Syntactic Information Processing in Fungal Electrical Networks: A Computational Framework

This dataset provides complete raw data, analysis code, and computational models supporting the manuscript "Syntactic Information Processing in Fungal Electrical Networks: Evidence from Schizophyllum commune and Computational Modeling" (Chowdhury, 2025).

Research Question: Do fungal mycelial networks encode information using grammar-like syntactic rules analogous to quantum error correction, rather than simple Boolean logic or random processes?

Contents:

1. Raw Electrophysiological Data

• 137.2 hours (494,044 seconds) of continuous voltage recordings from Schizophyllum commune mycelium

• 5 differential electrode channels sampled at 1 Hz

• Voltage range: ±78 mV (24-bit ADC resolution)

• Files: paste.txt, fast_spike_window.csv

2. Processed Spike Data

• 395 detected spike events from high-density activity window (59 minutes)

• Spike detection via derivative thresholding (dV/dt < -0.15 mV/s, 30s minimum separation)

• Binary spike trains and inter-spike interval distributions

• Files: spike_analysis_enhanced.csv, comprehensive_event_catalog.csv, spike_analysis_summary.csv

3. Information Theory Analysis Results

• Shannon entropy at multiple time scales (5s, 10s, 30s, 60s windows): H = 4.84 bits (10s)

• Kolmogorov complexity: K = 0.098 (90.2% compressible)

• Redundancy: 51.6%

• Markov baseline comparisons: 20 synthetic sequences with t-test results (p=0.560 entropy, p=0.007 complexity)

• Files: week1-2_information_theory_analysis.csv, summary_statistics.csv, window_comparison_summary.csv

4. Cellular Automaton Simulations

• Three 50×50 grid models: Random, Boolean logic, Syntactic error correction

• Noise injection experiments (0-30% bit-flip rates, 5% increments)

• 105 total simulation runs (3 models × 7 noise levels × 5 trials)

• Mutual information preservation metrics demonstrating 419× capacity advantage for syntactic model

• Files included in analysis package

5. Complete Python Analysis Code

• Spike detection algorithms

• Entropy and complexity calculations (gzip-based K approximation)

• Markov chain simulation and statistical testing

• Cellular automaton implementations with noise injection

• Directory: COMPLETE_PYTHON_CODE_PACKAGE

6. Visualization and Graphs

• Voltage traces with detected spikes

• Entropy vs window size plots

• Mutual information degradation curves

• ISI distributions

• Files: Analysis.dgraph, Fast_spiking_analysis.dgraph, Action-potential-like-spikes.dgraph, Symmetric_Spikes.dgraph

7. Experimental Protocol and Predictions

• Complete wet-lab replication protocol for independent verification

• Four falsifiable predictions with cost estimates (£0-£2,000)

• Files: week5-6_experimental_protocol.txt, week5-6_falsifiable_predictions.csv

Key Findings:

• Fungal spike patterns show high structural redundancy (90.2% compressible) suggesting grammar-like encoding

• Temporal ordering is Markovian (p=0.560) but content exhibits super-compressibility (p=0.007), indicating syntactic structure beyond simple transition rules

• Computational models demonstrate syntactic error correction provides 419× higher information capacity than Boolean logic under noise conditions

Methods: Secondary analysis of Schizophyllum commune electrophysiology data (following Adamatzky 2023, Scientific Reports 13:12808) combined with information-theoretic quantification and cellular automaton modeling.

License: CC BY 4.0 (Creative Commons Attribution 4.0 International)