The great Stanford Prison Experiment was eventually shut down after only six days - it was originally planned to last two weeks. A female research assistant performed a scheduled interview of the participants and was appalled at the conditions. She convinced ZIM that even he, acting as superintendent of the prison, had become an intimate participant in the simulation and his behavior was degrading like the sadistic guards.
Even though it ended early, ZIM claimed the experiment proved his hypothesis. Extreme situations such as incarceration can turn otherwise normal and healthy people into sadistic tormenters or pliable cows. These people aren't inherently evil; it is the circumstances which can push a person to commit evil acts.
In order to test this idea, ZIM had to create an elaborate fictional world. The subjects chosen to be prisoners were picked up at their homes by actual police officers, brought down to the station for an accurate booking for armed robbery and even deloused. The guards were told to refer to the prisoners only by the numbers stitched on their muslin smocks. The prisoners were forced to wear panty hose on their heads to de-individualize them by removing any details that hair color or length might lend. This kind of illusion was important. Any sense that what the prisoners were experiencing was contrived and their reactions would have been disingenuous. The world ZIM created was intensely real.
As computer technology improves and gives developers the chance to create more vivid environments, so does the ability for game designers to fabricate more compelling and lifelike worlds. After running around in Gears of War, it's difficult to readjust to the real world of my crappy Brooklyn apartment. But graphics are only part of the experience. Characters, story and, most of all, player choices are what makes a game feel real. It's all in the details, as ZIM knew.
Given where we are on the technology curve, can a game simulate reality to the point where it is an accurate test of human behavior? Or more importantly, if given a moral choice in a game, do our actions portray how we would react if given that choice in our daily lives? It's easy to say, after reading about the Stanford Prison experiment, "Bah, I'd never do that." And that assumption could be correct. Only about a third of the guards in the simulation actually exhibited sadistic tendencies. The results of the Milgram experiments, however, cannot be denied so easily.
Stanley Milgram, a Yale University psychologist and high school classmate of Dr. Zimbardo, conducted several similar experiments from 1961 to 1964. The subjects were told they were participating in a study testing the effect of electric shocks on learning. A subject and the victim (who portrayed another subject but was actually a confederate) were given slips of paper to determine which would be the teacher or the student. In fact, both slips of paper read "teacher" but the confederate always acted as if he received the student slip.