Can Researchers Withhold Information Simply Because Knowing It Might Change Participant Behavior?

Is “participants will behave differently if they know” enough justification?

Researchers often have a legitimate methodological concern: telling participants exactly what is being studied may influence their responses. Participants might guess the hypothesis, monitor themselves, try to help the researcher, resist what they think is expected, or otherwise behave differently because they know they are being studied for a particular reason.

It is tempting to move directly from that problem to a solution: do not tell them.

Research ethics requires another step. The fact that information can affect behavior explains why withholding it might be scientifically useful. It does not automatically establish that withholding it is ethically permissible.

Behavioral reactivity can justify consideration of withholding, not automatic permission

Why participant reactivity matters without becoming a blank check

Participants can change because they know what researchers are looking for

Human participants are not passive measuring instruments. Once people understand what a researcher is studying, they may interpret tasks differently, pay attention to particular cues, consciously regulate their responses, or infer what behavior seems expected.

In experimental research, these effects are often discussed in relation to demand characteristics, hypothesis guessing, evaluation apprehension, social desirability, and other forms of participant reactivity. The exact mechanism varies by design.

This creates a genuine methodological problem. If disclosure changes the phenomenon being measured, researchers may end up studying participants' reactions to knowing the hypothesis rather than the behavior they originally intended to investigate.

Research ethics guidance recognizes this problem

OHRP guidance on informed consent explicitly describes research in which participants are left unaware of a particular purpose because their responses might be biased if they knew beforehand what investigators were seeking. The guidance explains that an IRB may approve such an alteration when the applicable regulatory criteria are satisfied.

The Belmont Report similarly acknowledges that disclosure of pertinent research information can sometimes impair validity and recognizes incomplete disclosure in carefully limited circumstances.

So concern about behavioral reactivity is not ethically irrelevant. It can be part of a legitimate justification. The mistake is treating the existence of any possible behavioral influence as sufficient by itself.

Almost any information can change behavior somehow

If the standard were merely “this information could affect participants,” withholding would become extraordinarily easy to justify. Instructions affect behavior. Recruitment language affects expectations. The laboratory environment affects behavior. Knowing that one is participating in research may itself affect behavior.

The relevant threshold must therefore be more demanding. Researchers should identify a plausible connection between disclosure and a meaningful threat to answering the research question.

Ethical justification is much stronger in the second situation than in the first.

Researchers should identify the mechanism, not merely invoke “bias”

“Participant bias” can become a convenient umbrella term unless researchers specify what they mean. What information would participants receive? What would they infer from it? How would that inference change their behavior? Which outcome would then become difficult to interpret?

Consider a study of helping behavior. Telling participants, “We are measuring whether you help a stranger who drops several objects” could make helping unusually salient. A researcher can articulate the resulting problem: participants may intentionally help because they know helping is the dependent behavior.

By contrast, saying only that “full disclosure may bias results” provides little information about why the omission is necessary.

Methodological validity and ethical permissibility are separate questions

A technique can improve internal validity while creating ethical problems. Conversely, a highly transparent procedure can protect autonomy while making a particular research question difficult to investigate.

Ethical review exists partly because neither consideration automatically overrides the other. Researchers need to establish when withholding the full purpose can be justified without assuming that scientific validity settles the issue.

Under the U.S. Common Rule's general waiver or alteration provision, an IRB must make several findings. These include that the research involves no more than minimal risk, that the alteration will not adversely affect participants' rights and welfare, that the research could not practicably be carried out without it, and that participants will receive additional pertinent information afterward whenever appropriate.

A less informative consent process should not become broader than necessary

Suppose revealing one hypothesis would produce problematic reactivity. That does not establish a reason to conceal unrelated information about procedures, duration, compensation, confidentiality, foreseeable risks, or voluntariness.

Researchers should isolate the information creating the methodological problem and ask how much else can remain transparent. Sometimes an accurate but general description is sufficient, avoiding the need for a false cover story.

This distinction matters because withholding a specific study purpose does not necessarily require actively deceiving participants.

Changing behavior is not always a flaw

There is another question that researchers occasionally overlook: does participant awareness actually invalidate the construct being studied?

If the research question concerns spontaneous behavior, revealing the hypothesis may indeed undermine the measurement. But some research questions concern deliberate judgments, attitudes after reflection, stated preferences, or behavior under explicit information. In those cases, participants changing their responses after receiving information may be part of the phenomenon rather than contamination.

Researchers should therefore justify why reactivity is a threat to validity in this particular study rather than treating all participant awareness as bias.

Deception itself can also change participant behavior

Withholding information is sometimes presented as though it removes participant reactivity. It may instead replace one source of reactivity with another.

Participants can become suspicious that information is being withheld or that a cover story is false. Prior experiences with deceptive research may also shape expectations. The empirical consequences appear to depend on the context and form of deception.

For example, a 2022 preregistered study involving more than 2,000 crowdsourced participants compared true-purpose disclosure, incomplete disclosure, and false-purpose deception in incentivized honesty tasks. The researchers found no robust evidence that false-purpose deception changed honesty in those particular tasks, while participants nevertheless expressed reservations about researchers using false-purpose deception. The finding should not be generalized to all research settings, but it illustrates why researchers should not simply assume that deception removes behavioral effects.

The possibility of refusal is not the kind of behavioral change researchers should try to prevent

There is an important difference between participants changing the behavior being scientifically measured and participants changing their decision about whether to participate.

If researchers want to withhold information because people might refuse enrollment after learning it, the ethical problem becomes much more serious. Informed consent exists precisely so that prospective participants can use relevant information when deciding whether participation is acceptable to them.

The Belmont Report specifically states that information about risks should never be withheld for the purpose of eliciting cooperation. Preventing an informed refusal is therefore fundamentally different from preventing hypothesis awareness from contaminating an experimental outcome.

When changed behavior is the methodological problem

Now change one fact. Suppose the researchers want to conceal that participants will encounter a substantial foreseeable risk because disclosure might make fewer people enroll. That is not simply another strategy for controlling participant reactivity. It interferes directly with the person's ability to make an informed participation decision.

When “avoiding bias” becomes too easy an explanation

Turn “participants might behave differently” into a testable methodological argument

If withholding information seems necessary, make the reasoning explicit. Identify the information, the expected participant inference, the resulting behavioral change, and the scientific inference that would be compromised.

That exercise often reveals whether the concern is substantial or merely conventional. It may also reveal that only a narrow piece of information needs to remain undisclosed.

The ultimate question is not whether withholding would improve the study. It is whether the improvement establishes the kind of scientific necessity that can justify reduced disclosure under the framework governing the research.

Before withholding information to prevent participant reactivity, check your reasoning

Frequently asked questions about withholding information to prevent bias

Preventing behavioral change can be a reason for withholding, but not the whole justification

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.