Crouching amongst the wreckage of an apartment block I look through a shattered window and see an enemy running and trying to find a place to hide. I pick up my Intervention M-200, take a breath, get his head lined up in the cross hair and pull the trigger.
I have at various points in my life gone through phases of being, what you might call, a ‘gamer’. The scenario above was a frequent one that occurred when I used to play the previous Call of Duty game, Modern Warfare 2. I amassed an ‘on-line kill’ total of around 10,000. My ‘real life kill’ total is thankfully 0. So, it would seem that if you want to find out about me, monitoring my activities in the game would be not the most accurate way of finding information!
However, scientists have been using video games to look at real world psychology. In research published in ‘Cyberpsychology, Behaviour , and Social Networking’ virtual reality (VR) has been used to look at high-level social phenomena. The researchers, from Cardiff, outlined several advantages of using a virtual reality situation including, enhanced control, easy content modification and relative low cost.
Figure 1: A VR re-enactment of the Milford experiment
The study aimed to look at the reduced likelihood of helping bystanders whilst under varying time pressure during a task. This study is similar to the famous ‘Good Samaritan Study’ carried out by Darley and Baston, but places removes the focus on religion and instead looks at the virtual realm. Other studies that have crossed the VR-real life divide includes the infamous Milford experiment.
The participants were placed in the world of the popular game Half Life 2. This allowed the researchers more controll than other VR worlds such as SecondLife or World of Warcraft.
The study had two main parts. In experiment 1 the subjects were asked to complete a simple task (get to the exit of the 3D Maze), whilst (like the Good Samaritan study) the amount of time pressure they are placed under varies (24 participants were given time pressure, with 20 given less time pressure). Whilst completing this task the patients were given several opporunities to help virtual characters that requested assistance, both in the presence of ‘by standers’ and without.
Figure 2: The VR environment from experiment 2
In the second experiment, they looked more at the ‘bystander effect’ (the theory that when there are more bystanders there is a decreased individual responsibility to help as ‘someone else will do it’). To do this they increased the number of virtual characters in the maze and asked the participant to pretend as if they were real (although, surely a better way to have this effect would have been to have said they were being controlled by real people).
In both situations the maze was made up of 4 different sorts of rooms, an instruction room, 2 obsevation rooms, 14 assistance rooms and an exit room. In the observation room the participants could see one character helping another by smashing crates behind which they were stuck. In the assistance rooms the participants had the choice of whether they would help in a similar way. In both experiments half of the assistance rooms were populated with bystanders. But, experiment 2 had in each room as 19 opposed to the 4 in expiment 1. The bystanders “displayed various casual movements like turning their heads but ignored requests for assistance”.
The results interestingly were similar to those of their real world equivalents. Experiment 1 showed that increased time pressure decreases the help given to those in need (Figure 3), as in the Good Samaritan Study. Although the presence of bystanders had no significant effect in each time situation.
Figure 3: A graph showing that in a VR world increased time pressure decreases helpfulness
The lack of bystander effect in this experiment was attributed by the researchers to the VR bystanders not being convincing enough meaning that there was no “diffusion of responsibility ad fear of embarassment”. This was seens as the major drawback of the VR method.
However, in contrast to the findings of experiment 1, experiment 2 did show a significant difference between the bystander and non bystander rooms. This could be attributed to either of the two variables between the experiments, the number of bystanders or the participants being asked to pretend they were real.
The authors of the paper, Kozlov and Johanssen argue that VR is a useful tool in measuring factors that are difficult or impossible to control in the real, such as behaviour when escaping a burning building. However, I can’t help but feel that despite the success of the experiment a real world experiment, money no object, would be better at monitoring psychology as the lack of real world consequences is an issue for VR experimentation.
So, would I be a merciless killing machine if WW3 ever descends? No, I think most likely I would be sat quivering in a corner. But, the research does raise some very interesting questions about virtual behaviour.
To sum up my closing point here is a funny little video about killing your characters on the Sims, something which as a young teen thought was very funny (dont judge me!):
Kozlov MD, & Johansen MK (2010). Real behavior in virtual environments: psychology experiments in a simple virtual-reality paradigm using video games. Cyberpsychology, behavior and social networking, 13 (6), 711-4 PMID: 21142998
Darley, J., & Batson, C. (1973). “From Jerusalem to Jericho”: A study of situational and dispositional variables in helping behavior. Journal of Personality and Social Psychology, 27 (1), 100-108 DOI: 10.1037/h0034449
Slater M, Antley A, Davison A, Swapp D, Guger C, Barker C, Pistrang N, & Sanchez-Vives MV (2006). A virtual reprise of the Stanley Milgram obedience experiments. PloS one, 1 PMID: 17183667