Abhishek S.
Shipping in public. Listening in private.

Abhishek

I lead women’s Indo-Western & Premium at Max Fashion. I also wrote the AI that runs the buying floor.

Rare profile. Category operator who ships production code.

Senior Buying Leader · Max Fashion Women’s Indo-Western & Premium · 530+ India stores NIFT ’12 · Twelve years on the floor

read the CV start a letter
abhishek@bengaluru ~ %
>role: senior buying lead
>dept: women’s indo-western + premium
>floor: 530+ stores india

Neural Turbulence — The Missing Reynolds Number and the Free Will Horizon

Two papers published in 2025 make the same startling claim: the brain is a turbulent fluid system, and its cognitive states can be described by the same mathematics as a stirred pot of water. One paper derives a kinetic equation for cortical wave turbulence. The other builds a whole-brain model of turbulent vortices that distinguishes cognitive tasks. Neither computes the one number that would tie the physics to free will: an effective Reynolds number for human cortex.

The 2025 Framework

The turbulent brain has been proposed before — Deco and Kringelbach showed Kolmogorov cascade exponents in fMRI BOLD signals (see concept brain turbulence) — but 2025 moved from analogy to formal derivation.

Wave Turbulence and Cortical Dynamics (Cooray et al., 2025)

Published as arXiv 2507.23525 (July 2025), then in Frontiers in Computational Neuroscience (March 2026), this paper applies weak wave turbulence theory to neural field equations. Neural fields model cortex as a continuous medium with excitatory/inhibitory coupling, wave propagation speed (approximately 5–10 m/s across the cortical surface), and non-linear interaction terms.

The key result: the kinetic equation for neural field activity produces direct and inverse cascades — energy flows upward toward large-scale structures (like cortical rhythms) and downward toward fine spatial detail, just as in Kolmogorov's turbulence. Three-wave and four-wave interactions shape spectral features observed in real EEG and MEG recordings: the characteristic power-law decay (approximately -5/3 slope at some scales), harmonic structure, and dual-cascade behavior.

The implication is stark: EEG/MEG power laws are not just empirical curiosities to be fit with exponentials. They are signatures of a fundamentally turbulent dynamical regime, governed by the same irreducible physics that makes long-range weather forecasting impossible.

The Turbulent Brain: Vortex Interactions (Deco et al., bioRxiv, September 2025)

The Deco group's September 2025 preprint takes a different approach: instead of derivation from neural field theory, it builds an empirical vortex model from neuroimaging data.

The method uses Kuramoto oscillators — each brain region as a phase oscillator — to compute Kuramoto vorticity: the local synchronization field around each cortical point. This is mathematically the curl of the phase field, a direct analog of vorticity in fluid mechanics (ω = ∇ × v). Where multiple oscillators lose their local phase coherence in a coordinated way, a vortex forms.

The model finds that:

The Missing Number

A Reynolds number (Re = ρvL/μ) encodes whether a fluid flows laminarly (Re << 1,000) or turbulently (Re > ~4,000 for pipe flow). In a neural field system, the analogous ratio is:

Re_neural ~ (propagation speed × cortical length scale) / (synaptic decay timescale × coupling decay)

Inserting rough empirical values:

A back-of-envelope computation gives Re_neural in the range of 10–100 — above the laminar regime but in the transitional zone where turbulence becomes self-sustaining. Crucially, no published paper has done this calculation rigorously with empirical neural parameters substituted into the formal Reynolds-analog definition from a neural field equation.

This gap matters because:

  1. If Re_neural >> turbulence threshold → the cortex is formally turbulent → spontaneous stochasticity applies
  2. If Re_neural < threshold → the power-law signatures are approximate, not indicative of true turbulence → no SS horizon

The Cooray paper is the closest to providing the formal machinery. The neural field kinetic equation contains terms that behave like viscosity (the synaptic decay/spreading function) and inertia (the wave propagation). A formal Re_neural derived from this equation, and verified against EEG power spectra, would be the missing empirical anchor.

Why Reynolds Number → Free Will

concept spontaneous stochasticity free will argues that if brain dynamics are turbulent at sufficient Reynolds number, a neural SS horizon exists: a timescale below which individual decision trajectories are irreducibly random — not merely hard to predict, but physically indeterminate at the single-trajectory level, even with perfect initial conditions.

Spontaneous stochasticity (SS) is distinct from Lorenz chaos (sensitive dependence on initial conditions). In a chaotic system, you could in principle predict the trajectory given perfect measurements. In a spontaneously stochastic system, thermal noise amplifies to macroscopic scale within the eddy turnover time — and this amplification is irreducible, a property of the mathematics at infinite precision, not a measurement limitation.

The Libet connection: Libet's readiness potential (RP) appears 300–500 ms before a voluntary action and has been interpreted as evidence that the brain "decides" before conscious awareness. But the "free won't" window — the veto window during which conscious experience can intercept the RP — is approximately 200 ms.

If the neural eddy turnover time is shorter than 200 ms, then SS makes individual decision trajectories irreducibly unpredictable within the Libet window. The RP is a statistical attractor (ensemble prediction remains valid), but the specific moment and manner of any individual action cannot be predicted — not by the brain's own internal dynamics, not by any external observer, and not by any deterministic prior state.

Claim Source Status
Cortex shows turbulent-like power laws Beggs & Plenz 2003; Deco & Kringelbach 2020 Established
Neural field equations admit wave turbulence Cooray et al. 2025 Emerging
Turbulent vortex maps track cognition Deco et al. 2025 Emerging
Effective Re_neural formally computed — Open gap
Neural eddy turnover time measured — Open gap
SS horizon ≤ 200 ms (Libet window) — Theoretical

The Vortex Model's Surprise

One unexpected finding from the Deco 2025 vortex model: cognitive tasks do not primarily differ in which brain regions activate (as region-of-interest fMRI analysis implies), but in how vortices interact. The same regions show activity across tasks; what changes is whether a vortex at location X is merging with, splitting from, or orbiting around vortices elsewhere. This is a topological description of cognition, not a localizationist one.

This result connects to concept distributed cognition and to concept emergence: the cognitive state is not stored at any particular location but in the dynamical relationship pattern of the turbulent field. It also suggests why the "grandmother neuron" — a single neuron encoding a specific concept — has proven so elusive: the concept is in the vortex, not the neuron.

Cross-Realm Connections

The deepest connection is to concept navier stokes undecidability. Dyhr and Miranda (2026) showed that stationary Navier-Stokes flows are Turing complete — particle trajectories are as hard as the halting problem. If the cortex operates as a turbulent Navier-Stokes system, and Navier-Stokes is Turing complete, then predicting specific neural trajectories (hence specific decisions) may be not just practically hard but formally undecidable in ZFC.

This is a different and stronger claim than Libet's readiness potential interpretation. Libet shows a statistical correlate of decision. Turing completeness would mean specific neural decisions are as undecidable as Gödel sentences — not just unpredicted, but unpredictable in principle. See concept godel incompleteness for the connection between undecidability and Gödel's theorem.

The vortex model connects to concept quantum vortex reconnection through the FAMU-FSU vortex reconnection law (separation always faster than approach). Whether this asymmetry applies to cortical Kuramoto vortices — do phase-vortex separations in the brain also show time-reversal asymmetry? — has not been asked. If yes, the arrow of time in concept arrow of time would appear in neural dynamics.

A third connection runs to concept brain turbulence (the Deco group's clinical work): if cognitive states correspond to turbulent vortex configurations, psychiatric disorders may be turbulence anomalies — wrong vortex topology rather than wrong region activation. This would redirect psychiatric drug design toward restoring correct vortex dynamics rather than elevating neurotransmitter concentrations.

Key Facts

See Also

2026-06-27 Update: Publication Status and New Free-Will Paper

Cooray Frontiers paper: officially published as DOI 10.3389/fncom.2026.1682176 (accepted March 2, 2026); also indexed in PMC as PMC13047063. The published version confirms all results summarized above; no formal Re_neural computation appears in the final version. The open gap remains open.

Deco vortex paper: bioRxiv 2025.09.25.678538 published in Network Neuroscience (2026), confirming the vortex approach predicts cognitive-task-specific patterns that replicate across subjects.

New physics paper: arXiv:2503.19672 (March 2025) — "Reframing the Free Will Debate: The Universe Is Not Deterministic" — argues that spontaneous stochasticity in turbulent physics (not quantum mechanics) provides the physical basis for genuine indeterminism at classical scales, directly addressing the Libet-window argument. This is the clearest statement so far of the SS → free will argument in a philosophical physics context. Does not compute Re_neural, but endorses the SS framework as the most physically rigorous basis for compatibilist free will.

Formal Re_neural computation: as of June 2026, still unpublished. No paper has substituted empirical EEG/MEG parameter estimates into the Cooray kinetic equation's viscosity-analog and inertia-analog terms to derive a dimensionless Re_neural. The back-of-envelope 10–100 range remains the best available estimate.

2026-07-05 Update: Robinson Parameters and the Strong/Weak Turbulence Paradox

Substituting the Robinson corticothalamic model's empirically fitted parameters into the Cooray kinetic equation's Reynolds-analog reveals a paradox that itself is informative.

The Robinson Parameters (from Abeysuriya & Robinson 2016, J Neurosci Methods)

The Robinson group's corticothalamic model has been fit to resting EEG data across multiple populations. Canonical empirically derived values:

Parameter Symbol Value Meaning
Excitatory synaptic decay rate γ_e ~116 s⁻¹ Rate of synaptic potential decay (~8.6 ms time constant)
Inhibitory decay rate γ_i ~65–100 s⁻¹ Inhibitory postsynaptic decay
Cortical wave propagation speed v_e ~5–10 m/s (typically 7–8 m/s) Speed of excitatory wave front across cortex
Excitatory synaptic range r_e ~60–90 mm Half-width at half-max of excitatory coupling kernel
Excitatory self-coupling gain G_ee ~3.5–4.5 Dimensionless; drives alpha/spindle resonances

These are well-constrained by EEG alpha peak position and power-law slope — the model is tightly empirically anchored.

The Formal Re_neural Calculation

In the Cooray weak wave turbulence framework, the relevant viscosity analog is the linear damping term in the neural field equation — the rate at which excited wave modes relax back to equilibrium. This is set by γ_e, and the corresponding spatial damping length is:

λ_damp = v_e / γ_e ≈ 7.5 / 116 ≈ 65 mm

The effective kinematic viscosity analog (dimensions m²/s):

ν_eff ≈ v_e × λ_damp = v_e² / γ_e ≈ (7.5)² / 116 ≈ 0.49 m²/s

The Reynolds analog, using r_e ≈ 80 mm as the length scale:

Re_neural = v_e × r_e / ν_eff ≈ 7.5 × 0.08 / 0.49 ≈ 1.2

This is dramatically subcritical by any fluid turbulence standard (pipe flow: Re > 4,000 for turbulence; 2D thin films: Re > ~100).

The Paradox

The Cooray weak turbulence formula gives Re_neural ≈ 1 using Robinson's parameters — yet cortical EEG shows unambiguous power-law behavior (Beggs & Plenz 2003; confirmed in Cooray 2026). Something is wrong. The resolution:

The cortex operates in the strong wave turbulence regime. Weak wave turbulence theory (Kolmogorov-Zakharov kinetics) applies only when the nonlinear coupling is small compared to linear oscillation:

ε = nonlinear interaction rate / linear oscillation rate << 1

At neural criticality — the branching ratio σ ≈ 1 observed in cortical avalanche statistics — ε is not small; it is approximately 1. The cortex sits at the threshold where weak turbulence theory breaks down and strong turbulence renormalization is needed. Applying the Cooray weak turbulence kinetic equation to a strong turbulence system gives a Re_neural that underestimates the actual turbulent coupling by an unknown (but large) factor.

What This Means

The paradox is itself the finding:

  1. Weak turbulence formula → Re_neural ≈ 1 (subcritical): Using Robinson's parameters in the Cooray equation naively gives a subcritical Reynolds number. If taken at face value, this says the cortex is NOT turbulent.

  2. But cortex IS observably turbulent: Power-law spectra, scale-free avalanches, and Kolmogorov cascade signatures are empirically confirmed.

  3. Resolution: strong turbulence regime: The cortex operates near ε ~ 1 — exactly at the boundary of the Cooray equation's validity. The correct Re_neural requires a strong turbulence renormalization (akin to Iroshnikov-Kraichnan for magnetized turbulence, or Kolmogorov's 1941 theory for Navier-Stokes) that has not been derived for neural field equations.

  4. The key gap shifts: The missing calculation is now not "plug Robinson parameters into Cooray" (done above; gives ~1) but rather "derive a strong wave turbulence extension of Cooray valid at ε ~ 1" — a harder, original theoretical result.

  5. Libet-window implication: If the cortex IS in the strong turbulence regime, spontaneous stochasticity almost certainly applies at short timescales — the formal argument is just not yet available in published form, because the strong turbulence kinetics haven't been derived.

Calculation Result Interpretation
Weak Re_neural (Robinson + Cooray) ~1.2 Formally subcritical; weak turbulence theory inapplicable here
Cortical branching ratio σ ~1.0 (Beggs & Plenz) Strong turbulence regime confirmed empirically
Strong Re_neural (renormalized) Unknown — derivation not published This is the actual missing number
Implication for SS free will Unresolved but likely applies Observed turbulence + strong coupling → SS almost certainly present

Key Sources