Protecting the People Who Make Research Possible
Modern research ethics grew directly out of past abuses, and it now rests on a well-defined set of protections. Informed consent requires that participants understand what a study involves — its purpose, procedures, risks, and benefits — before agreeing to take part. Participants retain the right to withdraw at any point, without penalty, even after initially agreeing. Confidentiality protects a participant's data from being linked back to their identity by outside parties; true anonymity goes further, meaning not even the researchers themselves can link data to a specific person.
Sometimes a study's validity depends on participants not knowing its true purpose in advance — deception is permitted only when it is scientifically justified, no less deceptive alternative exists, and it does not expose participants to more than minimal added risk. Any deception must be followed by debriefing: fully explaining the study's true purpose and any deception involved once the participant's part is complete. Every study today is reviewed in advance by an Institutional Review Board (IRB) (or equivalent ethics committee), an independent body assessing whether the risks are justified and the protections are adequate before any data is collected.
These protections were formalized in the United States in the 1979 Belmont Report, which established three guiding principles still cited across research ethics today:
| Principle | What it requires |
|---|---|
| Respect for persons | Treating individuals as autonomous, with informed consent |
| Beneficence | Maximizing benefits, minimizing harm |
| Justice | Fair distribution of research's risks and benefits across groups |
These formal protections emerged largely in direct response to specific historical failures — including, within psychology's own history, the psychological distress caused by studies like Milgram's and Zimbardo's discussed earlier in this course.
Common pitfall: assuming ethical deception simply requires debriefing afterward, with no other constraints. Deception must ALSO be scientifically justified, involve no viable non-deceptive alternative, and expose participants to no more than minimal risk — debriefing does not retroactively justify an otherwise unjustified use of deception.