Consciousness under general anesthesia: Lecture by Tanmoy Bhattacharyya

Introduction and Index

The Paradox of Anesthesia and Consciousness

Consciousness under general anesthesia is not darkness, not sleep, and certainly not a mere pause in the film of experience. It is something far stranger: a reversible erasure of the very medium in which every other possible experience, dark or light, could ever appear. When propofol or sevoflurane is pushed into the vein or the lungs, the world does not fade; it ceases to exist in a way that cannot be remembered, cannot be imagined in advance, and, most disturbingly, cannot even be framed as absence once it is over. Patients wake up and say, with genuine bewilderment, “When did it start?” or “Turn it off, I’m still awake,” only to be told that the surgery ended twenty minutes ago. Time, self, pain, sound, light, thought: all vanish without residue, and then, just as cleanly, reappear.

The first thing modern anesthesia teaches us is that consciousness is astonishingly fragile. Less than one milligram of propofol per kilogram of body weight, a dose that leaves breathing, heart rate, and spinal reflexes largely intact, is enough to delete the entire contents of subjective reality for hours. The same brain that moments earlier was worrying about the children, rehearsing a conversation, or noticing the cold of the operating table suddenly stops producing anything that deserves the name experience. This is not the dimming seen in sleep or sedation; it is an on/off phenomenon closer to a power cut than to a volume knob. EEG confirms it: under deep anesthesia, the slow-delta waves of sleep are replaced first by burst-suppression (periods of complete electrical silence alternating with brief explosions) and then, at higher doses, by near-isoelectric traces. Perturb the cortex with transcranial magnetic stimulation during this silence and the evoked potential dies out locally instead of reverberating across the hemispheres as it does in the conscious brain. Complexity collapses. The brain becomes a collection of isolated islands that no longer talk to one another.

Yet the fragility is matched by an equally astonishing specificity. Spinal reflexes remain brisk, pupils still react to light, the hypothalamus continues to regulate temperature and hormones, and the brainstem keeps the heart beating. A fly could still walk across the patient’s cornea and trigger a perfect aversive withdrawal. The body is alive and responsive, but no one is home. This dissociation is so clean that anesthesiologists now speak of “disconnected consciousness” as a real, measurable state distinct from both coma and dreaming sleep. In disconnected consciousness the thalamocortical system is functionally severed from the ascending arousal pathways, and the result is a brain that can generate complex activity (alpha waves anteriorised under propofol, coherent gamma under ketamine) without any of it ever reaching the threshold of experience. The lights are on, the wiring is intact, but the electricity never reaches the bulb.

Different anesthetics carve consciousness at different joints, revealing its composite nature. Propofol and the halogenated ethers (sevoflurane, desflurane, isoflurane) act primarily through enhanced GABA-A inhibition and produce the classic “blackout” state. Ketamine, by contrast, blocks NMDA receptors and often leaves the patient in a dissociative trance: eyes open, EEG showing fast gamma activity, yet no coherent self or world. Patients later describe “being nowhere,” “floating in infinite white light,” or “watching my body from the ceiling” while simultaneously insisting they were not dreaming. Dexmedetomidine, an alpha-2 agonist, mimics natural sleep so closely that patients can be roused with a loud voice and will answer questions coherently before drifting back into a state indistinguishable from unconsciousness on EEG. Midazolam and the other benzodiazepines create profound anterograde amnesia while preserving a twilight awareness that patients later swear never happened. Each drug peels away a different layer: memory, self, agency, sensory binding, valence, time.

Pain under anesthesia is especially revealing. Even in deep surgical planes the spinal cord and thalamus still register noxious stimuli; C-fibre volleys arrive, spinothalamic tracts light up, and autonomic responses surge. But nothing hurts. The anterior cingulate and insula, the regions most tightly correlated with the unpleasantness of pain, fall silent under GABAergic drugs. Patients who are deliberately kept light enough to show autonomic responses (tachycardia, hypertension) during incision nevertheless wake up insisting they felt nothing. Pain without suffering is not a philosophical curiosity here; it is routine. Conversely, in the recovery room, when anesthetic levels are sub-hypnotic, patients sometimes report vivid, excruciating pain that occurred “during the operation” despite EEG evidence of burst-suppression at the time. Memory confabulation and state-dependent recall create a pain that was never consciously experienced yet is later felt as real. The phenomenal sting can be created retroactively by the act of remembering.

The return of consciousness is just as instructive as its departure. Propofol wears off in minutes, and the first thing to reappear is almost always orientation to sound: patients open their eyes when their name is called. Visual awareness follows seconds later, then language, then the narrative self. The order is remarkably consistent across individuals and drugs. It is the reverse of the induction sequence, suggesting that consciousness is rebuilt along phylogenetically older pathways first (auditory vigilance) and newer ones later (self-narrative). Children often wake up hallucinating vividly: animals, cartoons, parents who are not in the room. Adults sometimes report a momentary conviction that decades have passed or that they have died and returned. The newly rebooted brain briefly runs without its usual predictive priors, and the result is a fleeting window into how much of ordinary experience is top-down expectation rather than bottom-up registration.

Modern anti-psychotics, SSRIs, and anti-epileptics provide a slower, chronic version of the same lesson. Lamotrigine and levetiracetam can flatten emotional valence so thoroughly that patients say colour vision itself seems drained, yet acuity and contrast sensitivity are unchanged. Olanzapine and quetiapine sometimes produce a state patients describe as “being conscious but not caring that I am conscious,” a detachment from the fact of experience itself. MDMA and SSRIs in high doses can dissolve the subject-object boundary for hours, leaving a pure field of awareness without a centre. And every clinician who has managed post-operative delirium knows the terrifying moment when a patient is fully oriented, neurologically intact, yet describes a parallel reality (insects crawling out of the walls, relatives trying to kill them) with absolute conviction and no insight. Consciousness is intact, but its contents have become untethered from the shared world.

Taken together, the pharmacology of anesthesia and modern psychotropic medication demonstrates five inescapable conclusions:

1. Consciousness is not a single thing but a coalition of processes that can be taken apart selectively.
2. None of the usual candidates (gamma synchrony, recurrent processing, frontal-parietal integration, DMN activity) is individually necessary; each can be eliminated while leaving fragments of mentality intact.
3. The felt presence of a self, the sense of time flowing, the emotional colouring of experience, and the conviction that the world is real are all generated mechanisms, not bedrock.
4. Memory and consciousness are almost completely dissociable; vast stretches of brain activity can occur without laying down episodic traces, and traces can later be recruited to construct experiences that never actually happened.
5. The ordinary waking state we take as baseline is a highly edited, highly stabilised virtual reality maintained by continuous neurochemical fine-tuning. Change the dials even slightly and the entire theatre can collapse or be rewritten.

Anesthesia is the closest we have come, and perhaps ever will come, to a reversible lesion of subjectivity itself. It does not solve the hard problem; it makes the problem precise. The mystery is no longer why there is experience rather than darkness, but why there is ever a stable, coherent, self-containing movie at all when the projector can be switched off so easily and completely. When the drugs wear off and the patient looks at you and says “What happened?” the honest answer is: everything that makes you you was temporarily deleted, and now it has been switched back on as if nothing had occurred. The fact that this feels normal, that we treat it as routine, is perhaps the deepest mystery of all.

Tanmoy Bhattacharyya