How does general anesthesia switch off consciousness?

The question

On 16 October 1846, in Boston, a dentist named William Morton had a patient breathe ether while a surgeon removed a tumour from his neck. The patient felt nothing. Since then, general anaesthesia has been given hundreds of millions of times, and it has become one of the safest routines in medicine.

Yet the basic question remains open. A few breaths of a gas, or a few millilitres of a drug in a vein, and within a minute a person's inner world disappears; stop the drug, and it returns. What exactly do these molecules do to the brain to make experience stop? And because anaesthesia is the only way we can switch consciousness off and on reliably, on demand, the answer would tell us something about consciousness itself.

What we know

• Anaesthetics are chemically very different: ether, chloroform, nitrous oxide, the noble gas xenon, propofol, ketamine. Around 1900, Hans Meyer and Charles Overton noticed that their potency tracks how well they dissolve in oil, which for decades suggested they act by disturbing cell membranes as a whole.
• In 1984 Nicholas Franks and William Lieb showed that anaesthetics inhibit a pure protein, with no membrane present, at the concentrations that matter clinically. Today the main view is that anaesthetics act on specific proteins, above all receptors and ion channels on neurons.
• Different drugs have different main targets. Propofol and etomidate mainly strengthen inhibitory GABA-A receptors: mice with a single mutation in one of these receptors become largely resistant to them. Ketamine, nitrous oxide and xenon act mainly by blocking NMDA receptors, which carry excitatory signals.
• Anaesthesia is not simply a switched-off brain. Under propofol, the brain shows slow waves and strong rhythms in the frontal cortex. Under ketamine, activity can even increase, and patients often report vivid dreams.
• Unresponsive does not mean unconscious. In a large UK audit, awareness during surgery with later recall happened roughly once in twenty thousand anaesthetics. In studies where one arm is kept free of muscle relaxants, a few percent of patients squeeze a hand on command right after the start of anaesthesia, and most remember nothing afterwards.

Main hypotheses

Sleep circuits are hijacked

Many anaesthetics seem to act on the same brainstem and hypothalamic circuits that control sleep and wakefulness, pushing the brain into a sleep-like state. The idea explains why stimulating arousal centres can speed waking up in animals, but not why anaesthesia goes much deeper than any sleep.

The cortex stops communicating

Consciousness may require different cortical areas to exchange information, especially feedback from frontal to posterior regions. Anaesthetics would break this integration: local areas keep working, but the whole stops forming one picture. Magnetic stimulation studies support this: under propofol the brain's response stays local and simple, while under ketamine, with its dreams, it remains complex.

The thalamus as a switch

The thalamus relays signals to the cortex and was long considered the key target. It is now debated whether the thalamus switches the cortex off, or only follows changes that begin in the cortex.

No single mechanism

Perhaps each class of drugs reaches unconsciousness by its own route, and "anaesthesia" is a shared end point, not a shared mechanism.

What would settle it

• Finding the point in the brain where different drugs converge — or proving that there is none.
• A reliable measure of experience that doesn't depend on the patient's responses, tested on the transition into and out of anaesthesia.
• Being able to trigger or reverse unconsciousness by acting on a specific circuit, without a drug, and predict the result in advance.