- Wanted to find out if prison guards behaved brutally because they have sadistic personalities or rather because of the situation that creates such behaviour.
A mock prison was created in the basement of the psychology department at Stanford University (1973). Advertised for students to volunteer and extensive psychological testing was conducted in order to find the most emotionally stable to take part.
Roles were randomly assaigned. Prisoners were arrested in their homes by local police and were delivered to the fake prison. They were blindfolded, strip-searched, deloused and issued a uniform and a number that was to be used as opposed to their name. Guards were issued a different unifom complete with a wooden club, handcuffs, keys and mirror shades.
Social roles of the two groups were strictly divided with the prisoners rountines heavily regulated. There were 16 rules that were enforced by guards who worked in shifts, three at a time. Guards had complete power over the prisoners.
The guards eventually regarded their roles with enthusiasm and their behaviour became a threat to the prisoners psychological and physical health which meant the study had to be stopped after only 6 days as opposed to the intended 14.
Rebellions occured within 2 days. Uniforms were ripped and abuse was shouted whereby the guards retliated with fire extinguishers. The guards employed divide and rule tactics by playing prisoners off agaisnt one another. Prisoners were consistently harassed by the guards. The guards created plently of opportunites to enfore the rules and enjoyed punishing the smallest of misdemeanours. Prisoners became subdued, anxious and depressed. A prisoner had to be released on the first day as a result of showing signs of mental disturbance. Two more were released on the fourth day. As the guards identified more and more with their role their behaviour became more and more brutal and aggresive with some appearing to enjoy the power they had over the prisoners.
The experiment revealed the power of the situation to influence people's behaviour. Each individual involved in the study conformed to their roles. Even volunteers who only preformed certain functions behaved as if they were in a real prison as oppsed to a psychological study.
Control - In the selection of participants, emotionally stable individuals were chosen and randomly assigned roles in an attempt to rule out dispositional factors that may effect the findings. If the guards and prisoners behaved very differently but the roles were given only by chance then therefore their behaviour must be attributed to the situation. A strength as it increases the internal validity of the study and as a result we can be more confident in drawing conclusions about the influence of roles on behaviour.
Lack of realism - It has been said (Banuazizi & Mohavedi 1975) that participants were merely play acting (their performances based on their stereotypes of how they were supposed to behave e.g. one guard claimed he based his role on a character from Cool Hand Luke) rather than genuinely conforming to a role. Zimbardo argued that the situation was very real to the participants. Quantitative data gathered during the procdure showed that 90% of the prisoner's conversations were about prison life and one prisoner asked for parole as opposed to withdrawal from the experiment. This suggests that the situation was real to the participants which gives the study a high degree of internal validity.
Role of dispostional influences - Fromm (1973) accused Zimbardo of exaggerating the power of the situation to influence behaviour, and minimising the role of personality factors. For example, only roughly a third of the guards behaved in a brutal way. Another third tried to ensure the rules were applied fairly and the rest sympathised with the prisoners and gave them privilleges. This is a limitation as it suggests that Zimbardo's conclusion may be over stated.
Lack of research support - Reicher and Haslam (2006) replicated the study in the UK and it was broadcasted on the BBC. The findings of their experiment (Guards were uncomfortable with their power and the prisoners eventually took control of the prison and harassed the guards) were very different to those of Zimbardo. Social Identity Theory was used to explain this outcome as the only the prisoners shared a social identity as a group (refused to accept the limits of their assigned roles). A limitation as it direct challenges Zimbardo's conclusions.