tudes experiment. Journal of Perhave wished to argue? I would sugsonality and Social Psychology, 1974, gest, in answer to this, that guards 30, 468-476. in a real penitentiary could also be Rosenthal, R. Interpersonal expectasaid to be "playing a role" when they tions: Effects of the experimenter's first begin their jobs. Like the hypothesis. In R. Rosenthal & R. L. Rosnow (Eds.), Artifact in behavioral guards in the prison study, real research. New York: Academic Press, guards are given little or no training. 1969. Beyond what they have seen in films Schuck, J., & Pisor, K. Evaluating an or heard from the more experienced aggression experiment by the use of simulating subjects. Journal of Perguards, they really do not know what sonality and Social Psychology, 1974, to expect. Just like the guards in 29, 181-186. the prison study, they only know Zim'bardo, P. G. On the ethics of inthat they must be "tough" with the tervention in human psychological research: With special reference to the prisoners in order to maintain order. Stanford prison experiment. CogniIn what way, then, is the subsequent tion: International Journal of Cognibehavior of the guards in the study tive Psychology, 1973, 2, 243-256. any less "real" than the behavior of Zimbardo, P. G., Haney, C., Banks, novice guards in actual penitentiaW. C., & Jaffe, D. Stanford prison experiment. Stanford, Calif.: Philip ries? Certainly, a qualitative differG. Zimbardo, Inc., 1972. (Tape reence does not exist. Like real guards, cording) the prison study guards began slowly but, once disinhibited, soon freely agWILLIAM DEjONG gressed against the prisoners as a Stanford University matter of routine. I recognize that a difference of opinion on the interpretation of the Stanford Prison Experiment will perDemand Characteristics Are sist, in spite of the arguments I have Everywhere (Anyway) made. However, I have hoped to A Comment on the show that the findings of the prison Stanford Prison Experiment study should not...