Follow me on Twitter

Blog archive

We Participate In:

ABA Journal Blawg 100!

Subscribe to The Jury Room via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.


Your online avatar and your real-world behavior

Monday, February 17, 2014
posted by Douglas Keene

online avatarLast fall we wrote about how having a dark-skinned avatar in an immersive virtual reality experience can reduce your implicit bias against dark-skinned people. Now Illinois researchers show us that the avatar assigned in online gaming also influences behavior. How? If you are assigned to be a hero, you do good. If you are assigned to be a villain, you do not do good. Okay. That makes sense in an online role-playing game–-but the point of the research is that the online role-play (as hero, villain, or neutral geometric figure) made a difference in the real world after the online gaming ended. 

The researchers did two separate experiments:

Experiment 1: 194 undergraduates (95 male, 99 female, average age 20.3 years) were told they would be involved in two separate studies: a game usability test and a blind tasting test. They were randomly assigned an avatar (a hero–Superman, a villain–Voldemort, and a neutral avatar–a simple circular shape). After they battled their opponents for 5 minutes online, they were asked how much they identified with their avatar on a 7-point scale ranging from 1 (not at all) to 7 (very much). Then they were told they were moving on to the second study–the blind taste test. They were given tastes of both chocolate and chili sauce and then asked to choose one or the other to give a a (fictitious) future participant to consume. They were told the future participant would eat all of the food provided and the participants could pour as much as they wanted into a container for the future participant.

While the conscious level of identification with the avatar made no difference, participants with heroic/Superman avatars gave more chocolate to the fictitious future participant than did either those with villainous/Voldemort or neutral (circular) avatars. Participants with villainous/Voldemort avatars gave more chili sauce than did participants who had heroic/Superman avatars or neutral/circular avatars.

Experiment 2: 125 undergraduates participated (44 male, 81 female, average age 19.4 years). The design was the same except the researchers added a condition to test whether role play assignment (as either hero or villain) was more powerful in real world conditions than “common behavioral priming or perspective taking”. The researchers also simplified the experiment by dropping the neutral/circular avatar and focusing on how much chili sauce was given to the fictitious future participant. To clarify–some participants played the online game with either the heroic or villainous avatar while others were asked to be an observer of the online game for 5 minutes but “take the perspective of” either the hero or villain avatar/player.

Again, villains poured more chili sauce for the future participant than did the heroes. Those playing heroes poured less chili sauce than those observing and taking the perspective of heroes. Similarly, those playing villains gave more chili sauce than those observing and taking the perspective of the villain.

In short, say the researchers, “acting as a hero or villain” means you are more likely to repeat heroic or villainous behavior than you would if you are simply asked to “think of yourself as being” either the hero or the villain. They encourage game developers to design more heroic avatars to encourage more real-world prosocial behavior.

From a litigation advocacy perspective, you cannot ask people to actually “be” pro-Plaintiff/Prosecution or pro-Defense for a few minutes and then send them off to deliberate and expect a positive outcome for your case. All you can do is ask them to have empathy, or put themselves in the position of, your client. And, this research would say that isn’t nearly as powerful.

We would beg to differ. The researchers are talking about the impact of online gaming on the quantity of chili sauce or chocolate left for someone else to eat. It’s a little different from the real life and critically important issues often presented to jurors. We certainly don’t advocate asking jurors to take on the role of hero or villain overtly.

What we do advocate doing is asking jurors to take on the hero role (or to not take it on) without actually asking overtly.

From a Civil perspective: We would use the “be the very best you you can be” strategy. Obviously it would be a little differently framed for these two roles but the idea is to create empathy for either the Plaintiff or the Defendant. This would be akin to the request for the juror to take on a “hero” role. They are being asked to rise above petty bias and consider the situation from the position of the individual person involved. It is, in some ways, a “but for the grace of God, there go I” sort of strategy that encourages identification with the actual person or with the situational/genetic influences that may have driven the Defendant’s bad behavior.

From a Criminal perspective: Consider increasing the urge to punish the Defendant by focusing on the egregious nature of the crime. Create a virtual environment in which the jurors imagine that this is a threat to society that they can reduce. By focusing on the socially inappropriate behavior in the fact pattern, they become more inured to the aggressive punishment that is being sought. Of course, this is typically how a prosecutor would handle criminal cases, but the reason for doing so is endorsed by this research. Criminal defense strategy would be to focus on externalities, like the role that the defendant having been treated ruthlessly would affect him, or the way in which the conduct of anyone ‘primed’ him or her to act improperly. It is a way to make people identify with their own role in the trial, and orient them toward one frame of reference or another.

Yoon G, & Vargas PT (2014). Know Thy Avatar: The Unintended Effect of Virtual-Self Representation on Behavior. Psychological Science PMID: 24501111


%d bloggers like this: