What was wrong with the Stanford Prison Experiments?

To illustrate issues of authority, trust, and social construction in my teaching I use two seminal US studies: the Milgram studies (read more here) and Zimbardo's Stanford Prison Experiment (here; 1972). These pieces of research are of their time, reflecting how we then saw our world. While using both studies, I have instinctively felt more drawn to the work of Stanley Milgram: it felt more universal. For me, the Stanford "SPE" - as it is known - was slightly tainted by two things: the pointlessness of the Vietnam war and my personal ick re the creepiness of Professor Zimbardo.

In a nutshell, the SPE was an experiment run by Professor Zimbardo from Stanford University, interested in understanding the psychological effects of imprisonment on 24 male Uni students; randomly allocated 50/50 as either guards or prisoners (Banuazizi & Movahedi, 1973; Haslam et al., 2019; Le Texier, 2019). The SPE was held in a mocked-up jail - of flimsy cardboard in parts - in the university basement, for a fortnight (Banuazizi & Movahedi, 1973; Hanley & Zimbardo, 1998; Zimbardo, 1973). But after only 6 days, Professor Zimbardo apparently pulled the pin due to overly authoritarian guard behaviour and prisoner distress (Banuazizi & Movahedi, 1973; Hanley & Zimbardo, 1998; Le Texier, 2019)... but maybe related more to pressure from his girlfriend (Eisner, 2024, Le Texier, 2024). Participants self-selected via an ad paying $15/day for a "psychological study of prison life" (Le Texier, 2024, p. 58). 

This experiment made Professor Zimbardo a household name due to a riot at Attica prison occurring three weeks later, allowing him to emerge as a "reasoned critic" on prison culture at a time when it was front and centre in the media (Le Texier, 2024). The Professor was authoritative, articulate, and confident, and US TV loved him (Le Texier, 2019).

So, earlier this year when I stumbled across the work of Thibault Le Texier, a French academic at the Sorbonne who wrote a book about the SPE in 2018 (trans. from the French, Le Texier, 2024), I was interested to read on. Le Texier had gone to Stanford, had read and viewed the original files in the archives, had interviewed participants; and had made some interesting discoveries. Then Disney made a documentary about it (Eisner, 2024) and I watched that too. 

Le Texier found evidence of methodological improvisations, ethical transgressions, and a preparedness for the media (2019, 2024). The Professor as "Prison Superintendent" was interfering on the fly; having coached the guards via crib sheets and a training day; despite promising that participants could voluntarily exit the experiment, not allowing them to do so (Eisner, 2024; Le Texier, 2019, 2024). On day two of the SPE, the Professor began sending out press releases, inviting the media to visit the mock prison (Le Texier, 2019).

SPE participants were likely to have been operating on internal mental models of guard and prisoner behaviour, rather than solely responding to their situation: two years after the SPE, 81% students surveyed thought the participants could accurately guess the experiment’s aims, and 90% predicted that guards would be oppressive and hostile (Banuazizi & Movahedi, 1973; Le Texier, 2019). And 11 of the 17 'rules' used by guards came from a student experiment run twelve weeks earlier (Le Texier, 2019). The 'random' assignment of volunteers as either prisoners or guards was very well-organised ahead of time. Those who were assigned to be prisoners were picked up by real police officers on the day, and processed - fingerprinted, photographed etc - at a real police station before being transported to the Stanford basement (Eisner, 2024; Le Texier, 2024). 

Some students - both prisoners and guards - acted a part to aid the study (Eisner, 2024; Le Texier, 2019), perhaps in a similar way to Elton Mayo's studies at GE's Hawthorne plant (read more here; Perera, 2023). Further, the guards were apparently told that tough behaviour was necessary for study outcomes, in order to illustrate US penal system flaws. Guards gained the impression they were research 'assistants', pulling them into the observation team (Richter, 2020; Le Texier, 2024). 

The collection of data was surprisingly incomplete, with under 15% of the 6 days of the experiment filmed or sound recorded. Of the recordings made, it appeared that more "cinematic" events were recorded (Le Texier, 2019, 2024, p. 105), perhaps suggesting set-ups. The press releases written by Zimbardo outlined the experiment's "conclusions" about prisons being toxic from day two (Le Texier, 2019, 2024, p. 104). 

The big one for me though is that both the guards and the prisoners were psychologically damaged by the actions inflicted on each other, in addition to the lack of ability to leave the experiment while it was underway (Le Texier, 2019, 2024), and by reflecting back on the experiment in the years following (Eisner, 2024). Self-selection bias may have been a factor as a "study of prison life" may have attracted more authoritarian participants (Le Texier, 2024, p. 58); the low ecological validity of the prison may have meant that participants could not be fully immersed in their roles (Banuazizi & Movahedi, 1973); and the forced confinement "in the hole" would have increased the sense of powerlessness felt, even if the Uni basement prison was not particularly robust (Le Texier, 2024, p. 72). 

Perhaps instead of assuming that prisoners will become servile and guards will become militant, we should remember that even in this flawed experiment, two thirds of the guards were not cruel. Our takeaway could be that most of us are unlikely to become sadistic from just being allocated a role (Haslam & Reichner, 2008; Le Texier, 2024). 

Sam

References:

Banuazizi, A., & Movahedi, S. (1973). Interpersonal Dynamics in a Simulated Prison: A methodological analysis. American Psychologist, 30(2), 152-160. https://doi.org/10.1037/h0076835

Eisner, J. (director) (2024). The Stanford Prison Experiment: Unlocking the truth. National Geographic | Disney Plus.

Haney, C., & Zimbardo, P. (1998). The past and future of US prison policy: Twenty-five years after the Stanford Prison Experiment. American Psychologist, 53(7), 709-724. https://doi.org/10.1037/0003-066X.53.7.709

Haslam, A. & Reicher, S. (2008). The BBC Prison Study [home]. http://www.bbcprisonstudy.org/index.php

Haslam, S. A., Reicher, S. D., & Van Bavel, J. J. (2019). Rethinking the nature of cruelty: The role of identity leadership in the Stanford Prison Experiment. American Psychologist, 74(7), 809-822. http://dx.doi.org/10.1037/amp0000443

Le Texier, T. (2019). Debunking the Stanford Prison Experiment. American Psychologist, 74(7), 823-839. http://dx.doi.org/10.1037/amp0000401

Le Texier, T. (2024). Investigating the Stanford Prison Experiment: History of a Lie. Springer.

Perera, A. (2023). Hawthorne effect: Definition, how it works, and how to avoid it. Simply Psychology. https://www.simplypsychology.org/hawthorne-effect.html

Reicher, S. D., Van Bavel, J. J., & Haslam, S. A. (2020). Debate around leadership in the Stanford Prison Experiment: Reply to Zimbardo and Haney (2020) and Chan et al.(2020). American Psychologist, 75(3), 406–407. http://dx.doi.org/10.1037/amp0000627

Zimbardo, P. G. (1972). Pathology of Imprisonment. Society, 9(6), 4-8. https://doi.org/10.1007/BF02701755