Lucid Dreaming: Who Is Steering, and Where Did the Critic Go
A man cornered me by the booth at half four in the morning, still holding an empty cup, and told me he had been flying over Rotterdam an hour earlier. He meant in a dream — he had known it was a dream while it happened, and had decided where to go. He wanted to know whether that was the same thing I do to people on stage. I told him what I'll tell you: no, and also yes, and the interesting part is where the line falls.
What lucid dreaming actually is
Lucid dreaming is dreaming while knowing that you are dreaming. That is the whole definition, and it is smaller than the marketing around it. The dream carries on — imagery, emotion, the physiology of REM sleep — but a piece of ordinary self-knowledge switches on inside it: this is not the room, this is not happening, I am asleep. Control over the dream is common at that point but not part of the definition. Plenty of people become lucid and get as much say over the plot as a passenger has over a train.
Surveys keep landing in the same place: about half of people recall at least one lucid dream in their lives, and a much smaller group have them often enough to plan anything. So it isn't rare and it isn't a gift. It's a variable capacity, unevenly distributed — a sentence I could use about suggestibility without changing a word.
Why we know it is real and not a story told after waking
For a long time the honest objection was unanswerable: reports of awareness inside a dream are made afterwards, by an awake person, who is reconstructing.
Then a very tidy piece of work solved it. Keith Hearne got there first, in Hull in 1975, with a practised lucid dreamer called Alan Worsley; Stephen LaBerge arrived at the same trick independently at Stanford a few years later, and it was his version the field noticed. Both used one of the few parts of the body that REM sleep leaves running. Voluntary muscles are switched off — that's why you don't act out dreams — but the diaphragm and the eyes are among the exceptions, which is why you keep breathing and why eye movement still shows up on the electro-oculogram. So the agreement was made in advance: when you realise you're dreaming, move your eyes in a specific pattern, something no sleeping eye does by accident. The sleeper went into REM. The signal came out of the REM. The state was verified by the person inside it, from the far side of unconsciousness.
Notice the structure. It didn't argue about the experience; it found something measurable the experience could do — precisely what I keep asking of my own trade in is hypnosis real. Later work has gone further, with sleepers answering simple questions mid-dream through agreed signals. Sleep talks back now, in a very limited vocabulary.
The techniques, described as they actually work
A dream journal. Everything else rests on this and most people skip it. Written on waking, before the phone, before standing up — dream recall collapses within minutes of getting vertical. The point is not the archive. You cannot notice you are dreaming if you can't remember what your dreams are like, and after a few weeks you start seeing your own recurring nonsense: the same building, the same missed train, the same person dead for years standing in a kitchen.
Reality checks. Through the day, ask whether you're dreaming, and check it properly rather than rhetorically. Read a line of text, look away, read it again — printed words are notoriously unstable in dreams. The mechanism people miss is that this trains nothing about dreaming; it trains a habit of questioning, and habits leak across the border. Checks done absent-mindedly, twenty times a day without ever meaning it, don't leak. Intention is the active ingredient.
MILD — mnemonic induction of lucid dreams — sounds least like magic and works most like hypnosis. On waking from a dream you rehearse it, find the moment where you should have noticed, and take an intention back down with you: next time I am dreaming, I will remember that I am dreaming. Hold it rather than chant it, and drift off. Prospective memory, aimed at a place you can't consciously reach.
WBTB — wake back to bed. An alarm at roughly five hours in, ten minutes to half an hour up doing something mildly involving but not bright and not stimulating, then back to sleep. The reasoning is sound physiology: REM periods lengthen towards morning, so you drop back into a night that is mostly dream sleep with a mind slightly more awake than it should be. The most reliable amplifier in the set, and the one with a real cost, which I'll come to.
SSILD — senses initiated lucid dream — is usually run inside a WBTB. Cycle attention through sight, hearing and body sensation in slow rounds, then faster ones, without trying to produce anything, and fall asleep mid-cycle. Anyone who has run a body scan in self-hypnosis already knows this exercise: rotating attention fixation with a different label on the tin. What the cycling actually does is keep you loitering in hypnagogia, the threshold state, instead of passing straight through it.
What the evidence supports, and what it doesn't
Induction techniques work for some people, some of the time, and no method has been shown to work reliably for everyone. Reviews keep reaching the same tired conclusion: promising results, small samples, wide variation between individuals, outcomes measured mostly by asking the sleeper afterwards. Self-report is not nothing — the eye-signal work exists because someone refused to be satisfied with it — but most technique studies still run on it.
Research groups in Frankfurt and Osnabrück, among others, have kept at the harder questions: what distinguishes lucid REM from ordinary REM, whether regions associated with self-reflection come partly back online, whether the state can be nudged from outside. The picture that keeps emerging is a hybrid — REM sleep with a piece of waking self-awareness running inside it.
So: your odds go up with recall, intention and WBTB, and nobody can promise you a lucid dream on Thursday. Anyone promising one is selling.
Where this is genuinely the same machine as hypnosis
Expectation does heavy lifting in both. People who believe lucid dreaming is possible and expect to manage it manage it more often, and I can't tell you how much of that is belief creating the effect versus the effect creating the belief. The same circularity sits at the centre of my own work, unresolved.
Both also run on attention narrowed to one thing. Rotating through your senses at five in the morning and staring at a point on the ceiling until your lids get heavy are one manoeuvre pointed at two destinations — the machinery I've described at length in what trance actually is.
And MILD is autosuggestion: a phrase held with intent as you cross into sleep, aimed at a part of you that acts on it without your supervision. Identical move to the pre-sleep suggestion in sleep hypnosis, different instruction loaded. Both bet that material planted at the edge of sleep is accepted with less argument than material offered at noon.

One guided evening, written out word for word. For anyone who doesn't want to understand trance so much as run one tonight.
See the bookWhere the comparison breaks, and it breaks hard
There is no operator. Everything in hypnosis that matters — calibration, noticing that someone's breathing changed, the decision to slow down, the conversation between two nervous systems — has no counterpart in a dream. You are the room, the voice and the subject at once. Nobody is watching your face.
There is no test of response either. My trade runs on small checks: an arm that gets heavy, a hand that won't lift, a suggestion that lands or doesn't and tells the operator where they actually are. In a dream nothing can be tested, because the thing evaluating the test is made of the same material as the test. And the physiology is simply different. Trance is a waking state — the EEG finds no sleep in it at all, whatever the Greek name implies. Lucid dreaming is REM sleep with an extra feature.
The critical faculty is the part people get backwards. In hypnosis the critic goes quiet and the world stays real. In a lucid dream the critic wakes up — that is the entire event — and the world stays false. A reality check is a critical faculty deliberately reinstalled where it usually isn't. This is the one practice I know of that people take up specifically to get more self-monitoring, which makes it a useful mirror for everything the subconscious gets blamed for and credited with.
The part about your sleep, without a bow on it
WBTB works by cutting your night in half. If your sleep is already fine, an occasional broken night is a manageable price for a hobby. If it isn't fine — if you struggle to fall asleep, wake and lie there, or run on too few hours anyway — an alarm five hours into the night will make a bad situation worse, and no dream is worth that trade. Don't run it before a day that requires you to drive.
Sleep problems running for months belong in front of a doctor, not in a technique forum. For chronic insomnia, cognitive behavioural therapy for insomnia — CBT-I — is the first-line treatment with the evidence behind it, and it is not a consolation prize. If you have narcolepsy, or sleep paralysis that comes often or frightens you badly, get that looked at properly before you go deliberately fragmenting your nights. Nothing here treats or cures anything, and I am not offering it as therapy.
There's a quieter cost too. Chase lucidity hard enough and you sleep worse while thinking about sleep constantly — the exact mental posture insomnia feeds on. If bedtime starts to feel like a task, stop for a month.
What I actually take from it
I don't teach lucid dreaming and I don't sell it. What draws me is that it's the only altered state most people will ever practise on purpose, alone, with nobody's hand on the dial — and it circles the question my whole trade circles. Who is steering right now? Something in you built the street, the weather and the stranger's face, and something else in you noticed. Both were you. Neither was in charge the way you assume you are in charge at this moment, reading this.
People come to my work for the version of that question with another person in the room and consent written into it, which is what my books are for. But the machinery isn't foreign. It runs every night, unpaid, whether or not you turn up to watch. And unlike the past lives people go looking for in a chair, this one leaves a signal you can catch on an instrument — which, in a field this full of claims, I find almost moving.
— Mira
FAQ
How do you lucid dream if you have never had one? Start with recall, not technique. Write down whatever you remember the moment you wake, before moving or reaching for a phone, and keep at it for two or three weeks. Once you know what your own dreams look like, add reality checks done with genuine intent, then MILD on early-morning wakings. Most people who get nowhere skipped straight to the exciting techniques and never built the recall underneath.
Which lucid dream techniques actually work? The combination that comes up most consistently is a dream journal plus deliberate reality checks plus MILD, run inside a wake back to bed window. Results vary enormously between individuals, and induction studies are small and lean heavily on what sleepers report afterwards. Nothing works on demand for everybody, and any method sold as guaranteed is being oversold.
Is lucid dreaming the same as hypnosis? No, though they share expectation, narrowed attention, and suggestion given to yourself near sleep. The differences are structural. Hypnosis happens awake, usually with another person calibrating and testing responses; lucid dreaming happens in REM sleep with nobody outside the dream. And the critical faculty moves in opposite directions — quieter in trance, switched back on in a lucid dream.
Is wake back to bed bad for you? It deliberately interrupts your night, so it costs sleep quality on the nights you use it. For someone sleeping well, occasional use is a manageable trade. For anyone with existing sleep problems, shift work, or a demanding morning ahead, it is a bad idea. Sleep trouble lasting months should be assessed by a doctor, and CBT-I is the established first-line treatment for chronic insomnia.

One guided evening, written out word for word. For anyone who doesn't want to understand trance so much as run one tonight.

Eighteen complete sessions written out to read aloud — from the first agreement to the way back. The book for turning one attempt into a repertoire.

The foundation as a course rather than a reference: language, timing, reading the other person — in the order you actually learn them.