Is consciousness a steady lamp, or does the brain continually shift between what we see and what we do not know we see? A new preprint from Doris Tsao’s laboratory at Berkeley, published in October 2026, suggests the latter. The researchers showed that in the inferior temporal cortex and prefrontal cortex of primates, neural activity about 15 times per second—roughly every 60 milliseconds—alternated between representing the consciously perceived image and the suppressed image. The finding has not yet been peer-reviewed, but if confirmed, it moves the marker of consciousness from “a region” to “a moment.”
To study consciousness, researchers need a situation in which the stimulus remains constant while conscious experience changes. The familiar “vase or two faces” image is a simple example: the drawing on the page does not change, but we sometimes see a vase and sometimes two profiles. Tsao’s team used a method called binocular flash suppression: two different images are shown to the two eyes at the same time, and the researchers control which one reaches awareness and which one, although continuously present on the retina, remains outside awareness.
At the same time, Neuropixels probes recorded the activity of thousands of neurons in the “face patches”—specialized face-processing regions in the inferior temporal cortex—as well as in the prefrontal cortex. Previously, in 2020, the same group reported in the journal eLife that face cells represent both conscious perception and the suppressed stimulus; both could be decoded from neural activity within a single trial. The open question was: how does a single neural population carry two opposing contents at once?
Two images from one scene; the brain sends only one into awareness at any moment — Image: AI-generated / Gemini 3.1 Flash via OpenRouter
Why do they say consciousness is in “time,” not in “place”?
The answer lay in the temporal detail. When the stimulus and perception remained the same, the brain shifted about 15 times per second between two “micro-epochs”: conscious and suppressed. In conscious micro-epochs, there was a pure signal for conscious perception, and the researchers could reconstruct the seen image from neural activity. In suppressed micro-epochs, this signal inverted and made the suppressed image reliably reconstructable.
The same pattern appeared at the single-cell level: a neuron responded to the seen face in a conscious micro-epoch and to the suppressed face in the next. Still, the two states were not perfect mirror images; in conscious micro-epochs, feature tuning was enhanced, and the two periods could be reliably distinguished. Prefrontal recordings showed similar results. The authors conclude that conscious access requires a “temporal selection,” not merely a particular region; in Tsao’s own wording, conscious perception is located not in a brain region but in a specific “temporal phase.”
This idea is not unprecedented. Rufin VanRullen, in a 2016 review, showed that perception may operate in successive cycles, and in vision at least two rhythms, ~10 and ~7 Hz, coexist. Fiebelkorn and Kastner, in the rhythmic theory of attention, argued that sampling of the environment is tied to the theta rhythm (3-8 Hz), and Landau and colleagues used magnetoencephalography to show that attention alternates between two locations at a rhythm of about 4 Hz. The new finding lifts these rhythms to the level of the content of consciousness itself.
Limitations: The paper is a preprint and has not yet been peer-reviewed; the experiment was conducted in primates and only with face stimuli; and the claim of “overturning decades of theory” is currently the authors’ claim [Further verification needed: independent replication and generalization to humans and to other stimuli].
What did Freud anticipate, and what does this finding mean for psychoanalysis?
Freud, in his 1915 paper “The Unconscious,” stressed that unconscious content is not absent; it is active and influences behavior without reaching awareness. Ten years later, in “A Note upon the Mystic Writing-Pad” (1925), he speculated that the perceptual and consciousness apparatus does not operate continuously, but switches on and off in rapid periodic pulses, and he regarded this discontinuous operation as the origin of the concept of time. The resemblance to 60-millisecond micro-epochs is striking; but resemblance is not proof, and Freud was speaking about the structure of the mind, not about neurons in the temporal cortex.
A clinical distinction is also necessary: “suppression” in this experiment is perceptual suppression of an external image, not “repression” in the psychoanalytic sense—keeping a desire or painful memory away from awareness. Nevertheless, the finding offers a useful image: what remains outside awareness is present in the same networks and the same neurons, only displayed at another moment. This image fits the experience of the consulting room, where a slip of the tongue, a dream, or an unaccountable reaction shows that unconscious content has never truly been silenced.
Does this mean we lose consciousness 15 times per second?
No. In this experiment, the animal was conscious throughout and its perception remained stable. What alternated was the content represented by neural activity: one micro-period the seen image, the next the suppressed image. Our experience seems continuous because this rapid alternation is faster than we can sense.
How definitive is this finding?
Not yet definitive. The paper was posted as a preprint on bioRxiv in October 2026 and has not been peer-reviewed. The experiment was done in primates and with face images, and it needs to be repeated in other laboratories, with other stimuli, and in humans.
Is this the same as the Freudian unconscious?
Not directly. Perceptual suppression of an image differs from psychoanalytic repression of a desire or memory. But the finding shows that content outside awareness is present and active in the brain, and this is consistent with Freud’s hypothesis of an active unconscious and a discontinuous functioning of consciousness.
What relevance does this research have for daily life?
There is no immediate therapeutic application. Its value is foundational: if consciousness depends on precise temporal organization of neural activity, future research on anesthesia, coma, attention, and even artificial intelligence will need to attend to rhythms, not only to which region is active.
Conclusion
If the findings of Tsao’s team are confirmed, consciousness should be understood as a fast dance in time, not a steady lamp in one corner of the brain. The same neurons that represent what we see carry, a moment later, what we do not know we have seen. This image opens a new door for both neuroscience and psychoanalysis: what has remained outside awareness is not lost, and that is what makes it possible to reach it again in treatment.
Looking for professional support with your relationship or mental health?
Hesse JK, Lanfranchi FF, McGrory C, Shi Y, Engel TA, Tsao DY. Rapid alternation of conscious and suppressed stimulus representations in the primate brain. bioRxiv [preprint, not peer-reviewed]. 2026. DOI: 10.64898/2026.10.15.759299
Hesse JK, Tsao DY. A new no-report paradigm reveals that face cells encode both consciously perceived and suppressed stimuli. eLife. 2020;9:e58360. PMID: 33174836 · DOI: 10.7554/eLife.58360
Fiebelkorn IC, Kastner S. A Rhythmic Theory of Attention. Trends Cogn Sci. 2019;23(2):87-101. PMID: 30591373 · DOI: 10.1016/j.tics.2018.11.009
Landau AN, Schreyer HM, van Pelt S, Fries P. Distributed Attention Is Implemented through Theta-Rhythmic Gamma Modulation. Curr Biol. 2015;25(17):2332-7. PMID: 26279231 · DOI: 10.1016/j.cub.2015.07.048
Freud S. The Unconscious (1915). Standard Edition, vol. 14. London: Hogarth Press.
Freud S. A Note upon the ‘Mystic Writing-Pad’ (1925). Standard Edition, vol. 19. London: Hogarth Press.
Image Credits
Featured image — Image: AI-generated / Gemini 3.1 Flash via OpenRouter
