What is one of the key differences between video games and television? Interactivity. I’ve taken that fact as granted and ignored it while reading research articles. In any case, I went on a research binge to find all relevant studies that compared the effects of violent video games and television violence. The results reflect the kind of hindsight in psychology video game research, five studies. I have managed to get four of them, one was a thesis dissertation, so it was not possible to get my hands on.
To summarize why psychologists are concerned with video games is that unlike television, the audience or player is actively involved in violent behaviours. The very decision to aggress another person, animal or object primes or activitates aggressive thoughts, feelings and behaviours. To boot, such aggressive behaviours are rewarded by numerous stimuli, such as player survival, points, special effects, and such. Do this hundreds of time in a short span of time and, theoretically, you could have someone who might be easily provoked to be aggressive (no not violent, that’s a different story). Whereas people watching violent television are simply learning as an observer, but don’t get any practice in doing it. In sum, the one difference between video games and television is that players are active learners which is more powerful than passive learning.
So I decided to read all four studies, two studies were published in the 1980s and 1 study published in 2004 and another that’s recently published. Their conclusions conflicted with each other, but there are circumstances to these conflicting conclusions.
The studies share a fundamental procedure where pairs of participants are randomly assigned to either a violent or a non-violent condition. One of the pairs is assigned to play the violent or non-violent video game, while the other pair is assigned to watch that game.
Differences in their procedures:
Cooper and Mackie (1986) had 84 4th and 5th grade children, had them played either Missile Command (high violent), Pac-Man (low violent) or a maze-solving paper game (control). Play session is 8 minutes and the watching participants were in verbal contact with the playing participants, but could not have physical contact with the game or the player. Not a good idea, since they are still participating in the game.
Silvern and Williamson (1987) had 28 children (age range: 4-6 years old, average about 5 years old). They played either Space Invaders (video game) or watched Road Runner (television). Play time is 10 minutes and television time is 6 minutes. The group of children who had either played video games or watched television first, were then observed for aggression during free play for 10 minutes, then a day later, they switched to the other condition and then observed again for aggression during free play for 10 minutes. A good method in getting the effects from all participants, but it’s still small.
Tamborini et al. (2004) had 182 college students (age range:18-30 average age is 20.57). They were grouped into either in violent (Duke Nukem 3D) or non-violent (CoolBoarders 3) and the playing participants are either playing the violent video game game with Virtual Reality headsets and a toy gun or standard PC with a joystick. The watching participants are in a different room with no contact with the playing participant, watching on a television screen what is being played by the playing participant or if they are in the non-violent condition, then they watched a pre-recorded game session, but were told that they were live-feeds. Playtime is divided into 3 sessions of 5 minutes, interspersed by an administration of a questionnaire. Everthing seems good, except for the short play time sessions.
Polman et al. (2008) had 57 5th and 6th grade children. Most of them are in 5th grade, so the average age is around 10-11 years old. The participants come from the Netherlands. They had them played either Tekken 3 (violent) or Crash Bandicoot (non-violent) for 15 minutes. The watching participants watched the game from a seperate room with no contact with the playing participant. If Polman et al. had more participants that would have been great, but getting parental consent might affect the composition and characterisitics of the sample as some parents refused their child’s participation in the study. Who knows if those non-participating children would have changed the results of the study.
Differences in measuring aggression and other measures:
Cooper and Mackie (1986) had one group answer hypothetical situations of bad behaviour and good behaviour by another child, what sort of punishment and reward, respectively and how long would the punishment or reward would it be. The authors noted that kind of aggression measurement is controversial whether it does relate to aggression at all, reminds me of the noise-blast paradigm. The other group were directed to have some free play for 8 minutes with the toys provided by the researchers and were observed with which toy they played with. I’m not sure if playing a certain type of toy would mean anything. But I should ask my professors about that. Afterwards, the group switched to the other measures.
Silvern and Williamson (1987) used observational researcher reports from three free play sessions, the first is used as a baseline and is before the experiment. The second and third sessions are right after the experiment. Please refer to differences in procedures. Free play sessions last 10 minutes. The observational reports and toys used were from previous studies, this included a bobo doll.
Tamborini et al. (2004) had administered the Buss-Perry Aggression questionnaire for trait aggression and the prior video game use questionnaire where they were asked to name their 5 favorites video games and rated them on their level of violence and often they played them. They were administered three weeks prior to the experiment and were used as control variables. Druing the experiment, they administered the thought-lisitng procedure, which basically asked participants to write anything that comes to their minds for 4 minutes. Researchers would analyze for any hostile thoughts. Another measure administered during the experiment is the telepresence questionnaire, which basically asked how immersive or realistic the game environment is.
Polman et al. (2008) administered questionnaires after the experiment at different times, immediately, 1 hour or more than 1 hour after the video game experiment, one is about their gaming habits and another on aggressive behaviours. In particular, the children were asked to name another child in the study who displayed acts of physical, verbal or relational aggression. They were also asked if the intentions were for play or being mean.
Results on aggression from the four studies:
Cooper and Mackie (1986): the video games were significantly different in terms of violence, but they were rated under the mid-point scale. Quite obvious, since Missile Command isn’t so graphically violent. There was only one significant effect for video game exposure in how long they would punish/reward behaviours, that is those that played video games had longer punishment/reward than non-players and basically the aggressive nature of the video game had nothing to do with punishing or rewarding bad or good behaviours. As for the effect of playing Missile Command had no effect on boys’ preference on the types of toys during free play, but there was an effect for girls in that they played the aggressive-typed toy (a Shogun samurai that spits fists and darts) as often as the boys.
Silvern and Williamson (1987): There were significant differences between televison/video game and baseline. There was, however, no difference found between televison and video games in aggression during free play.
Tamborini et al. (2004): The non-violent watching condition was significantly lower in hostile thoughts than the three other conditions. However, there were no significant differences between the violent-VR, violent-standard console or violent-observation conditions. Looking at the relationship between telepresence and state hostility, there was no significance, even they added all the relevant factors that may influence the relation between video games and state hostility.
Polman et al. (2008): They’ve used Kruskall-Wallis analyses to get their results, analyzed boys and girls seperately and they cautioned that any results (significant or not) should be taken with caution because of small sample sizes, so I’m not sure what to make of it. Anyways, boys who played violent video game were significantly rated to be more aggressive than those who watched a violent video game. However, they were not significantly different from those who played a non-violent video game and the non-violent video game participants were not significantly different from those who watched a violent video game. For girls, there were no significant differences in either violent or non-violent video game nor whether they played or watched. Looking at effects across time, they found that boys who immediately finished the experiment were rated to be more significantly aggressive than those who rated after 1 hour or more than 1 hour of the experiment.
Where to start with my criticisms? Well let’s break the studies from the 1980s. Pac-Man, Missile Command and Space Invaders are far different from today’s video games because they aren’t as graphic, sophisticated or violent to today’s standard. Silvern and Williamson would get some praise in comparing an actual television show and a video game, but missed on several factors, such as pace, action, frustration, and narrative. Nevertheless, I might find it difficult to find a television show or movie that is equivalent to an FPS game, and I would not consider using the movie based on Doom. In effect, some might find this as evidence against the adoption of a universal media rating system for video games, television and movies.
Looking at Tamborini et al. (2004), I was disappointed that they made such a blunder of having participants play 5 minutes of video games. Although, they did it so because they would not want one questionnaire influence participants’ answers on another questionnaire. Furthermore, I’m not sure why they would want a game like Duke Nukem 3D, it makes me think that the study was done years before it was published. Anyways, at least it tells us that in the short term, video games and television are not so different and the effects might show if we play longer and that violent content increases our aggression.
Looking at Polman et al. (2008), I was also disappointed of the small sample size and the fact they need to use statistical techniques beyond my understanding to make sense of the data they have. Second, the children were watching a video game which is quite a different experience from watching television despite trying to improve on flaws from Silvern and Williamson. Third, they did not include whether watching a non-violent game might have an effect. However, I am impressed that they found significant effects between playing violent video games and watching violent video games, but it should be taken with grains of salt.
Besides directly comparing video games and television, some might compare the effect sizes of meta-analytical studies. However, what was found was that video game effects seem smaller than television violence (r = .26 vs. r = .31, respectively), perhaps this difference is due to the smaller studies on video games (see Polman et al., 2008).
In summary, the few comparative studies on the effects between watching violence and playing violence showed that more studies are needed. We also have to keep in mind of the characteristic differences between video games and television in order to determine which are relevant to affect individuals’ aggression. Therefore, it’s going to take a while to get a clear answer on whether interactive violent video games has a stronger effect than passively watching violence on a screen.
Cooper, J., & Mackie, D. (1986). Video games and aggression in children. Journal of Applied Social Psychology, 16(8), 726-744.
Polman, H., de Castro, B. O., & van Aken, M. A. G. (2008). Experimental study of the differential effects of playing versus watching violent video games on children’s aggressive behavior. Aggressive Behavior, 34(3), 256-264.
Silvern, S. B., & Williamson, P. A. (1987). The effects of video game play on young children’s aggression, fantasy, and prosocial behavior. Journal of Applied Developmental Psychology, 8(4), 453-462.
Tamborini, R., Eastin, M. S., Skalski, P., Lachlan, K., Fediuk, T. A., & Brady, R. (2004). Violent virtual video games and hostile thoughts. Journal of Broadcasting & Electronic Media, 48(3), 335-357.