Findings
Last updated 29 September 2026 · superseded entries kept below
Everything on this page describes what was measured in my own recordings under stated conditions. None of it is a claim about what fungi do, and none of it is settled.
Each entry lists what supports it and, more importantly, what has not been excluded.
§1Standing1 entry
Present in the paper dataset and not yet explained by anything I have been able to test for.
F1Circadian-like 24-hour rhythm under constant darkness
- Observation
- Pleurotus ostreatus blocks were recorded under constant darkness (Dec 2025–Apr 2026). Coherent circadian-like 24-hour rhythms in within-window signal variability appeared in 23 of 30 inoculated channels (77%) versus 4 of 14 hardware-matched control channels (29%; Fisher's exact test, p = 0.006).
- Support
- Peak hours clustered at a consistent phase (~13.7 UTC) across inoculated channels (Rayleigh R = 0.83) but not in controls (R = 0.054). The phase was reproduced across non-overlapping ESP hardware sets (Watson–Williams F = 0.14, p = 0.71 between recording periods).
- Not excluded
- 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.
- Entry
- Log · circadian-like 24-hour rhythm (2025-12-11)
§2Retired2 entries
Each of these looked like a result for a while. They took the most time to rule out, and they are the most useful entries for anyone attempting similar work.
R1Endogenous 6–14 second oscillation retired 2026-03
- Was
- A short-period spectral feature reproducing across two acquisition rigs, and the strongest surviving candidate from a systematic search of the 20-second to 30-minute band.
- Failed on
- It tracked ambient temperature cycling and persisted in uninoculated controls at comparable amplitude.
- What it cost
- Roughly four months. Because it reproduced across rigs, a third rig would not have caught it. The uninoculated control did.
- Entry
- Log · the 6–14 second oscillation is thermal (2026-03-09)
R2Resting-state offset separation retired 2026-05
- Was
- A clean separation in resting potential between conditions, of a size that looked immediately usable.
- Failed on
- The separation tracked which amplifier recorded the condition rather than the substrate. Swapping the rigs moved the effect with the hardware.
- Consequence
- Condition and rig are now counterbalanced by default, and any absolute-level comparison across rigs is treated as uninterpretable.
- Entry
- Log · the offset separation was the rigs (2026-05-17)
§3How to read this page
- Standing means present in current recordings, with the alternatives I have been able to test for excluded. That is a much weaker statement than established.
- Retired entries are not removed. They stay with the date and the reason, because the record of what failed is the part that is hard to reconstruct later.
- Reproduction across rigs is not sufficient. R1 reproduced across two rigs and was still an artifact. The uninoculated substrate control is what caught it.
Different results are welcome; see About.