A new theoretical paper published in The Neuroscientist proposes that the most basic feeling of being a conscious self originates in the brainstem rather than the outer layers of the brain. Led by Marshall Devor and Mark Baron of The Hebrew University of Jerusalem, the authors argue that raw sentience is computed in an ancient subcortical structure beneath the cortex, acting as a gatekeeper for attention.
The piece challenges dominant models that locate conscious awareness primarily in the cerebral cortex, the wrinkled outer layer associated with thought and language. The authors contend that while the cortex is vital for processing information, it is not the source of the raw feeling of being conscious. They point to a 2025 adversarial study of human brain activity that failed to confirm key predictions of cortex-centred theories, leaving open where subjective experience truly originates.
The authors differentiate between access consciousness—the things a person is aware of in the moment—and phenomenal consciousness, which they label raw sentience or the “me inside” that persists even with sensory deprivation.
Across clinical and animal observations, the notion that the cortex alone creates this raw sentience has faced repeated challenges. They cite a 2007 review noting that children born without a cerebral cortex can still display conscious awareness and emotion, and a 2015 clinical case study in which a patient with extensive cortical damage to pain-processing regions could still feel and respond to painful stimuli.
Crucially, tiny disruptions in a specific brainstem region can produce the opposite effect. In a 2001 rat study involving 828 microinjections across 182 animals, injections into a tiny node—now identified as the mesopontine tegmental anesthesia area—rapidly and reversibly abolished pain responses and other signs of consciousness, whereas nearby injections produced no such effect. Later work showed that volumes of 10 nanolitres or less could generate anaesthesia within seconds.
In experiments where cells in this area were permanently destroyed with ibotenic acid, outcomes varied with the extent of damage. A small, targeted patch did not plunge the rats into a coma; they woke from surgery and resumed sleep–wake cycles with some insomnia and a diminished response to injected anaesthetics. By contrast, larger, targeted lesions that covered a wider area led to six rats entering an irreversible coma, while six others with nearby off-target injections recovered normally.
From these results, the researchers argue that the conscious-switch circuitry in rodents resides within a 3-by-4 millimetre zone of the mesopontine tegmentum, surrounding the anaesthesia area and overlapping with the parabrachial complex, a key hub for incoming pain signals.
Building an evolutionary framework, the authors propose that pain signals carry an immediate survival meaning—what they term Darwinian fitness valence—once they enter the nervous system. They suggest that the brainstem evolved sentience hundreds of millions of years ago to translate these raw tissue-damage signals into the subjective feeling of pain, while the cortex later developed the sophisticated processing required to analyse sight and sound. Rather than a separate consciousness system inside the cortex, they propose that cortical processing feeds information down to the brainstem, where consciousness is ultimately tagged onto perception.
Their model aligns with a human neuroimaging study that mapped an 18‑node network linking brainstem pathways to higher brain regions to support wakefulness. While the study focused on arousal rather than subjective experience, the researchers note that it likewise places brainstem circuitry at the foundation of human consciousness.
Several caveats accompany the theory. Much of the brainstem evidence can be reconciled with cortex‑centred explanations that emphasise wakefulness rather than the origin of phenomenal experience. For instance, coma following brainstem injury or anaesthetic administration does not, on its own, prove that subjective experience emerges there. The distinctive support for this theory comes primarily from pain—the one sense the authors argue does not require the cortex.
Even on pain, it remains possible that cortical processes contribute to some aspects of the feeling. In epilepsy patients, for example, repeated stimulation of a site in the insula can reliably trigger pain, a finding some researchers interpret as indicating a cortical origin. Devor and Baron interpret such cases as signals spreading to brainstem pain pathways, but the possibility of a cortical pain centre is not definitively ruled out.
Several methodological caveats also apply. Much of the evidence relies on translating animal and nonverbal human data into inferences about subjective experience, and the coma‑inducing findings come from a relatively small number of rats reported briefly in the paper, with limited statistical detail.
Ultimately, the authors present a hypothesis rather than a settled theory. They have not yet identified the exact neurotransmitters or firing patterns that would “mark” cortical activity, nor do they claim to resolve the philosophical hard problem of how biology gives rise to first‑person experience. They propose that brainstem neurons might communicate via broad, chemical signalling rather than precise synaptic connections, a form of analogue processing that current silicon computers cannot replicate, and they call for experiments to test these cellular mechanisms directly.
The study, titled Brainstem Theory of Consciousness, is authored by Marshall Devor and Mark Baron.
