H567/03 Section B Option 2 - Haney et al. (1973) and reducing reoffending
Learn the design and findings of the simulated-prison research, evaluate its situational interpretation and ethics, and apply a carefully supported reform strategy.
Testing a situational explanation of prison behaviour
Haney, Banks and Zimbardo examined whether a prison-like situation could produce hostile and submissive behaviour among screened volunteers.
The simulation took place at Stanford in 1971; the prescribed Naval Research Reviews report was published in 1973. A dispositional account locates behaviour in enduring characteristics of prisoners or guards. The researchers instead explored situational influences such as roles, power, anonymity and loss of control.
A newspaper advertisement offered male volunteers $15 per day for a study of prison life lasting up to two weeks. Seventy-five respondents were screened through questionnaires and interviews about health, background and behaviour. Twenty-four were selected as comparatively stable, mature and minimally involved in antisocial behaviour. They were mainly White, middle-class male college students in the Stanford area; one participant was Asian.
Half were randomly assigned to guard and half to prisoner roles. Screening and random allocation reduced some pre-existing differences between the role groups; volunteering for a prison study still creates self-selection, and selection was not a representative random sample.
The report's sample paragraph contains an inconsistency: it first says 22 participated, then specifies 24 selected, two standby prisoners never called and one standby guard withdrawing. It explicitly states that the analysis used 10 prisoners and 11 guards. Distinguish the 24 selected from the 21 analysed instead of treating every reported total as the same quantity.
This was an experimental simulation with independent role groups and extensive observation and self-report, rather than a natural experiment in an existing prison. Role allocation was manipulated; many features of the simulated environment operated together.
Creating and observing the simulated prison
The researchers constructed a basement prison setting intended to create an unequal relationship between guards and prisoners.
A 35-foot corridor contained three converted cells, an enclosed yard and a very small solitary-confinement space. Prisoners remained in the setting continuously; guards worked three-person, eight-hour shifts and returned to ordinary life outside their shifts. Zimbardo acted as superintendent and a research assistant as warden.
Guards wore khaki uniforms and reflective sunglasses and carried symbols of authority, including batons and whistles. Prisoners wore numbered smocks, stocking caps and an ankle chain, and could not keep personal belongings. These arrangements reduced individual distinctiveness and emphasised different status and power.
At an orientation the day before prisoner induction, guards were told to maintain reasonable order and were explicitly prohibited from physical punishment or aggression. They were not given a complete script for every response, but did receive a task, administrative guidance and role expectations. They also helped prepare the setting. It is inaccurate to say that the guards received no instructions.
With cooperation from local police, prisoners were unexpectedly arrested at home, processed at a police station and blindfolded before being taken to the simulation. At induction they were stripped, sprayed with a deodorant preparation, given uniforms and addressed by numbers. Counts initially lasted about ten minutes but became longer; routines and access to privileges changed as guards exercised control.
Researchers directly observed and recorded interactions on audio and video. They also used questionnaires, mood and personality measures, guard reports and interviews after the study. Combining behaviour and self-report gave different perspectives, although recordings did not cover every location and observers also held institutional roles.
Behaviour, distress and the authors' interpretation
The simulation produced increasingly negative interactions and serious distress, but participants did not all respond identically.
Interactions were often hostile and impersonal. Guards frequently issued commands, while prisoners became more passive and dependent after initial resistance. Five prisoners were released early: four after extreme emotional reactions such as anxiety, crying or rage, and a fifth after a psychosomatic rash. The planned two-week study ended after six days.
The authors also described variation. Some guards were tough but fair, some engaged in cruelty and harassment, and some were relatively passive. Some prisoners coped better than others. A claim that every guard became abusive would misrepresent the report.
The researchers interpreted the findings as evidence for the power of the situation. Randomly assigned authority, uniforms and group norms could encourage guards to use control. Prisoners' reduced identity and arbitrary consequences could weaken initiative. The authors likened their passivity to learned helplessness: if an action does not reliably improve an unpredictable situation, efforts to control it may diminish. This is an interpretation, not a separately isolated test of that mechanism.
They argued that changing the institution and guard training deserved attention rather than explaining all problems as defective personalities. The study connects social roles, conformity and power to the psychological effects of imprisonment. It provides no follow-up measurement of recidivism and did not compare rehabilitation programmes.
The findings support a situational influence, not complete determinism. Differences within the same role leave room for individual choices, expectations and coping. Explaining how power shapes behaviour also does not remove moral responsibility for harmful actions.
Continue this lesson.
You’re previewing NovaLearn without an account. Create a free account to read the rest.
Create free account