Laboratories have been recording electrical activity in fungal networks since the 1970s. The phenomenon itself is not in dispute. What those signals correspond to is largely open, including whether they carry anything about the state of the organism and whether that could be read from outside it.
EarthTalk is my attempt to work out, in my own recordings, how much is signal and how much is instrumentation. I measure electrical activity directly from living mycelium using ordinary hardware, describe what I observe, and publish the methods so that anyone can check.
Most of what is here is negative. Several promising leads from the first year turned out to be temperature, electrodes, or amplifier differences rather than biology. Those are recorded in full, because they cost the most to find and are the more useful thing to hand to anyone attempting similar work.
Writing up circadian-like 24-hour rhythms recorded from Pleurotus ostreatus blocks under constant darkness (paper data, Dec 2025–Apr 2026). Longer term, working toward non-invasive, in-field detection of mycorrhizal colonization from electrical recordings. The method cannot do this yet.
What currently holds, what has been retired, and why. Start here to see where the project stands.
Dated entries. Each states the question, what was uncertain, what was done, what came back, and what it changed.
Step-by-step setup, equipment, and code, versioned. Superseded protocols are kept and marked rather than removed.
Descriptions of what was measured under stated conditions. None of these are claims about what fungi do. Full findings page →
| Circadian-like 24-hour rhythm under constant darknessCoherent rhythms in 23/30 inoculated channels vs 4/14 hardware-matched controls (p = 0.006). Room temperature and humidity also cycle at 24 h and controls were recorded later, so an environmental contribution is not excluded. What distinguished inoculated channels was a consistent peak phase that same-hardware controls did not show. |
| 2026-03 | Endogenous 6–14 s oscillationTracked ambient temperature cycling and persisted in uninoculated controls. Thermal and electrode artifact. |
| 2026-05 | Resting-state offset separationSeparation tracked which amplifier recorded the condition. Swapping rigs moved the effect with the hardware. |
| 2026-05 | The resting-offset separation came from the rigs | retired |
| 2026-03 | The 6–14 second oscillation is thermal | retired |
| 2026-02 | Where the noise floor comes from | method |
The hardware is deliberately ordinary, and anything described here can be reproduced without specialist equipment.
These apply to everything on this site rather than to any single entry.
If you run fungal blocks, build instrumentation, or analyse signals of this kind, I would like to hear from you, particularly if you tried something described here and it did not work. Negative results are welcome and will be published with attribution.
Moshe Lokshin, an entrepreneur based in Canada. I have built and sold companies in media, software, and real estate. I now work on this full time.
The longer question I am working toward is whether electrical measurement can say something useful about the biological state of soil, including non-invasive detection of mycorrhizal colonization in the field. That is a hypothesis rather than a plan, and most of this site is the record of finding out where it fails.