In the basement of Stanford University’s psychology building, in August 1971, a group of perfectly ordinary young men spent six days descending into something close to hell. Twenty-four undergraduates had answered a newspaper ad. Half were randomly assigned to play prisoners. Half were assigned to play guards. The plan was to run the simulation for two weeks. It was shut down after six days because, by all accounts, the situation had become unbearable.
The professor running the study, Philip Zimbardo, emerged with a story that would change how the world thought about human nature. Give ordinary people authority within a cruel system, he said, and they will become cruel. Strip ordinary people of their identity and dignity, and they will collapse. The conclusion embedded itself in textbooks, exam syllabi, and public conversation for fifty years.
The only problem is that most of it didn’t happen the way Zimbardo said. For exam purposes you need to know both versions — the classic account that examiners still ask about, and the modern critique that gets you the top-band marks. This study guide gives you both.
| Researcher | Philip G. Zimbardo · Stanford University (with Craig Haney, Curtis Banks, David Jaffe) |
|---|---|
| Year | August 14–20, 1971 (terminated after 6 days; planned for 14) |
| Location | Basement corridor, Jordan Hall, Stanford University |
| Sample | 24 male US college students, mean age c. 21, screened for psychological stability; volunteer + opportunity sample |
| Design | Controlled observation · random allocation to “prisoner” or “guard” condition |
| Aim | To investigate whether brutality in prisons arises from dispositional factors (the people) or situational factors (the system) |
| Key Finding | Guards rapidly adopted authoritarian, abusive behaviours; prisoners showed acute psychological distress — supporting the situational hypothesis |
| Published | Haney, Banks & Zimbardo (1973), Naval Research Reviews; later in International Journal of Criminology and Penology |
| Modern Status | Disputed. Le Texier (2018), Blum (2018) and the BBC Prison Study (2002) have challenged its scientific validity |
Zimbardo wanted to test the situational hypothesis — the claim that behaviour is shaped more by the situation a person is in than by their underlying personality. Specifically, he wanted to know whether the brutality observed in American prisons in the 1960s was caused by sadistic individuals being attracted to prison work (the dispositional view), or whether the prison environment itself created brutality in anyone placed inside it.
To isolate the situation from the person, he used random allocation. If ordinary, pre-screened students — randomly assigned, with no special traits — nonetheless began to behave brutally or fall apart, the situation must be responsible. That was the logic.
For AQA Paper 1 and IB Paper 2, you must be able to state the aim, sample size, and random allocation in one or two sentences. A common 4-mark question reads: “Outline the procedure of Zimbardo’s Stanford Prison Experiment (4 marks).”
The study Zimbardo described unfolded with the rhythm of a tragedy. Day one was almost playful. By day two, prisoners had staged a rebellion, and guards crushed it. By day three, one prisoner had begun screaming and had to be released. By day six, the simulation was being terminated — ostensibly because Zimbardo’s then-girlfriend (and now-wife), psychologist Christina Maslach, visited the site, was horrified, and confronted him.
- Day 1 · Sunday Aug 15 Arrivals & mock arrests Palo Alto police arrest “prisoners” at home. Guards receive briefing — including, we now know, coaching to be “tough.”
- Day 2 · Monday The rebellion & the crackdown Prisoners barricade themselves. Guards retaliate: push-ups, removal of beds, naked sleeping, humiliation rituals. Tone shifts sharply.
- Day 3 · Tuesday Prisoner #8612 — the first “breakdown” Participant Doug Korpi screams to leave. He later confirms the breakdown was performed.
- Day 4 · Wednesday Visiting day & the priest Parents visit. A Catholic priest, invited by Zimbardo, conducts mock pastoral interviews. Realism — and staging — escalate.
- Day 5 · Thursday Maslach visits Christina Maslach sees prisoners marched to the toilet with bags on their heads. She tells Zimbardo to stop.
- Day 6 · Friday Aug 20 Experiment terminated Eight days early. Zimbardo would spend the next 50 years narrating its meaning.
Zimbardo’s interpretation was clean and dramatic. He called it the situational hypothesis: when ordinary people are placed into systems that grant them power over others, the situation, not the person, drives behaviour. Cruelty is not a property of bad apples; it is a property of bad barrels. The guards in his basement were not predisposed to brutality — the experiment had randomly assigned them. And yet they brutalised. Therefore: anyone, under the right conditions, would do the same.
This conclusion travelled fast. It was used to explain the Holocaust. Abu Ghraib (Zimbardo would later testify for the defence of a soldier convicted there). Corporate cruelty. Totalitarianism. Bullying. Zimbardo’s bestseller The Lucifer Effect (2007) restated the case in full. For exams, you should know the three findings that became canonical:
- 1. Conformity to social roles
- Guards rapidly internalised an authoritarian role; prisoners adopted submission. This supports Social Identity Theory and the concept of role engulfment.
- 2. Deindividuation
- Uniforms, numbers, mirrored sunglasses, and the absence of names reduced personal identity. Both groups behaved in ways their out-of-role selves would not, supporting Zimbardo’s deindividuation theory (1969).
- 3. Situational over dispositional causation
- Because participants were randomly assigned and pre-screened for normality, the brutality could not be blamed on personality. The situation appeared to do the causal work.
Examiners love when students connect SPE to Milgram’s obedience study (1963). Both support the situationist tradition; both have been re-examined critically. A 16-mark essay question often asks you to compare and contrast Milgram and Zimbardo.
This is where exam marks are won and lost. A clean, balanced evaluation — using research terminology — will separate top-band answers from middle-band ones. Memorise at least three of each.
- Ecological realism. The simulation produced surprisingly realistic behaviours — “mundane realism” was high.
- Random allocation. Reduced participant variables; supports the situational interpretation.
- Real-world applications. Influenced prison reform and military training; cited in the Abu Ghraib case.
- Drove ethical reform. Helped trigger stricter IRB / ethics-committee requirements still in force today.
- Rich qualitative data. Detailed video, observation logs, and interview material.
- Lack of informed consent. Participants were arrested without warning; could not fully anticipate the experience.
- Failure to protect from harm. Participants experienced significant psychological distress.
- Experimenter bias. Zimbardo played “superintendent” — he was not a neutral observer.
- Demand characteristics. Participants behaved as they thought was expected of them.
- Sample bias. All male, American, college students — not generalisable.
- No control group. No baseline to compare against; weak internal validity.
- Failure to replicate. Reicher & Haslam (2002) produced opposite findings.
For decades, criticism circulated quietly within academic psychology. Then in 2018, two pieces of journalism cracked the case open. The French academic Thibault Le Texier published Histoire d’un mensonge (“History of a Lie”), based on years of archival work. Almost simultaneously, journalist Ben Blum published “The Lifespan of a Lie” on Medium. Together, they exposed material that contradicted Zimbardo’s narrative in four important ways — these are the points that win you the highest exam marks today.
Audio archives reveal warden David Jaffe explicitly telling guards on the night before the study: “We want to get you active and involved… every guard is going to be what we call a ‘tough guard.'” The guards did not discover cruelty — they were instructed to perform it.
Prisoner #8612, Doug Korpi, whose screaming exit became the study’s dramatic centrepiece, has stated on the record that he was performing, partly to be released so he could revise for graduate-school exams.
A substantial subset of guards refused to be cruel, remained friendly with prisoners, and were criticised by Jaffe and Zimbardo for being insufficiently committed to the role. This directly contradicts the claim of “inevitable” brutality.
Zimbardo was not observing — he was directing. He spoke to guards about their performances and encouraged escalation. The setting was less a scientific study than a well-funded piece of immersive theatre, later re-narrated as natural human behaviour.
In 2002, British psychologists Stephen Reicher (University of St Andrews) and Alex Haslam (University of Exeter, later Queensland) ran a carefully ethically-reviewed replication for the BBC. They recreated the prisoner-guard structure with comparable participants and observed what happened when nobody coached the guards.
The results were almost the opposite of Zimbardo’s. Guards did not spontaneously become tyrannical. Prisoners formed solidarity quickly. The simulation eventually collapsed into a brief egalitarian commune before researchers ended it. Published in the British Journal of Social Psychology (2006), the study argued that group identity and shared belief — not the situation alone — determine whether people behave cruelly or cooperatively.
Reicher and Haslam’s replication, broadcast as a four-part BBC series, found guards became less authoritarian over time, not more. The “natural slide into cruelty” did not happen when guards were not actively encouraged to drift that way.
Conclusion: Zimbardo had not discovered a universal law of human nature. He had discovered what happens when you instruct young men to abuse other young men in a basement.
Examiners reward the precise use of psychological vocabulary. These eight terms appear in nearly every mark scheme for SPE-related questions.
- Situational hypothesis
- The view that behaviour is determined more by the environment and situation than by individual personality traits. Central to Zimbardo’s argument.
- Dispositional hypothesis
- The competing view: behaviour is caused by stable internal traits (e.g. authoritarian personality). SPE was designed to test situational against dispositional explanations.
- Conformity to social roles
- The tendency for individuals to adopt behaviours expected of them given their social position. SPE is the textbook example.
- Deindividuation
- The loss of personal identity within a group, producing reduced self-awareness and increased anti-normative behaviour. Achieved in SPE via uniforms, numbers, and mirrored sunglasses.
- Demand characteristics
- Cues within an experiment that signal to participants what behaviour is expected, threatening internal validity.
- Experimenter effects
- Influence of the researcher’s behaviour or expectations on participant responses. Especially severe in SPE because Zimbardo was a participant.
- Random allocation
- Assigning participants to conditions by chance — used in SPE to equalise dispositional variables between prisoner and guard groups.
- Ecological validity
- The degree to which a study’s findings generalise to real-world settings. SPE is often cited as having high mundane realism but contested overall ecological validity.
The questions below are drawn from genuine exam patterns. Practise writing answers under timed conditions; that habit alone will lift your grade by a band.
(A) Obedience to authority · (B) Conformity to social roles · (C) Cognitive dissonance · (D) Bystander apathy
Is the Stanford Prison Experiment on the AP Psychology exam?
Yes. The Stanford Prison Experiment is named in the AP Psychology Course and Exam Description under Unit 4 (Social Psychology) as an example of situational variables, conformity to social roles, and research ethics. It also appears in the MCAT psych/soc section and the GRE Psychology subject test.
Is the Stanford Prison Experiment on A-Level Psychology?
Yes — it is required content on the AQA A-Level Psychology Paper 1 (Conformity to Social Roles), and is also covered by Edexcel, OCR, and GCSE Psychology specifications across the UK.
Does the IB Psychology exam cover the Stanford Prison Experiment?
Yes. The SPE is one of the most cited studies in IB Psychology, used in the Sociocultural Approach and Ethical Considerations topics for both HL and SL papers.
What was the Stanford Prison Experiment?
A 1971 study run by Philip Zimbardo at Stanford in which 24 male volunteers were randomly assigned to play prisoners or guards in a simulated prison built in the basement of the psychology building. Planned for two weeks, terminated after six days when guards became abusive and prisoners began breaking down.
What is the standard exam evaluation of the Stanford Prison Experiment?
Strengths: high ecological realism, useful policy applications, contribution to ethical reform. Limitations: lack of informed consent, demand characteristics, no control group, sample bias, experimenter bias from Zimbardo’s dual role, and failure to replicate (Reicher and Haslam, 2002).
Was the Stanford Prison Experiment a fraud?
Not in the sense of fabricated data. However, the 2018 investigations by Le Texier and Blum revealed that guards were explicitly coached to be tough, a key “breakdown” was performed, and the public conclusions Zimbardo promoted exceeded what the data supported. It is no longer treated as definitive science.
How should I structure a 16-mark essay on SPE?
For AQA: 6 marks AO1 (description — aim, sample, procedure, findings) + 10 marks AO3 (evaluation). Top-band answers include both classical critique (ethics, control, sample) and the modern 2018 reinterpretation, plus the BBC replication as counter-evidence.
- Haney, C., Banks, C., & Zimbardo, P. (1973). A study of prisoners and guards in a simulated prison. Naval Research Reviews, 9, 1–17.
- Zimbardo, P. (2007). The Lucifer Effect: Understanding How Good People Turn Evil. Random House.
- Reicher, S., & Haslam, S. A. (2006). Rethinking the psychology of tyranny: The BBC prison study. British Journal of Social Psychology, 45(1), 1–40.
- Le Texier, T. (2018). Histoire d’un mensonge: Enquête sur l’expérience de Stanford. La Découverte.
- Le Texier, T. (2019). Debunking the Stanford Prison Experiment. American Psychologist, 74(7), 823–839.
- Blum, B. (2018). The Lifespan of a Lie. Medium.
- AQA (2024). A-Level Psychology Specification 7182. Unit 1: Social Influence.
- College Board (2024). AP Psychology Course and Exam Description. Unit 4: Social Psychology.
