This is one of the reasons I think people err by reading too much into accounts from the 'Stanford Prison Experiment'.
The students then (as now) weren't a sample of all people, but a particular kind of privileged young male, especially deferential and trusting towards professors, and especially willing to role-play with confidence that some other authority was managing the consequences.
The participants in the Stanford Prison Experiment weren't Stanford students. I'm not sure any of them were. They were selected by Zimbardo from responses to an ad in Palo Alto. They were selected for their normal-ness, mentally and physiologically.
You're right that they may not have been exclusively Stanford students. But they were recruited through city and campus newspapers, limited to male college students who were in Palo Alto for the last two weeks of August 1971, and had to report to an on-campus location to sign up.
I've seen definitive references that some were Stanford students, and my guess under those conditions would be that most were.
But even if most were not: a cohort of 24 male college students (23 white; 1 asian) in 1971 was not representative of society at large. Those in comfortable Palo Alto for the last 2 weeks of summer break, even less representative. And those who arrived on campus in response to a classified ad for paid volunteers "needed for psychological study of prison life", less representative again.
(As one of the followup studies footnoted in Wikipedia discovered, the mere presence of the "study of prison life" phrase in the recruitment ad changes what kind of people respond.)
This is a very interesting observation, while I don't doubt the observation about typical elite university students is true, I wonder if the inverse observation about the "general population" is enormously less true. This would be necessary in order to expect a different result from said experiment.
In my experience, the person who feels no desire to conform whatsoever is an extreme aberration. Far more frequent are people that pretend to be such and actually simply flip the expected behaviour from a conformist norm and behave in a different way but still conforming to some social group.
Sure, they're individuals, just like the other members of their subculture / social clique, etc.
Eliezer Yudkowsky, in his excellent essay "Lonely Dissent" (http://lesswrong.com/lw/mb/lonely_dissent/): Lonely dissent doesn't feel like going to school dressed in black. It feels like going to school wearing a clown suit.
The students then (as now) weren't a sample of all people, but a particular kind of privileged young male, especially deferential and trusting towards professors, and especially willing to role-play with confidence that some other authority was managing the consequences.