Demand Characteristics in Hypnosis Studies

In any experiment, participants are not passive objects. They think about what is expected of them, form theories about what the study is testing and often try to be “good participants.” In hypnosis research, these demand characteristics are particularly influential because the entire social context of hypnosis communicates expectations.

What demand characteristics are

Martin Orne, who spent decades studying this problem, defined demand characteristics as the totality of cues that convey an experimental hypothesis to participants [1]. These include:

• Explicit instructions. What the experimenter tells the participant to expect.
• Implicit cues. The experimenter’s tone, setting, equipment and behavior.
• Prior beliefs. What the participant already believes about hypnosis from media, friends or previous experience.
• The experimental structure. The fact that a study is measuring specific outcomes tells participants those outcomes are expected to change.

Why hypnosis is especially vulnerable

Several features of hypnosis research amplify demand characteristics:

The hypnosis label. Simply telling someone they are being hypnotized activates a set of expectations: you should relax, you should respond to suggestions, your experience should change. These expectations exist before the first suggestion is delivered.

Social compliance. Hypnosis involves a one-on-one interaction where one person (the hypnotist) directs another’s experience. The social pressure to comply is inherent in the format, distinct from any specific hypnotic process.

Self-report dependence. Most hypnosis outcomes are based on what participants report: pain levels, anxiety, satisfaction, sleep quality. Self-reports are directly influenced by what people think they are supposed to report.

Unblinded participants. In most hypnosis trials, participants know whether they received hypnosis. This asymmetry guarantees different expectations between groups.

The real-simulating paradigm

Orne developed an elegant approach to detect demand characteristics [2]. In the real-simulating paradigm:

1. High hypnotizables are hypnotized normally
2. Low hypnotizables are instructed to fake being hypnotized — to act as if they were deeply hypnotized
3. The experimenter does not know which participants are which
4. The key question: can the experimenter tell the difference?

If simulators produce the same behaviors and reports as genuinely hypnotized participants, those behaviors may reflect demand characteristics rather than hypnosis. If genuine hypnotic subjects show something the simulators cannot produce, that difference is more likely a specific effect of hypnosis.

This paradigm has limitations:

• Simulators may overperform, trying harder than genuinely hypnotized people
• It requires a secondary experimenter who does not know group assignments
• It tests laboratory tasks, not clinical outcomes
• It is labor-intensive and rarely used in clinical trials

How demand characteristics affect clinical trials

In a typical clinical trial of hypnosis:

Feature	How demand characteristics operate
Recruitment	People self-select, often with positive expectations
Allocation	If unblinded, expectations diverge immediately
Treatment delivery	The hypnotist’s confidence communicates expected outcomes
Outcome measurement	Self-report allows adjustment toward expected results
Follow-up	Memory for improvement may be biased by expectations

Every stage is vulnerable. For the implications of comparison group choice, see Why Comparison Groups Matter.

Can they be controlled?

Strategies exist but are imperfect:

• Credible sham conditions can match some expectations, but participants may detect differences
• Objective outcomes (medication use, physiological measures) are less susceptible than self-report
• Blinding the assessor helps when the person measuring outcomes does not know group assignments
• Checking blinding — asking participants which group they think they are in — can reveal whether expectations are balanced

No single strategy eliminates demand characteristics. The best trials use several strategies together.

What this means for interpreting results

When a hypnosis trial finds positive results using self-reported outcomes in an unblinded comparison with a wait-list, demand characteristics are a plausible explanation for a large portion of the effect. This does not mean the effect is “fake” — participants may genuinely feel better — but it does mean the improvement may not be caused by hypnotic suggestion specifically.

For a broader discussion of how these methodological issues affect evidence interpretation, see What Is a Randomized Controlled Trial?.

What this means in practice

Consumers of hypnosis research should be aware that the social context of hypnosis naturally generates expectations that influence outcomes. Positive trial results are more convincing when they use objective outcomes, blinding, credible comparison groups and checks on whether participants’ expectations were actually matched. The more a result depends on self-report in an unblinded comparison, the more cautious the interpretation should be.

When to see a clinician first. Demand characteristics are a research concern, not a clinical danger. If you are considering hypnosis for a health condition, discuss the evidence with a qualified practitioner who can distinguish between well-designed trials and those vulnerable to these biases. The strength of the evidence, not the impressiveness of the claims, should guide your decision.