The Stanford Prison Experiment: What It Actually Showed About Power (and What the Legend Gets Wrong)

About this psychology article

For fifty years, one story dominated how psychology explains cruelty: ordinary people, given a guard's uniform and a position of power, will descend into abuse almost automatically. The real histor...

In August 1971, a group of young men answered a newspaper ad offering fifteen dollars a day to participate in a psychological study of prison life. They were college students, screened for psychological stability, randomly assigned to be either "guards" or "prisoners" in a simulated jail built in the basement of a Stanford University building.

The study was scheduled to run for two weeks. It was shut down after six days.

By most tellings, what happened in those six days became one of psychology's most famous cautionary tales: guards, given uniforms, sunglasses, and authority, spontaneously became cruel and abusive; prisoners, stripped of identity and autonomy, became passive and broken; and the conclusion drawn was stark — ordinary, healthy people will behave monstrously if placed in a situation that grants them unchecked power over others. The implication reached far beyond the basement: it was used to help explain the Holocaust, prison abuse scandals, and corporate misconduct, framed as proof that situations, not individual character, are what primarily determine whether a person becomes an abuser.

Decades later, when journalist Ben Blum and historian Thibault Le Texier gained access to the original audio recordings, interview transcripts, and archived materials from Philip Zimbardo's own study, a considerably more complicated story emerged, one that changes what the experiment can actually be said to prove.

What the Study Was Designed to Show

Zimbardo's stated hypothesis was that the brutality regularly observed in real prisons resulted primarily from the prison environment and the roles it assigned, rather than from the personal characteristics of the specific individuals involved. To test this, he needed evidence that psychologically healthy people, with no predisposition toward cruelty, would behave abusively purely as a function of being placed in a guard role, and would become distressed and submissive purely as a function of being placed in a prisoner role.

Participants were randomly assigned to their roles by coin flip, controlling for any pre-existing difference between who became a guard and who became a prisoner. Prisoners were arrested at their homes by actual local police as part of the realism of the setup, brought to the basement "prison," stripped, deloused, and given numbered smocks in place of names. Guards were given uniforms, mirrored sunglasses, and told they had complete authority to maintain order, with no physical violence permitted.

What Actually Happened, According to the Archival Record

The popular narrative holds that guard cruelty emerged spontaneously, simply from being placed in the role. The archival recordings tell a different story. In one recorded exchange, Zimbardo's research team, functioning simultaneously as prison "superintendent" and lead researcher, a dual role that would be considered a significant methodological problem today, directly coached a guard identified in later accounts as "John Wayne" to be tougher on the prisoners, explicitly framing this as necessary for the study to produce useful data.

That guard, David Eshleman, stated in later interviews that he had deliberately intensified his behavior specifically because he understood, correctly, what the researchers wanted to see, describing his escalating cruelty as a conscious performance rather than a personality change induced by the role itself. Other guards, whose behavior was less dramatic, received considerably less attention in Zimbardo's own subsequent retelling of the study, which emphasized the most extreme behavior as though it were representative of the guard group as a whole.

Similarly, the prisoner who had the most famous breakdown in the study, referred to in accounts as "Prisoner 8612," stated in a 2017 interview that his distress had been substantially performed, undertaken deliberately so that he would be allowed to leave the study, which he had grown to find tedious rather than genuinely traumatic, and that once he understood he could not simply request to leave, faking a breakdown became his exit strategy.

Why This Matters More Than a Historical Footnote

These details aren't minor corrections to an otherwise sound study. They strike directly at the central causal claim the experiment is famous for. If the most dramatic instances of guard cruelty were substantially the product of direct researcher coaching rather than spontaneous role-induced transformation, and if the most dramatic instance of prisoner breakdown was a deliberate performance rather than genuine psychological collapse, the study's core evidence for "the situation alone produces this behavior" becomes considerably weaker than its reputation suggests.

This doesn't mean the study revealed nothing. It means the study cannot support the specific, strong claim it became famous for: that ordinary people will spontaneously and predictably become abusive purely by virtue of an assigned role, absent any other influence. What it may instead demonstrate is something narrower and, in some ways, more interesting — that people are highly responsive to perceived expectations from a legitimate authority figure about how they're supposed to behave, particularly when that authority figure is actively directing and reinforcing a specific performance. That reframing moves the Stanford Prison Experiment much closer to the mechanism documented in The Psychology of Obedience: Why Good People Follow Evil Orders , where direct experimenter pressure, not simply situational role assignment on its own, was the active ingredient producing troubling behavior.

The Study Zimbardo Rarely Compared Himself To, But Should Have

A useful contrast comes from a study run almost simultaneously, in 1973, by researcher David Rosenhan, in which healthy volunteers had themselves committed to psychiatric hospitals by feigning a single symptom, then behaved entirely normally once admitted. Hospital staff, operating within their own institutional roles and expectations, continued to interpret ordinary behavior through the lens of the assigned "patient" label, in some cases documenting completely mundane actions, like note-taking, as symptoms of pathology. The full study is covered in The Rosenhan Experiment Explained: Sane People, Insane Places .

Rosenhan's study shares a thematic core with Zimbardo's, but with a methodological structure that better isolates the actual variable of interest: it demonstrates how strongly an assigned label and role can shape how a person's behavior is perceived and interpreted by others operating within an institutional structure, without requiring the same degree of direct researcher intervention that undermines confidence in what Zimbardo's study actually proved.

Where the "Situations Shape Behavior" Idea Still Holds Up

None of this means situational psychology as a broader field is wrong, only that this particular, famous study is a flawed demonstration of it. Considerably more methodologically sound research supports the underlying principle that context and social structure meaningfully shape behavior. Milgram's obedience studies, despite their own ethical controversies, involved a cleaner experimental design isolating the effect of a direct authority figure's instructions on ordinary participants' willingness to cause apparent harm to a stranger.

The broader psychological principle connects usefully to The Terrifying Psychology of Elizabeth Holmes & Theranos , where organizational structure, deference to authority, and diffusion of individual responsibility within a company hierarchy allowed sustained deception to continue for years, sustained not by uniquely evil individuals but by a structure in which questioning authority carried a steep social and professional cost.

How the Study Continues to Be Misused

The Stanford Prison Experiment is frequently invoked as though it proves that literally anyone, given power over others, will inevitably abuse it — a framing used in classrooms, corporate ethics training, and popular commentary on institutional abuse. The corrected history suggests a narrower and more actionable lesson: people are powerfully influenced by what they believe a legitimate authority figure wants from them, particularly under direct pressure and reinforcement, and this susceptibility can produce troubling behavior even in psychologically healthy, carefully screened individuals. That's a real and important finding. It's a different, more specific claim than "power alone corrupts," and treating them as interchangeable has led to five decades of drawing a somewhat different lesson than the actual evidence supports.

What Actually Helps, Given the Corrected History

Recognize direct pressure from authority as the more precise mechanism, not role assignment alone. Interventions aimed at reducing institutional abuse are more effective when they target explicit and implicit signals from leadership about what behavior is expected or rewarded, rather than assuming any person placed in a position of power will abuse it regardless of surrounding influence.

Build in structural accountability, not just individual character screening. Because the study's participants were screened for psychological health and still, under direct pressure, behaved in troubling ways, this suggests that structural checks, oversight, and clear limits on authority matter more than trusting good character alone to prevent abuse.

Question studies that produced findings suspiciously aligned with what the researcher expected to find. The Stanford Prison Experiment's core flaw, direct researcher coaching toward a predicted outcome, is a useful general caution about any study where a highly involved researcher is also actively shaping the conditions being observed.

See How You Respond to Authority and Structural Pressure

Understanding how legitimate authority shapes behavior, even in people who would never predict it of themselves, is a useful lens for examining your own tendencies under institutional or hierarchical pressure.

→ Take the Manipulation Detection Assessment

Free. No account required.

Want a deeper look at how you respond to authority pressure and directional influence specifically?

→ Explore the full PsychOS Assessment Lab

Frequently Asked Questions

Was the Stanford Prison Experiment faked?

Not entirely faked, but significantly compromised. Archival recordings reveal that Zimbardo's team directly coached at least one guard to escalate cruelty, and the most famous prisoner breakdown was later described by the participant himself as substantially performed, both of which undermine the study's central claim of spontaneous, role-induced behavior.

Why was the study considered so important for so long?

Its dramatic, cautionary narrative offered a simple and compelling explanation for real-world atrocities and institutional abuse, and it was taught for decades before the archival record, made available through later investigative work, revealed how much direct researcher influence had shaped the outcome.

Does the corrected history mean situational psychology is wrong?

No. Other, more methodologically sound studies, including Milgram's obedience research, support the broader principle that context and authority significantly shape behavior. The correction applies specifically to this study's claim that role assignment alone, without direct coaching, produces spontaneous cruelty.

What happened to Philip Zimbardo's reputation after this came out?

Zimbardo has disputed some elements of the critical reassessment, but the core archival evidence, including recorded audio of the coaching, has been independently verified by multiple journalists and historians, and the study is now widely taught alongside its methodological controversies rather than as an uncomplicated demonstration of situational power.

Is the Stanford Prison Experiment still used in psychology education?

Increasingly, it's taught specifically as a case study in research ethics and the dangers of researcher bias and dual-role conflicts of interest, alongside its original, now heavily qualified, claims about the psychology of power.