Mushrooms and other fungi may communicate electrically through what is known as the mycelium network. New research reveals differences between species, described as having their own languages, formed through trains of electricity spikes, with each "language" typically having about 50 trains, equivalent to words.

The neurons in our brains communicate through changes in electrical potential, “However,” the paper in Royal Society Open Science notes, “almost all creatures without nervous system produce spikes of electrical potential.”

For fungi, these can take the form of clusters of spikes known as trains, which humans can capture with detectors inside or outside their cells. The filaments, known as hyphae, which form the network can join fungi underground over vast distances, arguably making entire ecosystems into an Avatar-like superorganism.

Professor Andrew Adamatzky has previously reported that oyster fungi have spikes of two lengths – around 2.6 minutes and 14 minutes long, respectively. That's all one needs for a digital language, albeit one that would make Entish look hasty. As Adamatzky put it in his new paper, “this indicates a possibility that mycelium networks transform information via interaction of spikes and trains of spikes in manner homologous to neurons.”

Adamatzky has also previously shown that the trains change in response to being touched, exposed to light, or after experiencing chemical changes in their environment. Others have found potential communication occurs between fungi and plants, rather than being purely fungus to fungus. It is thought the signals could communicate the discovery of rich food resources or threats and may contribute to trading networks. 

For the new paper, Adamatzky tested four fungi species to see if they spoke the same "language". Each of those he tried turned out to be quite different – one had long intervals between spikes, another a richer array of “words” formed from varying combinations of high- and low-frequency pulses.

To begin the process of translating at least one fungus "language" into English, Adamatzky looked for recordings of trains that were so similar to each other that they could be considered to represent the same "word" being used multiple times. This is a major challenge. Even unfamiliar human languages can bamboozle us as we struggle to work out whether two similar sounds represent the same word pronounced marginally differently, or words with entirely different meanings. Distinguishing "words" in mushroom has to be much harder.

Fortunately, linguists have developed many word analysis tools. Applying some of these, Adamatzky was able to identify certain trains repeated often enough and in a similar fashion that they probably represent "words". He also reveals the distribution of spike train lengths matches word lengths in human languages. As with human languages, some tend to use longer "words" than others – with Cordyceps militaris packing an average of 8.9 spikes into a "word", while Omphalotus nidiformis uses just 3.3. By contrast, English has 4.8 letters to a word, and Russian has six.

Based on this, Adamatzky found some of the species studied have vocabularies of 50 "words", although none used more than 15-20 frequently. Of the four species studied, Schizophyllum commune, commonly known as “split gills” have the most complex "sentences", but with thousands of fungi still to be tested, it's unlikely Adamatzky hit on the most advanced species straight away.

Dr Dan Bebber of the University of Exeter remains skeptical, telling The Guardian: “Though interesting, the interpretation as language seems somewhat overenthusiastic, and would require far more research and testing of critical hypotheses before we see ‘Fungus’ on Google Translate.”

Correct or not, Adamatzky is already putting his discoveries with mushrooms to use. The University of the West of England, Bristol, where he is based, is constructing a new building with fungi sensors built in. The fungi will respond to changes in light, temperature, and pollution, with the building having the capacity to respond, keeping the interior better adapted to its human occupants.

“Acting as a massively-parallel computer, the building will control devices depending on the environmental conditions,” Adamatzky said in a statement when the building was announced. The use of biological sensors will save energy other smart buildings require to build, run, and recycle their detectors.