Count Back From Ten

The seam

Here is your own experience, laid out end to end. There is nothing missing from this diagram -- that is what makes it a strange diagram.

Now the confession

We did not take four hours from you. We can't. And we could have printed a fake clock reading half past five -- but you would have known it was a lie, and the lie was never the point, because of this:

That is the exhibit. You cannot feel the length of an interval you weren't present for. A gap of eleven seconds and a gap of eleven hours are, from the inside, exactly the same size -- which is no size. The only way anyone finds out how long they were gone is by asking somebody, or by looking at a clock. Your own experience will never tell you, because your own experience is not on the list of things that were happening.

Which is why this room did not need to steal your afternoon. It only needed to steal a number you had already started saying.

What is actually happening under there?

It is not sleep

The comparison everyone reaches for is wrong, and knowing why is most of what this room has to give you. Exhibit № 20 scrubbed you through an ordinary night: sleep is a busy, structured, cycling thing. You dream. You turn over. A loud enough noise brings you back, because sleep has a door and the door can be knocked on.

General anaesthesia has no door. Nothing anyone can shout will bring you back -- not pain, not your own name, not the fire alarm. You return when the molecules leave, and not one second before. The brain state is not deep sleep; on an EEG it does not look like deep sleep; and at surgical depth it can be driven all the way to burst suppression, where the cortex falls silent for seconds at a stretch and then twitches out a burst of activity, over and over. There is no natural sleep that does that.

The measurement that can tell

Here is the part that should not work and does. Fire a magnetic pulse into the cortex and record the echo with EEG -- the field calls this zap and zip, because the second step is to run the echo through a compression algorithm and see how badly it compresses.

A conscious brain gives a complicated answer: the pulse spreads, comes back, reverberates in a pattern that is both widespread and intricate, and compresses badly. An anaesthetised brain gives a simple one: a big dumb local thud that dies where it started, and zips down to almost nothing. The number that falls out is the perturbational complexity index, and it separates conscious from unconscious states reliably enough to draw a single line through them.

Two things about that index are worth your time. The first is that it is the most clinically useful thing consciousness science has produced -- it can find the lights on in patients who cannot move or speak or answer, which used to be undetectable and is now, sometimes, a number on a screen.

The second is a beautiful complication. Ketamine anaesthesia scores high. Patients under ketamine are just as unresponsive as patients under propofol -- and they come back reporting vivid, elaborate, sometimes alarming experiences. The index sees it. So the measure is not tracking responsiveness, and it is not tracking arousal; it appears to be tracking whether anybody is home, which is a much stranger thing for a compression ratio to know. Its family lives next door in Exhibit № 29.

The rare, serious edge

This room would be dishonest if it stayed charming, so: sometimes it fails. Accidental awareness during general anaesthesia is when the lights come back on mid-operation, and the largest audit ever run on it -- the UK's fifth National Audit Project, which asked every anaesthetist in the country -- put reports at roughly 1 in 19,600 anaesthetics.

The detail that matters is what makes it terrible rather than merely unpleasant. With a neuromuscular blocking drug on board -- a paralytic, used so surgery is possible -- the rate rises to about 1 in 8,200. Without one, it falls to about 1 in 136,000. Awareness is bad. Awareness while paralysed, unable to move a finger to tell anyone, is a different order of bad, and it is why the audit exists and why the profession went looking for its own worst outcome and published the number.

That is what good calibration looks like when it costs something. There is a whole room about it: Exhibit № 33.

The switch nobody can explain

Now step back and look at what humanity actually has here.

We possess a reliable, reversible off switch for consciousness. We throw it hundreds of millions of times a year. It works on almost everybody, it can be titrated, and we can bring people back on schedule -- an achievement so complete that we have stopped finding it remarkable and now find it merely routine, which is the highest compliment engineering ever receives.

And we cannot tell you what it switches off.

The chemistry is known: propofol works largely on GABA_A receptors, turning up the brain's main inhibitory signal. But that is the story of how the drug quiets neurons -- not the story of why quieting those neurons, in that pattern, causes the world to stop being experienced by anyone. The leading accounts say that what breaks is integration: the parts stop talking to each other in the way that makes them one thing, and the one thing is what you were. That is a real hypothesis with real evidence behind it and it is a very long way from settled.

So: the switch is on the wall, it is beautifully engineered, it is used every minute of every day -- and it is wired into something nobody has ever located.

The bit to stare out of a window about

You did not experience darkness back there. You did not experience waiting, or silence, or floating, and you certainly did not experience four hours. Those are all things that happen to somebody. The gap has no inside.

Which leaves the sentence this room has been walking toward. For that interval -- yours was seconds; a patient's is an afternoon -- the most honest description available is that there was nobody there. Then the drug wore off, and a person assembled, and that person had your memories and your opinions and your name, and got up and went home and never once doubted the continuity.

If that lands strangely, you have two rooms waiting for you. Exhibit № 19 asks whether the thing that reassembled is the same thing that went under. And Exhibit № 24 asks what, exactly, was missing while you were gone -- given that your body was warm, your heart was beating, and your brain was demonstrably doing things the whole time.

You have been conscious for this entire exhibit. So far as anyone can tell.

Take it further

Emery Brown's work on the neurophysiology of anaesthesia is the readable way in, and he is very good on why "it's just sleep" is worth arguing about. The complexity measure is Casali and colleagues (2013), developed in Marcello Massimini and Giulio Tononi's lab -- see perturbational complexity index. The awareness numbers come from the 5th National Audit Project (British Journal of Anaesthesia, 2014), which is sobering and worth reading precisely because a profession did it to itself.

One honest note about what we just did to you. Nothing was taken except the rest of a sequence you had begun -- a real interruption, not a simulated unconsciousness, and the difference is total. You were present throughout. We would tell you if you hadn't been, though you would have to take our word for it, which is the situation everybody is in about this and always has been.