Confidence & Performance
Hypnosis for Test Anxiety: It Beat Relaxation at the Same Dose
Of everything on this site, this is the topic where I can make the strongest honest case. In 2022 hypnosis was randomised against progressive muscle relaxation — an active comparator, matched minute for minute — and it won, at the end of treatment and again two months later. That is a rarer result than the industry lets on.
Why the comparator is the story.Most encouraging hypnosis trials beat a waiting list, which mostly proves that doing something beats doing nothing. Zhang’s trial did not use a waiting list. It gave the control group real relaxation training, for the same amount of time, and hypnosis still came out ahead.
That is the test that usually stops hypnosis in its tracks. Here it did not.
Written by Ernest McLeodClinical hypnotherapist. An Adelaide practice, seeing clients across Australia by video. Fourteen years in practice. About Ernest →
I earn a commission when someone buys a session through this site. The number that cuts against the headline is in the same section as the headline, not buried at the bottom.
What did the best trial of hypnosis for test anxiety find?
Zhang and colleagues published in the World Journal of Psychiatry in 2022. They took 90 medical students who had screened as highly anxious — not a general student sample, people with the problem — and randomised them to hypnosis or to progressive muscle relaxation, at an equal dose.
Progressive muscle relaxation is not a sham. It is a standard, taught, evidence-supported technique, it is free, and you can learn it off the internet this afternoon. Setting it as the control is the researchers choosing the hard road.
Hypnosis beat it. At post-test, t = −3.664, P < 0.001. At two-month follow-up, still ahead: t = −2.943, P = 0.004. The advantage did not evaporate the week the sessions stopped, which is where a lot of relaxation effects go.
There was also a mechanism finding worth having. Hypnosis shifted attentional bias to threat — the automatic pull of attention toward whatever looks threatening, measured rather than self-reported — and progressive muscle relaxation did not. Something changed in how those students’ attention behaved, and it changed only in the hypnosis arm.
And the half of the same trial that argues the other way.When the authors counted how many people achieved clinically significant improvement — a change big enough to matter in someone’s life rather than on a scale — the two groups came out 64.0 per cent with hypnosis against 62.22 per cent with progressive muscle relaxation.
That is nearly identical. So the group averages separated reliably, the follow-up held, the attention measure moved — and roughly the same proportion of people in each arm got meaningfully better. Both of those are true at once, and any page giving you only the first one is selling rather than reporting.
How I read that, for what it is worth. Hypnosis did something specific that relaxation did not, and it is not a large enough something to change who crosses the line into “meaningfully better”. If free relaxation training is available to you and you will genuinely do it, do that first. If you have tried it and stalled, this trial is a real reason to think the hypnotic version is not just the same thing with a different name.
Which part of hypnosis is doing the work?
Boutin and Tosi went at that question in 1983, in the Journal of Clinical Psychology, and theirs is a test anxiety study rather than a near neighbour. Forty-eight nursing students, split four ways rather than the usual two. Four arms is rare and it is useful, because it lets you watch each ingredient being added and see which one moves the result.
Hypnosis alone beat both the placebo condition and the untreated control. Then the finding that has aged best: hypnosis combined with cognitive work beat hypnosis alone.
Forty years on, the pooled literature says the same. Valentine and colleagues’ 2019 review of hypnosis for anxiety concluded that hypnosis “was more effective in reducing anxiety when combined with other psychological interventions than when used as a stand-alone treatment”. Ramondo and colleagues, pooling 48 comparisons across 1,928 people in 2021, found adding hypnosis to CBT produced an extra d = 0.25 to 0.41 at the end and 0.54 to 0.59 at follow-up.
The practical translation is dull and consistent. A recording is something you run alongside your study, your past papers and your sleep. It is not a substitute for knowing chapter six.
Where the evidence runs out
Once, properly, and then I will stop.
Ninety students is a single trial in one country, in one profession, against one comparator. It has not been replicated. One good trial is a good reason to take something seriously and not a reason to treat the question as settled.
And the reason I made so much of the active comparator earlier is that hypnosis usually loses that test. Scheffler and colleagues’ 2018 meta-analysis of burns care, 21 randomised trials and 660 patients, shows it cleanly. Against treatment as usual, g = 0.69 [0.40 to 0.98] — solid. Against an attention control, g = 0.21 [−0.11 to 0.54], not significant. The gap between those two figures was itself significant at P < 0.001. Much of what looks like a hypnosis effect elsewhere is the effect of someone paying attention to you for half an hour.
Which is exactly why Zhang matters, and exactly why I am not going to extend it. It is evidence about test anxiety. It is not evidence about anxiety in general, and it is not evidence about exams you have not studied for.
On format, the news is better than most people expect. Yim and Derbyshire ran the subgroup analysis in 2026 and found no significant difference between self-hypnosis, therapist-delivered and taped delivery, p = 0.94. Montgomery and colleagues compared live against audiotape head to head in 2002 and reported no significant differences between methods of induction. The subgroups were small and that work was in pain. But where anyone has checked the question this site depends on, a recording has held up.
What should you actually do before an exam?
The free thing first, because it nearly matched hypnosis on the outcome that counts people rather than points: learn progressive muscle relaxation. It was the control group in the best trial on this page and 62.22 per cent of that group improved clinically. It costs nothing and it is on healthdirect.
If you want the hypnotic version, or you have already tried relaxation and it did not shift anything, the session that matches this is Overcoming Exam Nerves, which has its own page here covering what exam nerves are and are not. The catalogue also carries Failing School Tests, Test Result Anxiety and Driving Test if your version is more specific than a general exam.
Each is US$18.95, about A$27, an instant download inside the publisher’s ninety-day refund window. Run it in the weeks before, alongside the study rather than instead of it — that is the arrangement Boutin and Tosi, Valentine and Ramondo all point at. If the room you dread is an audience rather than an exam hall, the evidence is thinner and it is set out here.
Is it safe?
Bollinger went through adverse events recorded across hypnosis clinical trials in 2018 and found serious adverse events attributable to hypnosis reported at zero per cent, with other adverse events at 0.47 per cent. His own caveat: those trials treated medical conditions and procedures rather than psychiatric ones, so the safety data is thinnest where people are most unwell. Never listen while driving.
If the anxiety is running well past exam season — if it is most days, about most things — that is worth a GP appointment rather than a download. A mental health treatment plan opens ten subsidised psychology sessions a year. Hypnotherapy attracts no rebate.
Common questions
Does hypnosis work for test anxiety?
It is the best-supported topic on this site. Zhang 2022 randomised 90 highly anxious medical students to hypnosis or progressive muscle relaxation at an equal dose. Hypnosis came out ahead at post-test, P < 0.001, and at two months, P = 0.004, and it shifted attentional bias to threat where relaxation did not. One trial, not yet replicated.
Is hypnosis better than progressive muscle relaxation?
On the group averages in that trial, yes, and the advantage held for two months. On the number of people who achieved clinically significant improvement, the two were nearly identical: 64.0 per cent with hypnosis against 62.22 per cent with relaxation. Relaxation is free, so a reasonable plan is to try it first and treat hypnosis as the next step.
Does hypnosis help with exam anxiety specifically?
Test anxiety and exam anxiety are the same thing under two names, and both trials here studied students facing assessment. Boutin and Tosi 1983 tested 48 nursing students in four arms and found hypnosis alone beat a placebo condition, with hypnosis plus cognitive work beating hypnosis alone. Use it alongside revision, not in place of it.
Does a recording work as well as seeing a hypnotherapist?
Where anyone has tested it, comparably. Yim and Derbyshire’s 2026 subgroup analysis found no significant difference between self-hypnosis, therapist-delivered and taped delivery, p = 0.94, and Montgomery 2002 compared live against audiotape directly and found none. Those subgroups were small and the work was in pain rather than in exams.
Will it help if I have not studied?
No, and nothing in these trials suggests otherwise. What was measured was anxiety, attentional bias and clinically significant improvement in anxious students, not marks. Calm is worth having on the day because it lets you use what you know. It does not put anything there that revision has not.
Why do other sites quote much bigger numbers than this page?
Because the big ones do not survive checking. The percentages circulating in the hypnotherapy industry generally trace to uncontrolled case series, to self-selected clients of the author, or to no primary source at all. The figures here are from randomised trials, each quoted with the comparator it was measured against.
Free and confidential, open now. Tap to call.
Where these numbers come from
- Zhang Y, Yang XX, Luo JY, et al. Randomized trial estimating effects of hypnosis versus progressive muscle relaxation on medical students’ test anxiety and attentional bias. World J Psychiatry. 2022;12(6):801-813. doi:10.5498/wjp.v12.i6.801
- Boutin GE, Tosi DJ. Modification of irrational ideas and test anxiety through rational stage directed hypnotherapy [RSDH]. J Clin Psychol. 1983;39(3):382-391. doi:10.1002/1097-4679(198305)39:3<382::aid-jclp2270390312>3.0.co
- Valentine KE, Milling LS, Clark LJ, et al. The efficacy of hypnosis as a treatment for anxiety: a meta-analysis. Int J Clin Exp Hypn. 2019;67(3):336-363. doi:10.1080/00207144.2019.1613863
- Ramondo N, Gignac GE, Pestell CF, et al. Clinical hypnosis as an adjunct to cognitive behavior therapy: an updated meta-analysis. Int J Clin Exp Hypn. 2021;69(2):169-202. doi:10.1080/00207144.2021.1877549
- Scheffler M, Koranyi S, Meissner W, et al. Efficacy of non-pharmacological interventions for procedural pain relief in adults undergoing burn wound care: a systematic review and meta-analysis of randomized controlled trials. Burns. 2018;44(7):1709-1720. doi:10.1016/j.burns.2017.11.019
- Yim TH, Derbyshire SWG. A systematic review of hypnosis for clinical pain relief. Eur J Pain. 2026;30(8):e70369. doi:10.1002/ejp.70369
- Montgomery GH, David D, Winkel G, et al. The effectiveness of adjunctive hypnosis with surgical patients: a meta-analysis. Anesth Analg. 2002;94(6):1639-1645. doi:10.1097/00000539-200206000-00052
- healthdirect Australia. Relaxation techniques – relieving stress and anxiety [Internet]. Canberra: healthdirect Australia; [cited 2026 Sep 24]. healthdirect.gov.au
- Bollinger JW. The rate of adverse events related to hypnosis during clinical trials. Am J Clin Hypn. 2018;60(4):357-366. doi:10.1080/00029157.2017.1315927
- Better Health Channel (Victorian Government Department of Health). Hypnosis [Internet]. Melbourne: State of Victoria; [cited 2026 Sep 24]. betterhealth.vic.gov.au
Related
- Overcoming Exam NervesThe session itself, and what exam nerves are and are not.
- Hypnosis for public speakingThe same fear, an audience instead of a paper, and thinner evidence.
- Does hypnosis work?The whole evidence base in one place, good and bad.
Learn the free technique first, because it nearly matched this one on the outcome that counts people. If it does not shift anything, this is the topic where I can point at a trial that beat it.
The session on this page
Every one is an instant MP3 from the publisher at US$18.95, about A$27, with ninety days to change your mind. I earn a commission; it changes neither what you pay nor what I list.
Overcome Exam Nerves
The only session here written for exams. One recording, practised in advance.
These are self-help audio sessions, not medical treatment, a diagnosis, or a substitute for professional care. No outcome is claimed for any session sold on this site, and the trials described here are small, few and not yet replicated. The evidence concerns anxiety in students facing assessment; it says nothing about marks. Anxiety that runs beyond exam season is a reason to see a GP, who can open ten subsidised psychology sessions a year. Better Health Channel advises avoiding hypnosis with severe depression, psychosis, or a drug or alcohol problem. Never listen while driving or operating machinery, and do not stop prescribed medication without your doctor’s approval.