Skip to content
NatureHQ

Claim check

Do fungi have a language of 50 words?

Exaggerated

Something true sits underneath, stretched well past what was shown.

Fungal mycelium really does produce electrical spikes, and that has been recorded since the 1990s. The vocabulary came from an analyst deciding how long a gap ends a word, and no fungus was ever shown to respond to one.

The claim as it circulates

“Fungi communicate using electrical signals that form a vocabulary of up to 50 words.”

Where you may have met it: International news coverage of a 2022 study, widely syndicated; Social media posts about fungal intelligence and the "wood wide web"; Popular books and talks about fungal networks

Taking it apart

What was claimed
That the electrical activity of fungal mycelium constitutes a language, with a lexicon of up to about 50 word-like units whose length distribution resembles that of human languages.
What was actually observed
Voltage was recorded between electrode pairs inserted into the mycelium or fruiting bodies of four fungal species. Spikes were identified in the traces and grouped into trains wherever the interval between consecutive spikes exceeded a chosen value. The lengths of those trains were then compared statistically with word lengths in several European languages.
What the evidence supports
That fungi produce spontaneous voltage fluctuations, that these arrive in clusters rather than evenly, that the pattern differs between species, and that it changes after events such as a wound, a flame or contact with a piece of wood. All of that is a real measurement and has been independently observed since the 1990s.
What it does not support
That any grouping carries meaning. The word-like units exist because a threshold was chosen; a different threshold produces a different vocabulary. A resemblance between two length distributions is a statistical similarity, not evidence of shared function. Most decisively, no receiver was tested — nothing in the work asks whether any fungus behaves differently for having received a train, which is the minimum a communication claim needs.

It is worth being clear about which part NatureHQ disputes, because it is a narrow part. The recordings are sound and the phenomenon is real and interesting. A mycelium may spread over square metres with no nerves and no brain, and it behaves as one organism: it thickens cords towards food, abandons unproductive regions and recovers their material, and fruits all at once across its whole extent. Electrical activity that propagates and responds to local events is exactly what coordination inside a single body might look like.

That is a different phenomenon from communication between individuals, and the difference is the whole question. The paper itself is more careful than its coverage was, and states that other explanations cannot be excluded. Specialists said so at the time. The experiment that would move this forward is not complicated: deliver a recorded or synthesised spike train to a mycelium, and see whether anything about its growth, its enzymes or its fruiting differs from a control that received a scrambled version. Until somebody does that, the word "word" is doing all the work.

What fungi demonstrably do coordinate

  • How fungi coordinate

    Chemical signals, hyphal fusion, and the electrical recordings in full

  • Fungi

    What a fungus is, and why a mushroom is only part of one

The claims underneath

Each one carries its own evidence, scope and caveats. Expand any of them to reach the studies.

Fungi have not been shown to have words. The spikes are real; the vocabulary is an interpretation.

Popular claim, unsupported

Widely repeated, with no good evidence behind it.

The identification of word-like units in fungal electrical recordings depends on an analyst-chosen inter-spike interval threshold for grouping spikes into trains, and on a statistical similarity between train-length and word-length distributions. No experiment has demonstrated that any fungus alters its behaviour in response to a received spike train, which is the minimum requirement for a communication claim.

Who this applies to
Claims that fungal electrical spiking constitutes a language or vocabulary, as reported for four species.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Basidiomycota, Ascomycota

You may have heard

“Fungi talk to each other using a vocabulary of about 50 words.”

The fifty comes from cutting a voltage recording into groups wherever the gaps exceed a threshold the analyst picked, then noting that the resulting group lengths resemble word lengths in English. Nothing in the work asks whether a fungus responds to a group, and until something does, the word "word" is doing all the work.

Why we rate it this way, and what the caveats are
Popular claim, unsupportedModerate confidence

The underlying recordings are sound and NatureHQ treats them as such. What is unsupported is the step to language, which rests on an analytical choice and an analogy rather than on a test — and the test that matters, whether a receiver responds, has not been run.

How far it can be extended

The analysis covers four species under laboratory conditions and tests no receiver.

Caveats

  • This is not a claim that fungi lack internal signalling. They plainly coordinate growth across a mycelium, by chemical and probably electrical means.
  • The original paper is more cautious than the coverage it generated.

Still unanswered

  • Whether a receiver experiment — delivering a recorded train to a fungus and measuring its response — would find anything.
  • What function, if any, the spike trains serve within the mycelium.

Last reviewed 2026-08-30

The evidence (2 studies)

Fungal mycelium produces electrical spikes, and the pattern changes when the fungus is disturbed.

Well supported

Good evidence backs this, though some details remain open.

Electrode recordings from fungal mycelium and fruiting bodies show spontaneous voltage fluctuations resembling action potentials, arriving in irregular trains, whose rate and pattern alter following localised stimulation such as wounding, heat or contact with a food source.

Who this applies to
A small number of cultured species recorded under laboratory conditions with implanted electrodes.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Basidiomycota, Ascomycota
Why we rate it this way, and what the caveats are
Well supportedModerate confidence

The phenomenon has been recorded independently since the 1990s and is not in doubt. Its scope is narrow — few species, laboratory conditions — and separating genuine biological activity from electrode artefact over long recordings is a persistent difficulty.

How far it can be extended

Recordings exist for only a handful of species, and electrode implantation is itself a wound, which is one of the stimuli that changes the signal.

Caveats

  • Activity within one mycelium is coordination within a single body, which is a different phenomenon from communication between individuals.
  • What the spikes do, if anything, is not established by their existence.

Still unanswered

  • Whether the spikes carry information, and if so what kind.
  • Whether any fungal behaviour changes in response to a spike train arriving from elsewhere.

Last reviewed 2026-08-30

The evidence (2 studies)

Go deeper

How fungi coordinate

Fungal mycelium really does produce electrical spikes. The vocabulary of fifty words came from choosing where to cut the recording.

Related things people ask

Last reviewed 2026-08-30