Psychology › Section B · Psychological research › Topic B.4
Research ethics
Modern psychology's ethical machinery - review boards, consent forms, debriefings - was not built on goodwill; it was built on outrage. This topic covers the rules researchers must follow today, the failures that forced those rules into existence, and why the field keeps enforcing them on itself.
Why research needs rules
Every method in this course places demands on human beings: their time, their trust, sometimes their distress. Ethics is not an add-on to research design; it is a constraint the field owes its participants. The rules exist because the alternative was tried. The guidelines described in this topic - the review boards, consent forms, and debriefing obligations - trace directly to documented abuses, and their enforcement is a professional obligation, not a courtesy.
The Institutional Review Board
In the United States, any institution that receives federal research funding must run an Institutional Review Board (IRB), and human-participant studies there cannot proceed without its approval. The IRB is a committee of administrators, scientists and community members that reviews research proposals - what participants will do, what risks they face, how data will be protected - before a single participant is recruited. The community members are the important detail: the researcher's peers are not the only judges of whether a study is acceptable. The IRB meets regularly, and its approval is a precondition, not a formality after the fact.
Informed consent
Every approved study begins with informed consent: a written account of what taking part will involve and what risks it may carry. The form also sets out the participant's rights: taking part is a free choice, anyone may stop at any point with no penalty, and their data will be kept confidential, with any legal limits on that promise stated. For participants under 18, a parent or legal guardian gives consent, and the young person is usually asked to agree (assent) as well. Consent is a conversation in writing: the participant can decline or quit at any point, and the researcher's obligation continues the whole way through.
Deception and debriefing
Sometimes a study's validity requires that participants not know the exact research question - telling people "we are testing whether background music slows your reading" would change how they read. Deception is allowed only when the study's value justifies it, no honest design would work, and it is not expected to cause physical pain or severe distress. In exchange, the rules require a full debriefing at the study's conclusion: the purpose of the study, how the data will be used, why the deception was necessary, and how to get more information. The participant leaves knowing the truth - a debt the study must pay before it closes.
The Tuskegee Syphilis Study
In 1932 the US Public Health Service enrolled 600 Black men near Tuskegee, Alabama, offering free medical care, meals and burial insurance. About 399 had syphilis and 201 did not; the men with the disease were never told they had it. Penicillin was the standard cure by 1947, yet it was withheld, and some men were steered away from treatment elsewhere. Over 40 years, many died of syphilis or its complications, and some wives and children were infected. After a Public Health Service worker, Peter Buxtun, took his concerns to the press, the study was exposed and ended in 1972. The outrage led to the National Research Act of 1974, which shaped the modern IRB system.
Animal research and the IACUC
Much psychological research involves animals - mostly rodents and birds - and the rules extend there too. Every animal research proposal is reviewed by an Institutional Animal Care and Use Committee (IACUC): institutional administrators, scientists, veterinarians and community members charged with ensuring humane treatment. The committee requires experiments to minimise animals' pain and distress, conducts semi-annual inspections of animal facilities, and can stop a project cold: no animal research proceeds without IACUC approval. The ethical reasoning mirrors the human case: the researcher's obligations do not shrink because the subject cannot sign a form.
Worked example: a study's ethical audit
Walk a hypothetical proposal past every gate. A researcher wants to test whether negative feedback affects mood, planning to give some participants harsh written criticism. 1. IRB review: does the design's benefit justify the discomfort? Are safeguards in place - a limit on how harsh the feedback is, a plan to repair mood afterward? 2. Informed consent: the form describes "completing tasks and receiving evaluative feedback," honestly, including the risk of feeling bad - without revealing the exact hypothesis. 3. Confidentiality: mood data stored without names. 4. Debriefing: afterward, every participant learns the true purpose, why the feedback was scripted, and how to access support if anything lingers. 5. Check: each gate exists because a past study failed it - the audit is the Tuskegee lesson applied in advance.
Worked example: where the line falls on deception
Two hypothetical proposals withhold information; judge each against the rules. 1. Proposal A tests whether background music slows reading but is described as "a reading study." The hidden detail is harmless and the study could not work if revealed: likely acceptable, with full debriefing. 2. Proposal B tells participants they failed an intelligence test (the result is assigned at random) in order to cause real distress. The deception is aimed at harm, and a debriefing may not fully undo the hurt. 3. Verdict: an IRB would likely approve A and demand that B be redesigned or rejected. 4. Check: the question is not "did the study withhold anything?" but "could the deception harm, and can debriefing repair it?"
Common mistakes to avoid
First, treating ethics review as a formality after the design is set: the IRB judges the design itself, and its questions improve it. Second, reading informed consent as a single signature: it is ongoing, revocable, and participants can quit without penalty at any time. Third, assuming deception is forbidden: it is regulated - justified, not harmful, and followed by a full debriefing. Fourth, citing Tuskegee as ancient history: it ended in 1972, and the National Research Act of 1974 that followed still shapes IRBs. Fifth, exempting animal research from the rules: the IACUC reviews it, with inspection powers. Sixth, confusing confidentiality (data protection) with anonymity (no identity collected) - consent forms promise the first, and often both.
Key takeaways
Research ethics is enforced, not merely advised. No human-subject study proceeds without IRB approval - administrators, scientists and community members reviewing proposals in advance - and every approved study begins with informed consent: a written description of procedures and risks, the right to withdraw without penalty, confidentiality, and parental consent for minors. Deception is allowed only when justified and not expected to cause serious harm, and it requires a full debriefing: purpose, use of data, reasons for the deception. The machinery owes much to the Tuskegee Syphilis Study - 600 men enrolled in 1932, about 399 with syphilis left untreated even after penicillin became standard by 1947, exposed in 1972, followed by the National Research Act of 1974. Animal research answers to its own committee, the IACUC, with the power to refuse any project. The participant's protection is not the researcher's discretion; it is the field's law.
Flashcards
Flashcard
Institutional Review Board: administrators, scientists, community members
Test your understanding
Question 1 of 5
Before recruiting a single participant, a study involving human subjects must be approved by:
Question 2 of 5
A study uses deception to hide its hypothesis. What does it owe participants at the end?
Question 3 of 5
The Tuskegee Syphilis Study is central to research ethics history because:
Question 4 of 5
Animal research proposals are reviewed by:
Question 5 of 5
Under informed consent, a participant who becomes uncomfortable halfway through:
