If I understand it correctly, the researchers imply that people's actions in VR simulations (vs text-based questions) would be closer to what people would do IRL, because "This change in decisions reflected in the autonomic arousal of participants, with dilemmas in virtual reality being perceived more emotionally arousing than the ones in text":
In the light of this model, we hypothesize that in VR participants could have been more sensitive to outcomes because they witnessed distressing consequences (gory deaths of virtual humans) of their actions and emotions motivated them to act in order to minimize the distress by choosing the best of two emotionally aversive options in which either one or numerous (2 or 5) deaths occur.
While there was no easy way to ethically test what experimental subjects would do in a perceived real-life trolley problem, I suspect that the conclusions drawn by the authors are the opposite of what is actually happening. If you look at the VR simulations, they do not appear as real people. In fact, they feel less real to me than those in text-based trolley problems. Indeed, GTA players think nothing of running over a (VR) pedestrian for minimal gain, while almost no one would do it IRL. Thus people go for a utilitarian choice when the situation appears to be less realistic.
Another interesting effect this study demonstrates is selective priming: subjects give more utilitarian answers on text-based dilemmas after running through a VR version, but not vice versa. The authors hypothesize that "gory deaths of virtual humans" make the utilitarian choice more convincing. I suspect that another possibility might be that people treat textual humans more like virtual humans after a VR experiment, but not vice versa, because visual stimuli are stronger than text-based and the image that "virtual humans are not real" persists through the text-based dilemmas.
I suspect that the conclusions drawn by the authors are the opposite of what actually happening
Yes. Real human lives, which are indicated in a text based question, are sacred quantities. Virtual lives – especially if perceived as avatars – are not sacred quantities. There might be a cost ,depending on the context (it's 'mean') , but measurable and comparable.
To what extent were participants encouraged to answer the question as though there were real people, and not virtual people or avatars? Even with extensive coaching, it might be difficult to suspend the conditioned insensitivity we need for playing video games. (Did they look at the effect, if any, of previous video game experience?)
THe simulations look like they might have been developed using the tech from Half-Life 2, but with terrible quality animations. If the simulations were highly immersive, I might freak out because zombies. They also look less realistic than sequences seen in a number of popular violent video games (some of which offer considerable applications to apply utilitarian or unutilitarian choices.
Telling people with no exp. on violent video games to play Mass Effect all the way through, and record all their choices, and hesitations might be interesting for the cost.
A new study indicates that people become more utilitarian (save more lives) when viewing a moral dilemma in a virtual reality situation, as compared to reading the same situation in text.
Abstract.
Full paper
Video of simulations