Last Updated: 28 Sep 2016
Nodes feeding into this one should explicitly control confounding factors. Some studies have found levels of competition and frustration in a video game to be better predictors of aggression after playing the game.
Studies that don't control for factors such as the above aren't by definition flawed; however, some may take studies feeding into this node as more valid in their conclusions.
Craig A. Anderson, Akira Sakamoto, Douglas A. Gentile, Nobuko Ihori, Akiko Shibuya, Shintaro Yukawa, Mayumi Naito and Kumiko Kobayashi From page 1070, "Figure 1 displays the results of this path analysis. As expected, the gender of participant strongly predicted HVGV and physical aggressiveness. Boys played more violent video games and were more physically aggressive than girls. Furthermore, physical aggressiveness at time 1 was an extremely good predictor of physical aggressiveness at time 2, consistent with much previous research that shows that the best predictor of future aggression is history of past aggression. Of primary importance, however, is the finding that across 2 very different cultures HVGV predicts physical aggression 3 to 6 months later, even after controlling for previous aggressiveness and gender."
This study compared groups who played violent games and groups who played non-violent games for a twenty minute period in terms of their physiological reactions to violent footage (shootings, stabbings). From the abstract: "Participants who previously played a violent video game had lower heart rate and galvanic skin response while viewing filmed real violence, demonstrating a physiological desensitization to violence."
Data from 34 participants used. For physiological results, from page 536: "Our first major prediction was that P300 amplitudes to violent images would be smaller for violent game players than for nonviolent game players, and that this effect would not generalize to other negative images. A general linear model showed that the predicted interaction between the 3- level image type variable (neutral, negative, violent) and video game violence exposure scores (continuous variable) was significant, F(2, 64) D12.03, p< .0001 (GreenhouseGeisser adjusted; D.99). Separate regression analyses within each image type showed that, as expected, P300 amplitudes to violent images decreased as a function of increased violent video game exposure, t(32)D ¡4.66, p< .0001, b D ¡.20 (see Fig. 1A)." For aggression results from sound blast test, from page 537: "Finally, a separate regression analysis showed that the more time participants spent playing violent video games, the more they aggressed against their ostensible partner, t(32) D3.88, p< .001, bD.17, rD .57 (see Fig. 2B)."
From page 1682, under Main Analyses: "As expected, people who played violent video games expected more aggressive responses from the main characters in the stories than did people who played the nonviolent video games, F(1, 220) = 7.40, p < .007 (see Figure 3). People who played a violent video game were more likely to expect the main characters to say or do something aggressive, F(1, 220) = 8.14, p < .005, d = 0.38. For example, one person who played a violent video game expected the main character in the “car accident” story to “shoot or stab the other driver.” People who played a violent video game were more likely to expect the main characters to have aggressive thoughts and ideas, although the effect was not quite significant, F(1, 220) = 3.69, p < .06, d = 0.26. For example, one person who played a violent video game expected the main character in the “going to a restaurant” story to think about “setting the table cloth on fire.” People who played a violent video game also were more likely to expect the main characters to feel angry and aggressive, F(1, 220) = 6.17, p < .02, d = 0.33. For example, one person who played a violent video game expected the main character in the “persuading a friend” story to feel “very pissed off.” "
Last Updated: 28 Sep 2016
Teena Willoughby, Paul J. C. Adachi, and Marie Good
This study looked at self-assessed aggression and violent video-game playing of high school students at four points in time. Numerous variables were taken into consideration. Table on page 6 provides correlation data.
From the discussion section, page 11: "The current study is the first to demonstrate a relation between sustained violent video game play and trajectories of aggressive behavior over the entire span of high school. The first set of analyses revealed that adolescents who played violent video games across many years of high school also reported steeper increases in aggression over time compared to participants who reported less sustained play. This finding was significant even after controlling for several potentially important third variables (i.e., nonviolent video game play, overall video game play, gender, parental education, number of computers in the home, number of at-risk background factors, academic marks, depressive symptoms, delay of gratification, involvement in sports activities, peer deviance, friendship quality, parental relationship quality, parental control, and school culture), and expands on initial evidence of a 2-year longitudinal link between violent video games and aggression (e.g., M???ller, & Krahe???, 2009). Moreover, the standardized coefficients were in the low .20 range, suggesting a small effect, consistent with findings reported in past short-term longitudinal studies (see Anderson et al., 2010)."
The study feeding into this node looked at sports games in two categories: traditional sports games and counterparts with extra violence. The study members rated the levels of difficulty and frustration in the games. There was a statistically significant difference in these potentially confounding characteristics between the violent and non-violent games. However, I'm placing this node feeding into n2, "Comparison controls for confounding factors such as competition," since the differences were small and were used as covariates in the analysis. From page 734, under aggressive cognition: "A difference score was calculated for each participant by subtracting the average aggressive word reaction time from the control word reaction time. A positive score indicates that the participant identified aggressive words more quickly than control words; thus, larger scores indicate greater aggressive cognition accessibility. Violent game participants displayed higher levels of aggressive cognition accessibility than nonviolent game participants, Ms = 26.28 and 16.85, F(1, 107) = 6.78, p < .05, d = .50. Men were higher in aggressive cognition than women, Ms = 26.40 and 16.73, F(1, 107) = 7.12, p < .05, d = .52. The game by sex interaction was non-significant, F(1, 107) = 0.17, p > .05. None of the game rating covariates significantly predicted aggressive cognition accessibility, all Fs < 1."
This study used the same story stems as n22, "Violent Video Games and Hostile Expectations" Under the results section, page 5: "The first hypothesis of the current study set out to replicate past research indicating a general adverse effect of video game violence on aggressive behavior. Fig. 1 displays the mean number of aggressive responses for each video game. Consistent with the first hypothesis, a one-way ANOVA with contrast codes (Kirk, 1995) comparing the three violent video games to the three nonviolent video games revealed that the violent video games (i.e., Mortal Kombat, Return to Castle Wolfensten, and Doom 3) produced significantly more aggressive responses than nonviolent video games (i.e., Project Gotham Racing, Tetris World, Top Spin Tennis; t (161) = 2.48, p < .05). It is also noteworthy that the mean number of aggressive responses for the three nonviolent video games did not differ from each other (F (1, 79) = .67, p > .05), nor did the mean number of aggressive responses for the three violent video games (F (1, 82) = .60, p > .05)."
Youssef Hasan, Laurent Bègue, Michael Scharkow, Brad J. Bushman This study used both the noise blast test and the story stem exercise to gauge aggressive behavior and expectations of aggression. The diagram on page 226 provides a strong visual. From the abstract: "After gameplay, participants could blast a confederate with loud unpleasant noise through headphones (the aggression measure). As a potential causal mechanism, we measured hostile expectations. Participants read ambiguous story stems about potential interpersonal conflicts, and listed what they thought the main characters would do or say, think, and feel as the story continued. As expected, aggressive behavior and hostile expectations increased over days for violent game players, but not for nonviolent video game players, and the increase in aggressive behavior was partially due to hostile expectations."
Last Updated: 28 Sep 2016
Example of a filling-in missing letters task
Page 935 of http://www.researchgate.net/profile/Bruce_Bartholow/publication/237931511_Violent_and_nonviolent_video_games_produce_opposing_effects_on_aggressive_and_prosocial_outcomes/links/00b49532ba52eae566000000.pdf
"Participants are given a booklet containing 98 word fragments, each with one or more missing letters. Participants are given 5 min to finish as many items as possible. Completed words are then coded as nonviolent, violent, ambiguous, or non-words using an included coding guide (available from Craig A. Anderson), and the percentage of violent words (ambiguous words are counted ???) compared to the total number of words constitutes the aggressive cognition score."
The conclusions of the first two experiments in this study strike me as standing on shaky ground. The data on the upper right of 936 in table 2 appears to show that taking a break after playing a non-violent game causes an increase in aggression comparable to that of playing a violent game without a break. The third experiment has seemingly more solid results. Note that the word completion task in experiment three also looked at pro-social words. From the word completion task section on page 939: "There was a significant main effect for Game on both aggressive words, F(2, 108)=6.72, p < .01, and prosocial words, F(2,108)=3.19, p=.045. Planned contrasts found that violent game players chose significantly more aggressive words (M=16.9%) than nonviolent game players (M=13.3%), F(1,66)=7.38, p < .01. Additionally, violent game players used significantly more aggressive words than control participants (M=12.9%), F(1,74)=11.97, p < .001" From the story completion task section of 939: "There was a significant main effect for Game on aggressive items, F(2,108)= 6.57, p b.01, and prosocial items, F(2,108)=5.28, p < .01. Violent game players again chose significantly more aggressive story completions (M=19.7%) than nonviolent game players (M=12.9%), F(1,66)=10.3, pb.01. There was a marginally significant effect when comparing violent game players to control participants (M= 16.0%), F(1,74)= 3.62, p= .061; the nonviolent condition chose significantly fewer aggressive responses than control, F(1,76)=5.26, p=.025."
Last Updated: 28 Sep 2016
From bottom of page 141 and top of 142:
"A lexical decision task was used to assess the activation and cognitive fluency of aggressive concepts. The task entailed quick classification of letter strings as words or nonwords."
From the first page, 139:
"Typical experimental paradigms for testing cognitive fluency compare one or more violent video games with one or more nonviolent video games with respect to their fostering aggressive solutions to word fragment completion tasks (Carnagey & Anderson, 2005) or their speeding up aggressive concepts in word pronunciation tasks (Anderson & Carnagey, 2009; Anderson & Dill, 2000) or lexical decision tasks (Glock & Kneer, 2009). For purposes of experimental control, the accessibility of aggressive concepts is often defined by the difference between reactions to aggressive and control target words or stimuli, with the nonaggressive control stimuli typically being neutral or negative but nonaggressive."
Page 144: "The data show that the violent video game primed aggressive concepts, which is well known from the research literature and is successfully replicated in this study. Furthermore, positive concepts were primed by the violent video game as well. While the facilitation observed for aggressive test words by the violent video game is nothing new and quite expected, the facilitation seen with the positive words is new and may pose problems for theories that define violent video games as eliciting negatively valenced concepts, at least in the initial phase. The results are at odds with the assumption of the General Aggression Model and the desensitization hypothesis that violent video games typically elicit negative thoughts and emotions, but are in accord with the high popularity of violent video games." There's also a diagram on page 143 of the results.
Alessandro Gabbiadini, Paolo Riva, Luca Andrighetto, Chiara Volpato and Brad J. Bushman Bottom of page 454, top of page 455: "Consistent with many previous studies, violent game players were more aggressive (M = 7.84, SD = 1.55), than were nonviolent game players (M = 6.79, SD = 1.33), t(165) = 4.41, p < .001, d = 0.69."
This study examined two variables, violent v. non-violent game and online v. offline game. From the results section: "A 2 (type of video game: violent vs. neutral) x 2 (setting: online vs. offline) analysis of variance (ANOVA) on the amount of chili sauce (aggression measure) revealed a significant main effect of type of video game, F(1, 97) = 8.63, p = .004, ?p2 = .08. Participants who had played the violent video game were more aggressive (M = 16.12, SD = 15.30) than participants who had played the neutral video game (M = 9.06, SD = 7.65)." The diagram under the results section has error bars overlapping for the violent and non-violent offline games, but it appears they were still statistically different.
Two responses to this challenge: (1) I just modified the Topic Statement to say in its body that what it is actually claiming is, "The Preponderance of the Evidence Indicates Violence in Video Games Leads to Aggression."
I think it could very reasonably be challenged in its previous form on the basis that it is not clear the evidence is conclusive. However, all the evidence presented on this diagram so far is claimed to support this thesis, so barring other evidence being introduced, it seems established.
(2) The studies being referred to are the citations presented on the right in the full topic mode, nodes n20,n9, etc. I haven't previously authored statements on this topic, and it seems to me a lot of them could use more text in the body, but the intent was to organize the studies by the sub-phenomena they study, to decide point by point what is real, which makes sense as an organization. This challenge may be appropriately applied to the individual studies that have been posted.