Research

A Golden-Ratio Pattern in Resting EEG

2026-03-03

The golden ratio has been studied as a mathematical and aesthetic object for centuries. This piece reports a narrower, empirical finding: a pattern in resting-state EEG spectral ratios, published in a peer-reviewed journal, that is separate from any claim about consciousness, intuition, or anomalous cognition.

In a study published in Frontiers in Human Neuroscience(March 2026), I analyzed eyes-closed resting-state EEG from 320 participantsacross two public datasets — PhysioNet EEGBCI and the LEMON Mind-Brain-Body project. The question: do the alpha/theta frequency ratio in resting EEG cluster closer to φ ≈ 1.618(the golden ratio) or to the harmonic ratio 2:1?

Peer-reviewed finding, sole-author study.This was a single-author analysis of two existing public datasets, not a new controlled experiment with founders, entrepreneurs, or intuition tasks.

What did the study actually measure?

Brain activity is dominated by oscillations grouped into bands: delta, theta, alpha, beta, gamma. We focused on the relationship between theta (4–8 Hz) and alpha (8–13 Hz) spectral centroids, and asked a precise question: for a given person, how close is the ratio of their alpha centroid frequency to their theta centroid frequency to φ, compared with 2:1?

To answer it, I built a metric called the Phi Coupling Index (PCI):

PCI = log( ( |R − 2.0| + ε ) / ( |R − φ| + ε ) )

where  R = f_alpha / f_theta,   φ = 1.618,   ε = 0.01

Positive PCI means a participant's ratio sits closer to φ. Negative PCI means it sits closer to the harmonic 2:1 ratio.

How often was the ratio closer to φ than to 2:1?

Across 320 participants, the mean theta-alpha ratio was 1.677— about 3.6% from φ, and closer to φ than to 2:1 on average. 80% of participantshad PCI > 0 (closer to φ). The pattern held across both datasets, which used different equipment, countries, and sampling rates. This is an association in resting EEG, not evidence that φ causes cognition or that the brain generally operates at φ.

Could this just be measurement noise from posterior alpha?

A reasonable objection is that posterior theta estimates can be contaminated by nearby alpha activity — a measurement artifact rather than a genuine theta signal. To check this, I re-ran the analysis using frontaltheta channels (Fz, F3, F4), which are less exposed to that specific leakage pathway.

If the result were driven by posterior-alpha leakage, the frontal version should weaken. Instead, the correlation between PCI and theta-alpha spectral-centroid convergence strengthened from r = 0.497 to r = 0.718. The stronger frontal-theta result argues against one simple posterior-alpha leakage or spatial-mixing explanation. It does not rule out every other measurement or analytic artifact, and independent replication with different preprocessing pipelines would strengthen the finding.

Why might a non-harmonic ratio matter?

In 2010, Belinda Pletzer and colleagues at the University of Salzburg showed mathematically that two oscillators coupled at the golden ratio tend not to phase-align repeatedly — their excitatory peaks drift relative to each other rather than locking. This is offered as a theoretical framework for why a non-harmonic ratio could, in principle, reduce repeated phase alignment between theta and alpha rhythms. It is the opposite dynamic from harmonic coupling, where peaks line up at predictable intervals.

Separately, Roopun and colleagues (2008) in Frontiers in Integrative Neuroscienceproposed that cortex may use non-harmonic frequency ratios to let functionally distinct rhythms remain simultaneously active with less cross-talk. The present result is consistent with that theoretical framework, but this study did not directly measure phase alignment, cross-talk, cognition, or behavior — it measured a spectral-centroid ratio at rest.

Exploratory hypothesis, not an established mechanism: harmonic ratios (like 2:1) are sometimes associated with synchronization and focused states, while non-harmonic ratios (like φ) are sometimes associated with desynchronized, resting states. Whether this maps onto insight, creativity, or intuition is an open empirical question this study did not test.

What does this mean in practice?

I am a researcher and an entrepreneur with a long-standing interest in how people combine rapid judgment with deliberate analysis. This EEG result did not measure founders, decision-making, or training effects — it measured a resting-state spectral ratio in two public datasets.

I use φ as a personal design and research motif that this finding happens to resonate with. That is a creative and conceptual choice, not a claim that the golden ratio is a validated biomarker of anything beyond the specific spectral relationship this paper reports.

Read the full peer-reviewed paper (sole-author study) in Frontiers in Human Neuroscience (Sec. Brain Imaging and Stimulation, Vol. 20, 2026): DOI 10.3389/fnhum.2026.1781338 . Open data & code on GitHub .