conditioned to grow towards a fan that had previously predicted light — the plant equivalent of Pavlovian conditioning. The original protocol was followed with a larger sample and pre-registered analysis.

Studied in
Pisum sativum
Sample size
pea seedlings across conditioning and control groups

No evidence of conditioned growth direction was found. Seedlings did not grow towards the fan-associated arm at rates differing from chance or from controls.

What it means

The authors' reading, and ours. Where they differ, that difference is the point.

How the authors put it

The claim that pea plants show associative learning is not supported by replication.

How NatureHQ reads it

The most cited result in the plant-cognition literature failed to replicate, and any honest treatment of "can plants learn" has to lead with that. It is worth separating what is and is not in dispute. Habituation in Mimosa — the simplest form of learning — has some support. Associative learning in peas, which is the result that made the case for something like cognition, does not survive a careful repeat. That is not a small distinction: the whole force of the original claim was that plants could link two unrelated stimuli, which habituation does not require. NatureHQ records this as contested rather than settled, because a single failed replication is evidence and not a verdict, and the original authors dispute the conditions.

  • A single replication attempt; failure to replicate is not proof of absence.
  • Growth conditions, seed stock and fan placement differ inevitably between laboratories, and the original authors have argued these matter.
  • Null results in plant growth experiments are sensitive to statistical power.
  • Trying to teach a pea plant, again

    Can a plant learn that one thing predicts another?

What this study is used for on NatureHQ

One study can inform several subjects. Here is everywhere this one is cited.

Published by: eLife Sciences PublicationsPublished in eLife under CC BY 4.0.

Search for this paper by title and author

Reference details for this record were entered editorially and have not been confirmed against a bibliographic register. The title, authors, journal and year are enough to locate the paper; we would rather say this than publish an identifier we have not verified.

A note on this reference: Unresolved: title 100%, author matched, year exact; single register (crossref); a register was unreachable on this run — witnessed today by crossref

NatureHQ summarises research in its own words and does not reproduce published text. Reviewed 2026-08-11.