Article
Consciousness isn't in a brain region — it's a 60-millisecond switch
You've been told consciousness lives in the prefrontal cortex, or the posterior hot zone, or some specific brain region. The search for a "consciousness spot" has driven neuroscience for three decades.
A preprint posted October 16, 2026 by the Tsao Lab at UC Berkeley suggests the whole framing was wrong. Consciousness, they found, isn't where at all. It's when.
The experiment. Postdoc Janis Hesse and colleagues placed Neuropixels probes — high-density silicon electrodes — into two brain regions of rhesus macaques: inferotemporal cortex (associated with high-level visual processing) and prefrontal cortex (linked to decision-making and awareness). They used a binocular flash suppression paradigm: show one image to one eye, then flash a different image to the other eye. The flashed image is consciously perceived; the first remains present but suppressed from awareness. The setup is a classic tool for dissociating what the brain sees from what the animal reports seeing [1].
What they found. During a constant stimulus and percept, the brain was not stable. Neural activity alternated between two coordinated states roughly 15 times per second — every ~60 milliseconds [1].
During what the authors call "conscious micro-epochs," neural activity carried a pure signal of the consciously perceived image — accurate enough to reconstruct the seen face from neural firing patterns alone. During "suppressed micro-epochs," the same neurons flipped: they now tracked the suppressed image [1].
The same neurons, in the same brain regions, carrying both the seen and the unseen — switching representation every 60 ms [2].
The same pattern in both front and back of the brain. Recordings in inferotemporal cortex (the visual pathway) and prefrontal cortex (the "executive" region) produced qualitatively identical results. The authors write: "both inferotemporal and prefrontal cortex rapidly interleave conscious with suppressed content, indicating that conscious access requires temporal selection beyond localization" [1].
This matters because decades of consciousness research have been framed as a turf war. Is consciousness in the front? The back? The Global Neuronal Workspace model (Dehaene & Changeux) emphasizes prefrontal hubs. Other theories argue posterior sensory areas are sufficient. Tsao's finding suggests both sides may have been asking the wrong question — the brain wasn't choosing a region for consciousness; it was choosing a moment.
Why previous studies missed it. If the same neurons encode both seen and unseen content at different moments, then studies that average neural activity over time — which is most of them — would produce only weak modulation. This neatly explains why prior experiments found that high-level visual cortex is only "weakly modulated" by conscious perception: they were averaging over alternating micro-epochs and washing out the signal [1][2].
The 60-ms cycle in context. Fifteen alternations per second is in the range of theta-band oscillations (4-8 Hz are slower; 15 Hz is in the beta range). It's also remarkably close to the rate of binocular rivalry itself — when each eye sees a different image and perception alternates spontaneously. The paper suggests these micro-epochs may be the building blocks of conscious access: the brain doesn't have a unified conscious state; it cycles through conscious and unconscious modes continuously.
What this doesn't mean. The preprint has not yet been peer-reviewed (it is posted on bioRxiv under a CC-BY 4.0 license). The experiments were done in macaques, not humans — though macaque visual cortex is the closest animal model we have to human vision. And "alternating every 60 ms" doesn't mean we're unconscious half the time; it means the brain's representation of content is temporally multiplexed, not spatially dedicated [1][2].
What it changes. If consciousness is a temporal phase rather than a spatial property, the entire search for "the neural correlate of consciousness" shifts from mapping regions to timing circuits. The question becomes not where in the brain but when in the cycle — and what mechanism coordinates the alternation across distributed networks. Doris Tsao, announcing the preprint on X (2.1M views as of Oct 17), described it as revealing "a pure correlate of conscious awareness in the brain during rivalry" that "overturns decades of thinking" [3].
Whether it holds up to peer review or not, the paper has already done something valuable: it forces the field to consider that the answer to "where is consciousness?" might be "anywhere, depending on when you look."
Sources:
- Hesse, J.K., Lanfranchi, F.F., McGrory, C., Shi, Y., Engel, T.A., & Tsao, D.Y. (2026). Rapid alternation of conscious and suppressed stimulus representations in the primate brain. bioRxiv. https://doi.org/10.64898/2026.10.15.759299
- Glitchwire Science Desk (2026, Oct 17). Tsao Lab Reports the Brain Toggles Between Conscious and Suppressed Content 15 Times a Second. https://glitchwire.com/news/tsao-lab-reports-the-brain-toggles-between-conscious-and-suppressed-content-15-t/
- Doris Tsao on X (2026, Oct 17). Thread announcing the preprint (contains the quoted line "a pure correlate of conscious awareness in the brain during rivalry"). 2.1M views as of Oct 17. https://x.com/doristsao/status/2111316522098733078