Learn it → Apply it → Save as Skill
Ordinary, decent people can do serious harm when a system and a situation are built to produce it, not because their character suddenly changed, but because roles, authority, anonymity, and incentives pushed harder than individual disposition could resist. Psychologist Philip Zimbardo built this argument into a full framework after directing the 1971 Stanford Prison Experiment, and later used it again as an expert witness in the Abu Ghraib prisoner abuse case. The same lens now shows up constantly in management, compliance, and culture design: if you want to prevent misconduct, the more reliable lever is usually not finding "better" people, it is redesigning the system around the people you already have.
Dr. Wallace’s Favorite Mental Models is designed to turn useful ways of thinking into reusable tools. Each model gives you a fast visual overview, deeper interactive learning, practical application exercises and an AI-ready Markdown skill you can save for future use.
Instead of merely reading about mental models, you can gradually build your own AI skills library: a collection of portable reasoning skills that can help ChatGPT, Claude, Grok and other AI assistants apply the mental models you value.
Zimbardo's framework separates three levels of explanation for conduct: the Person (individual character and disposition), the Situation (immediate roles, group dynamics, and environment), and the System (the institutions, incentives, and power structures that create and legitimize the situation). The central claim of situationism is that System and Situation are usually underweighted compared to Person when people explain why wrongdoing happened, even though they are often doing most of the causal work.
Important: these are the documented facts of the 1971 study, not an invented illustration. Philip Zimbardo directed 24 male college students, randomly assigned as "guards" or "prisoners," in a mock prison built in the basement of Stanford's Jordan Hall, paid $15 a day. Graduate researcher Christina Maslach visited on day five, objected to the brutality she witnessed, and challenged Zimbardo directly on the study's ethics, a conversation that led him to end it after six days instead of the planned fourteen.
The model is most useful the moment someone frames a conduct problem purely as a people problem, "we just need better hires" or "that person was simply unethical," and you want to check whether the system around them is quietly making bad conduct more likely.
The goal is not to memorize a philosophy of free will. It is to notice, on purpose, when an explanation for misconduct has stopped at "who did this" without asking what about the role, the incentives, or the oversight made it more likely to happen at all.
Separate explanations into Person (individual character), Situation (immediate roles, group dynamics, environment), and System (institutions, incentives, oversight). Most informal explanations stop at Person and never reach the other two.
Look for the specific forces that tend to override character: assigned roles and uniforms, diffused or absent authority and accountability, anonymity and deindividuation, time pressure, and incentive structures that quietly reward the behavior you say you do not want.
Compare "replace the people" against "redesign the roles, oversight, and incentives." The situationist bet is that changing the system around people usually moves outcomes faster and more durably than changing which individuals are in it.
Situational and systemic pressure explains the mechanism behind misconduct. It does not erase the fact that individuals still made choices inside that system and can still be held accountable for them.
Open a use case to see what the model changes.
Philip Zimbardo served as an expert witness for the defense of U.S. Army Sergeant Ivan Frederick, a guard convicted in the 2003 to 2004 Abu Ghraib prisoner abuse scandal. He argued that a chaotic chain of command, dehumanizing conditions, and a near-total lack of oversight were central to understanding how a previously unremarkable soldier participated in abuse, not that this erased individual responsibility, but that it explained the mechanism.
When the Wells Fargo fake-accounts scandal became public in 2016, the pattern was not a handful of rogue employees. An aggressive internal cross-selling culture and quota pressure led many ordinary bank employees to open large numbers of unauthorized customer accounts over several years, widely reported as involving millions of accounts. A single bad actor explains a single incident; an incentive structure explains thousands of similar ones.
At a generic company, repeated harassment complaints against different managers in different departments often share a common root: no channel to report safely, no real consequence for the behavior as long as results are good, and a culture where staying quiet protects a career more than speaking up does. Firing the latest offender without changing the reporting and incentive structure usually just produces the next one.
When a large share of a cohort cheats on the same assignment or exam at the same time, it is rarely because an unusual number of individually dishonest students happened to enroll together. It is more often a sign of intense, zero-sum grading pressure, weak proctoring, and a visible sense that "everyone does this," which is the situational and systemic side of the same mechanism.
Hazing that persists across graduating classes in fraternities, sports teams, and military units is rarely explained by a continuous supply of unusually cruel individuals. It usually persists because the role of "new member" and "veteran member" is built into the group's structure, oversight is thin or absent, and the ritual is handed down as the normal cost of belonging.
When a clearly written policy is routinely ignored by otherwise competent, rule-following employees, the more useful question is usually not "why are our people dishonest." It is what the daily incentives, time pressure, and unofficial rewards in that role are actually telling people to do, which often contradicts the official policy on the wall.
This is an illustrative example designed to teach the method.
Expand each item for the research behind the model and what it does not claim.
Psychologist Philip Zimbardo laid out this argument in his 2007 book "The Lucifer Effect: Understanding How Good People Turn Evil." The framework examines behavior at three levels: the System (institutions, power structures, rules), the Situation (immediate environment, roles, group dynamics), and the Person (individual disposition and character), arguing that System and Situation are usually underweighted relative to Person in how people explain wrongdoing.
Zimbardo directed the study at Stanford University from August 14 to 21, 1971, six days into a planned two weeks, with a research team that included Craig Haney and Curtis Banks. 24 male college students were randomly assigned as "guards" or "prisoners" in a mock prison built in the basement of Jordan Hall. Prisoners were arrested by real Palo Alto police, strip-searched, and given ID numbers; guards wore uniforms and mirrored sunglasses and carried wooden batons. Participants were paid $15 a day. Guards engaged in escalating psychological abuse, including restricting bathroom access, humiliation, and degrading tasks; some prisoners rebelled, others became passive and withdrawn. Graduate researcher Christina Maslach visited on day five, objected to the brutality she witnessed, and challenged Zimbardo on the study's ethics, and her objection led him to end the experiment early, after six days.
Not as cleanly as its popular retelling suggests, and this page should be honest about that. French researcher Thibault Le Texier published a detailed 2018 investigation, "Debunking the Stanford Prison Experiment," drawing on Zimbardo's own archives, arguing that guards were actively coached and directed toward aggressive behavior rather than spontaneously inventing it, a claim several original participants later corroborated. Separately, the 2002 BBC Prison Study by Alex Haslam and Steve Reicher deliberately did not direct participants' behavior, and produced materially different results: the groups did not spontaneously descend into tyranny the same way. Mainstream psychology today treats the Stanford Prison Experiment as a flawed, likely-staged demonstration rather than rigorous science.
Because the broader situationist thesis, that systems and situations strongly shape conduct, has independent support elsewhere. Stanley Milgram's obedience experiments at Yale University in the early 1960s are the classic companion evidence: ordinary volunteers, directed by an authority figure, administered what they believed were escalating and dangerous electric shocks to another person, and a majority complied far further than most observers predicted. Real-world organizational cases, including Abu Ghraib and the Wells Fargo fake-accounts scandal, add further support from outside the laboratory. The honest summary is that the Stanford Prison Experiment is scientifically discredited in its particulars, while the broader point it became famous for illustrating is separately supported by other research and real cases.
Hannah Arendt's concept of the banality of evil, from her 1963 book "Eichmann in Jerusalem: A Report on the Banality of Evil," covers the 1961 Jerusalem trial of Adolf Eichmann, a chief logistical architect of the Holocaust. Arendt argued Eichmann was not a sadistic monster or fanatical ideologue but an unremarkable bureaucrat whose participation stemmed from thoughtlessness, reliance on clichéd official language, careerism, and a failure to think independently about what he was doing, evil enacted through ordinary bureaucratic conformity rather than innate wickedness. This reading is genuinely contested: critics including David Cesarani and Bettina Stangneth argue Arendt saw only part of the trial testimony, and that later evidence, including Eichmann's own private recorded statements, shows more personal antisemitic conviction and pride in his role than she credited. Treat this as a contested historical judgment, not a settled fact, and as a related but distinct idea: Arendt's focus is bureaucratic thoughtlessness inside a system, while this model's focus is the wider set of situational and systemic pressures that shape conduct.
It is a flawed, likely-staged demonstration. Thibault Le Texier's 2018 investigation and the 2002 BBC Prison Study both undercut it as rigorous science, even though the broader situationist point survives on other evidence.
Explaining the mechanism behind misconduct is not the same as excusing it. People still make choices inside constrained situations, and individual accountability for a specific act does not disappear because the system made that act more likely.
Even in Milgram's experiments, a meaningful share of participants refused to continue. Situational pressure changes the odds of misconduct; it does not make every individual's choice identical or inevitable.
The Fundamental Attribution Error is about misjudging someone else's behavior as dispositional rather than situational. This model is about how systems and situations actually shape conduct, a related but different claim.
Groupthink describes how conformity pressure distorts a group's decisions. This model is broader: it covers how systems, incentives, and roles shape individual and organizational conduct, not only group decision-making.
A new incentive structure or oversight process does not retroactively fix harm that already happened, and it still needs enforcement and monitoring over time, not just a one-time redesign.
if an individual knowingly broke a rule they understood, situational analysis explains the mechanism; it is not a moral alibi that removes their responsibility for that specific act.
not every instance of wrongdoing traces back to a system. Sometimes one person's choice really is the best explanation, and reaching for a systemic story can become its own way of avoiding a plain conclusion.
it is a tool for designing better systems and understanding mechanism, not a script for explaining away every scandal as "the system's fault" without checking whether that is actually true in this case.
This browser tool structures your thinking. It does not send your information anywhere or perform AI analysis.
Save as Skill converts the model into portable Markdown instructions that teach an AI when and how to apply situationist analysis, what questions to ask, what evidence to seek, what mistakes to avoid, and how to structure its analysis.
Save the skills you find useful and organize them into your own reusable library. The Markdown is intentionally model-neutral, so it can be used wherever an AI system accepts uploaded files, project knowledge, custom instructions or reusable text context. Platform capabilities differ, so this is a portable skill file rather than a claim of one-click installation.
No. The model explains mechanism, why ordinary people are more likely to behave badly under certain conditions, it does not supply a moral exemption. Even Zimbardo's own use of the framework, as an expert witness in the Abu Ghraib case, was offered to help explain how a soldier's participation in abuse came about, not to argue that no one involved should be held responsible. Keeping both ideas in view at once, mechanism and accountability, is the actual skill this model is teaching.
Treat a recurring conduct problem as a design question before treating it as a hiring or firing question. Ask what roles, incentives, oversight gaps, and group pressures are present, change those first, and keep individual accountability for what already happened as a separate, still-necessary track rather than the only track.
When a pattern of misconduct repeats across different people in the same role, stop asking only "who did this" and start asking "what about this role makes this outcome likely." Redesign the system around that answer, and treat the Stanford Prison Experiment as a flawed but useful illustration of the idea, not as proof of it.