Virtual Torture

Computer-generated torture creates real-world ethical problems

| January 11, 2007

In the 1960s Dr. Stanley Milgram set off a firestorm of controversy with his scientifically illuminating but ethically questionable experiments on obedience. Milgram tricked volunteers into thinking they were punishing others (played by actors) with electrical shocks. The idea was to see how far the volunteers would go when prompted by an authority figure. Some of the participants in the tests were traumatized by what they had done, and more than 40 years later, the ethics of the tests still serve as fodder for heated debates.

Today, as Kerry Smith reports for BioEd Online, modern researchers are trying to sidestep many of the ethical quandaries involved in the experiments by recreating them in 'immersive virtual environments.' What these experiments have found is that torture doesn't need to be real for people to act like it is.

In a recently released study, Professor Mel Slater of University College London recreated Milgram's obedience experiments in a virtual reality. Slater had his subjects administer electrical shocks to a computer-generated woman and tested their reactions. Half of the participants could see an image of the woman, while the other half communicated with her only through text. All of the participants knew that the woman wasn't real, yet many of them became increasingly anxious as the experiment progressed, experiencing increased heart rates and sweaty palms. According to Smith, 'these measures are nearly impossible to fake,' indicating that participants were actually affected by their virtual actions. In the end, the participants who could see a computer-generated image of their victim were more likely to stop the tests than those who communicated solely through text.

Considering the results of the experiment, the wider implications of computer-generated torture become even more disturbing. Writing for the San Antonio Current, Aaron Delwiche calls 2006 'a banner year for torture,' citing recent political developments, horror movies, and the myriad 'torture manuals' available on the internet for the computer game 'The Sims.' On one website, the narrator describes her torture victims, saying, 'They scream & gnash their teeth, begging me, their cruel, heartless deity, to have mercy. I am laughing with much glee.' Delwiche believes that our gravest concerns should be with the torture atrocities taking place in the real world, like at Abu Ghraib, and our personal culpability in them. Nevertheless, his analysis of torture in the Sims virtual world -- 'We're not just watching the atrocities unfold on the screen when playing these games; we're actually directing the pain ourselves' -- raises new concerns in light of Slater's research.



As Slater's tests prove, just because the torture is virtual doesn't mean its implications aren't real. This tension between the real and virtual worlds has many people uneasy about the virtual torture experiments. Professor Slater insists that his experiment 'reopens the door to... an area of research that is otherwise not open to experimental study for ethical reasons.' But some question this conclusion. Although the subjects knew that the experiments were in virtual reality, their reactions to the experience were entirely real -- creating a new twist on an old ethical dilemma. As a post on the New Scientist's technology blog puts it: 'If the same distress was caused, surely the experiment is just as unethical?'

Go there >> Virtual Reality Shocker