Fungal machines: building computers and sensors from nature

Getting your Trinity Audio player ready...

• Fungal machines offer enormous potential for the future of computing.
• But first you have to get your head around what they do and how they do it.
• Fungal networks are the single largest organisms yet discovered on earth.

You might not believe that you can find free computers in the forest or pick them up in the fresh produce aisle of the supermarket for just a few dollars per kilo. But if you consider fungi to be capable of sensing and computing – and we’ll get into the arguments as to why you should – the very idea of information processing takes on a different form. Fungal machines, compared with laptops, smartphones, and other mass-produced electronics, represent computing’s unconventional side – enabling designs that are self-assembling, self-repairing, and biodegradable.

If fungal machines sound like an interesting prospect, then who better to learn from than pioneers in the field? And, for the first time, experts have teamed up to create a 425-page encyclopedia covering everything that is known so far about the ability of fungi to sense and compute the world around them. Andrew Adamatzky – editor of Fungal Machines: Sensing and Computing with Fungi and director of the Unconventional Computing Laboratory at UWE Bristol, UK – writes that fungi possess almost all the senses that humans do. “They sense light, chemicals, gases, gravity, and electric fields,” he explains.

Fungal Machines in five sections

Published by Springer, the book is divided into five main sections: fungal electricity, fungal sensors and wearables, fungal electronics, fungal computing, and fungal language and cognition. The first emphasizes that electricity is one of the key factors in shaping the growth and development of fungi – noting that the branching of mycelium (the thread of so-called hyphae below the stalk and fruiting body) is induced by electric fields.

Spiking electrical behavior, curiously reminiscent of neurons firing in the human brain, can be observed in fungal networks. And, pointing to the chemical sensing potential of fungi, experts have shown how the spiking frequency is dramatically reduced in the presence of chloroform – an anesthetic. When the vapor is eliminated from the test environment, electrical activity returns to its pre-anesthesia level.

Researchers are exploring the idea of embedding mycelium networks in wearables. Fungal skin has been shown to respond to loading – hinting at a potential use case as a touch sensor in robotics. These thin, flexible layers also have the ability to sense light. What’s more, as mentioned, fungal machines open the door to durable designs that can self-repair and grow, as well as being scalable and low-cost to produce.

Lab tests, described in the book, show that fungal wearables can withstand being kept in high-humidity conditions for months. And, signposting the ability for logical processing, computer models reveal that fungal colonies can implement a range of Boolean functions.

On TechHQ, we’ve written about how slime mold hit the headlines thanks to its ability to solve complex logistical problems – predicting the layout of metro systems and national road networks. However, fungal machines could prove to be even more versatile.

Adding an analog element to their digital capabilities, mycelium could offer an organic replacement for traditional capacitors, taking on the role of these electronic elements in signal filtering. The capacitance of living tissues can also be a proxy for ripeness – for example, in gauging the firmness of apples or the sugar content of citrus fruits.

FPUs – fungal processing units

In nature, it’s believed that fungi use electrical activity as a means of coordination. Mushroom populations, through networks of mycelium, can be huge. In the Pacific Northwest region of the US, fungal colonies can be found on the scale of several hectares, and fast-propagating electrical signals let these giant structures behave like a single giant organism – in fact, with apologies to the blue whale and the giant redwood, mycelium networks qualify as the single largest organisms yet discovered on the planet.

It’s possible that mycelium could provide a ready-made sensor network for monitoring forest health – for example, with electrical signals generated by fungal machines alerting us to environmental trauma.

In the book – Fungal Machines: Sensing and Computing with Fungi – the contributors discuss how mycelium responds to hydrocortisone, which is analogous to the hormone cortisol – a marker of stress in humans. And given how fungi are able to respond to changing concentrations of the chemical, organisms may offer insights into our own state of well-being.

Considering the future of computing – chemical AI and beyond.
Fascinating report@JT_bluebird1 https://t.co/QB6BMvAoIn

— Pier Luigi Gentili (@Pier_Complexity) October 9, 2023

Genetically, fungi are closer to animals than they are to plants, which could explain why they are more perceptive than many people would imagine. Some researchers argue that the complexity of fungi’s spiking communications rivals, or even exceeds, that of European spoken languages. And this may point to mycelium’s role as a messenger between plant, insect, and animal species.

Being able to decode this language could reveal a great deal and unconventional computing experts are busy trying to decipher these complex signals. Others too, are waking up to the idea of trying to listen to what fungi and plants are saying.

Data Garden – the North American firm behind Plant Wave – gives users the chance to hear microfluctuations in conductivity between two points on a plant expressed as human-audible sound. However, the company says that the tool isn’t designed to be used as a monitor of plant health. Instead, Plant Wave functions as a MIDI controller, triggering notes and interacting with harmonic algorithms.

Those wanting a scientific primer on getting started with FPUs should begin with Adamatzky’s breakthrough paper, Towards fungal computer, included within the Springer-published book’s section on fungal computing and also available online. Progress so far includes implementing sets of logical gates in single-colony mycelium networks through electrical impulses.

It should be said that computing with mycelium networks isn’t fast. Fungal machines won’t compete with silicon chips on speed, but they could turn out to be much more powerful in other ways. The book concludes by delving into the fascinating world of fungi consciousness and fungal minds.

The future of computing could turn out to be one where we care for our devices in a way that’s closer to looking after a houseplant than it is to plugging in and switching on a laptop. Certainly, unconventional computing concepts present a fresh take on information processing and the broad range of opportunities for fungal machines.

A stroll through the forest will feel different once you’ve read this book.