Placebo & expectation · Oct 7, 2026

Expectation, Placebo and Hypnosis: Why Comparison Groups Matter

The same hypnosis result can look large or disappear depending on what it was compared with.

What the evidence supports

What a hypnosis study can conclude depends on its comparison group. Against a wait-list, hypnosis often looks effective; against a credible sham or another active treatment, the gap usually shrinks and sometimes disappears, as it did for falling asleep in insomnia trials. Hypnosis is not simply placebo, since some sham-controlled trials still find an added effect, but expectation clearly contributes, and separating the two is still an open research problem.

11 sources cited · sources checked Oct 16, 2026 · Read the explainer

Printed research charts on a desk under a magnifying glass, with a pencil and a mug — Reading past the headline means reading the methods.

Research methods · Sep 16, 2026

What Is a Randomized Controlled Trial? Reading Hypnosis Research

What a randomized controlled trial is, why hypnosis trials are hard to blind, and a six-question checklist for judging any hypnosis study you read.

“A single positive hypnosis RCT is a reason to look closer, not proof.”

10 sources · checked Oct 16, 2026

Measurement · Aug 26, 2026

What Is Hypnotizability? How It’s Measured and What It Predicts

What hypnotizability means, how scales measure it, how stable it is, whether training changes it and how well it predicts who benefits from hypnosis.

“Strongly predicts response in laboratory pain studies, but in clinical studies it explains only a small share of outcomes.”

12 sources · checked Oct 16, 2026

Worked example

One result, two answers

A 2015 meta-analysis of hypnotherapy for insomnia. Against a wait-list, it shortened the time to fall asleep. Against a sham, there was no significant difference. Eleven of the 13 studies had low methodological quality.

“Hypnotherapy beats doing nothing” and “hypnotherapy has not been shown to beat a credible sham” are different claims.

From the comparison-groups explainer

Time to fall asleep, standardized mean difference

Point estimate shown for the wait-list comparison. The sham comparison is drawn as a band around zero to show “no significant difference”, not a measured interval. Left of zero favours hypnotherapy.

Reading a study

Six questions for any hypnosis trial

Hypnosis trials are harder to run well than drug trials: participants cannot be blinded, most outcomes are self-reported, and many trials are small.

From the RCT explainer

Were participants randomized, with allocation concealed?
What did the comparison group receive?
Was the primary outcome registered in advance, and is it the one reported?
How many people dropped out, and was the analysis intention-to-treat?
How large was the effect, and how wide were the intervals?
Who took part, and do they resemble the people you care about?

.71

Figure in brief · hypnotizability

In one cohort, hypnotizability scores correlated at .71 across 25 years, making it one of the more stable traits psychologists measure. Researchers still disagree about what the scores actually reflect.

From the hypnotizability explainer