Joseph Murphy’s bedtime technique, and what sleep research says

What exactly does Murphy ask you to do?

Murphy’s premise is that the conscious mind guards the door and the subconscious mind, as he calls it, takes in whatever gets past it. The guard relaxes when the body does: “Physical inertia favors mental passivity,” he writes in chapter 2. So the moment just before sleep is, for him, the best time to hand something over.

Chapter 6 sets out ten techniques, and several of them use this moment; three spell out the method. The passing-over technique: relax, then hand the request over “in the reverie-like state”. The Baudoin technique, named after Charles Baudouin (Murphy spells it Baudoin), whose formula he quotes: “condense the idea… sum it up in a brief phrase… repeat it over and over again as a lullaby.” And the sleeping technique: in a sleepy state, repeat a short phrase over and over, for five or ten minutes, night and morning. In chapter 8 he adds the condition that matters most: no effort, no strain (“Easy does it”).

What happens in the brain as you fall asleep?

Falling asleep is not a switch. You pass first through the lightest stage of sleep, called N1, before deeper stages. Researchers call this border between waking and sleep hypnagogia. It is close to the moment Murphy describes.

In 2021, a team led by Lacaux asked 103 people to solve a number task with a hidden shortcut, then gave them a 20-minute rest while recording their brain activity. Of those who spent at least 15 seconds in N1, 83% found the hidden rule, against 30% of those who stayed awake: nearly three times as many. The effect vanished in those who slid into deeper sleep.

The limits matter. Sleep stage was not assigned at random, it was one task in a lab, and the study does not establish why it works. But the finding supports one thing Murphy noticed: the threshold of sleep is a distinct mental moment, not just tiredness.

Can you steer what the mind does at sleep onset?

Partly. In 2020, Haar Horowitz and colleagues tested targeted dream incubation with Dormio, a hand-worn device that tracks sleep onset and plays a recorded cue, “remember to think of a tree”. During a 45-minute afternoon nap, it woke people every few minutes to ask what was going through their mind. In the sleep group (25 people), 45 of 67 hypnagogic dream reports after the cue (67%) mentioned a tree, against 2 of 69 (3%) without it.

One detail in that study keeps it honest. People who stayed awake and heard the same cue mentioned a tree in 52% of their reports, and the difference from the sleep group was not statistically significant. So the cue steered attention, awake or drowsy. That is evidence that a theme rehearsed as you drift off shapes what passes through your mind in the next few minutes, and nothing more: no study shows it shapes the next day’s events.

Does sleeping on it help?

Murphy’s chapter 13 summary says it in three words: “Sleep brings counsel.” The best-known test of that idea, by Wagner and colleagues in Nature (2004), trained people on a number task with a hidden rule, then let them sleep 8 hours or stay awake. More than twice as many found the rule after sleep.

The condition is the important part: sleep did not help people who had not worked on the task first. Sleep seems to reorganise what you have already been working on. It does not deliver answers to questions you never engaged with. The practical version of Murphy is “work on it by day, then hold it lightly at night”.

Where the technique needs changing

The phrase. Murphy’s lullaby is often a statement about yourself, such as being free of a habit. In experiments by Wood and colleagues (2009), people with low self-esteem who repeated “I’m a lovable person” felt worse than those who did not; people with high self-esteem felt better, but only to a limited degree. A short scene works around this: you picture an ordinary moment instead of arguing with yourself.

The content. Picturing an idealized finished result is the version with the least support. In a series of four experiments, positive fantasies about an idealized future left people less energized, on physiological and behavioural measures, than fantasies that questioned that future, negative fantasies or neutral ones (Kappes & Oettingen, 2011). At bedtime, choose a moment of doing (the first hour in the new job, not the job offer), and leave the planning for daylight.

The effort. Here Murphy and the research agree. In Wegner’s white-bear experiments (1987), people told not to think of a white bear kept thinking of it, and when later asked to think of it, did so more than people asked to from the start. Those given one specific thought to turn to showed less of that rebound. Gripping an idea, or checking whether “it worked”, keeps the mind busy. Choose one scene, relive it once, slowly, then let it go.

A gentler version, step by step

The table maps each of Murphy’s steps to what the research supports and to the matching step in Dream Architect’s night ritual. Screen names are quoted as they appear in the app on 20 October 2026.

Common snags

You fall asleep before the end. That is the point of the technique, so it is not a failure. Write the scene before you lie down, so the part that needs a clear head is done while you still have one.

Your mind wanders to tomorrow’s problems. Murphy’s advice in chapter 18 is to bring wandering attention back to your goal. Do it gently, once or twice, and let the rest go; fighting the drift is the effort he warns against.

The phrase feels false. Make it smaller and more concrete until it is something that could plausibly happen on a normal day.

What this technique is not

It is not hypnosis, though Murphy cites hypnosis pioneers, and it is not a sleep aid. Nothing in the research says that holding an idea at sleep onset will bring it about. What it can do is end the day on something you chose rather than on a screen or a worry. If falling asleep is regularly hard for you, talk to a doctor rather than relying on a bedtime technique.

How we checked

We read the 1963 text on 20 October 2026 in a full transcription on the Internet Archive and searched each chapter for the drowsy or sleepy state (named in 7 of 20: chapters 2, 3, 5, 6, 7, 8 and 18) and for the minutes before sleep (phrases such as “prior to sleep”, “as you go to sleep” or “lullaby”: 18 of 20, all but chapters 17 and 20). Study figures come from each paper’s abstract or full text, read the same day; the exact percentages of the 2004 sleep study are not in its abstract, so we give only “more than twice as many”. App screen names and defaults were read from the app’s code the same day.

Questions people ask next