Neural Noise

Encoding-decoding model of neural transmission and neuronal communication

2 min read

  • New insights into an encoding-decoding-based neuronal communication model have been reported for neural transmission in the wired brain.1
  • The work provides conceptual relevance to neural decoding and computational neuroscience; Nature, more theoretical, tier-2.1 1

Gardner updates

  • Neural transmission in the wired brain is modeled via an encoding-decoding-based neuronal communication framework; conceptually relevant to neural decoding and computational neuroscience. 1

Weekly enrichment (2026-07-20)

  • The source paper (Translational Psychiatry, 2025) analyzed resting-state EEG from five public datasets (TDBRAIN, SRM, MODMA, REO, FEP) totaling 1,668 participants aged 5–89, spanning healthy controls and people with MDD, ADHD, OCD, schizophrenia, and Parkinson’s.2
  • Methods: two minutes of eyes-open/closed EEG were average-referenced and filtered (0.1 Hz high-pass, 50/60 Hz notch), decomposed with the Stockwell transform (0.002 s time and 0.3 Hz frequency resolution), and inter-hemispheric alpha (8–12 Hz) envelope coupling was assessed by sliding-window Spearman correlation (10-point window, ~100 correlation values per second).2
  • The central finding is a “Beating” pattern: the frontal hemispheres’ alpha envelopes alternate between full synchronization and desynchronization several times per second, attributed to interference between two signals of slightly different frequencies.2
  • The pattern held across every ipsilateral/contralateral electrode pair, across frequency bands, in both eyes-open and eyes-closed states, and across all ages, which the authors interpret as evidence of an inherent communication mechanism.2
  • They propose an encoding-decoding communication model in which frequency modulation encodes a digital-like binary code (0s and 1s) transmitted in packets carrying data and metadata under biological rules shared by sending and receiving regions, explicitly analogized to wired electronic transmission.2
  • As a candidate biomarker, synchronization was significantly lower and desynchronization higher in people over 50 versus younger participants, and ADHD patients showed lower desynchronization than age-matched controls.2
  • The authors frame this digital-like scheme as directly exploitable for brain-computer interaction, robotics, and neuroprosthetics, connecting the basic-science model to applied neural decoding.2
  • This complements the broader view (perspectives in Cell and related reviews) that encoding and decoding are “two sides of the same coin” — downstream neurons decode and re-encode upstream information, while engineered decoders read out neural representations for BCIs.34
  • Scaling the idea computationally, recent foundation-model work (Neural Encoding and Decoding at Scale, NEDS) jointly learns encoding and decoding across 83 animals in the International Brain Laboratory dataset, reaching state-of-the-art bidirectional translation between neural activity and behavior.5

Footnotes

  1. https://news.google.com/rss/articles/CBMiX0FVX3lxTFBEcDdvcEQ3RlhNbm1HendZa0dpRHZ3eGZHWVJKQ0ZxVGcyUksxdXp5a1M3bFFLUkVUUmZvemNQd0UyQjlWVDY2MVcxUFBwaE11THZObThmSWpPTDJpbmtn?oc=5 ↩ ↩2 ↩3 ↩4

  2. https://doi.org/10.1038/s41398-025-03506-0 ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7

  3. https://pmc.ncbi.nlm.nih.gov/articles/PMC11637322/ ↩

  4. https://www.sciencedirect.com/science/article/pii/S0092867424009802 ↩

  5. https://proceedings.mlr.press/v267/zhang25bw.html ↩

Graph View