Research Presentation

AGGRESSIVE BEHAVIOR Volume 38, pages 281–287 (2012)

Effects of Prosocial, Neutral, and Violent Video Games on Children’s Helpful and Hurtful Behaviors Muniba Saleem∗, Craig A. Anderson, and Douglas A. Gentile

Department of Psychology, Center for the Study of Violence, Iowa State University, Ames, Iowa

: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :

Recent research reveals that playing prosocial video games increases prosocial cognitions, positive affect, and helpful behaviors [Gentile et al., 2009; Greitemeyer and Osswald, 2009, 2010, 2011]. These results are consistent with the social-cognitive models of social behavior such as the general learning model [Buckley and Anderson, 2006]. However, no experimental studies have examined such effects on children. Previous research on violent video games suggests that short-term effects of video games are largely based on priming of existing behavioral scripts. Thus, it is unclear whether younger children will show similar effects. This research had 9–14 years olds play a prosocial, neutral, or violent video game, and assessed helpful and hurtful behaviors simultaneously through a new tangram measure. Prosocial games increased helpful and decreased hurtful behavior, whereas violent games had the opposite effects. Aggr. Behav. 38:281–287, 2012. C© 2012 Wiley Periodicals, Inc.

: : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : :

Keywords: prosocial media; media violence; video games; aggression; helping

INTRODUCTION

Effects of Prosocial, Neutral, and Violent Video Games on Helpful and Hurtful Behaviors

A major development in mass media over the last 25 years has been the advent and rapid growth of the video game industry. From the earliest arcade- based console games, video games have been imme- diately and immensely popular, particularly among young people. Additionally, the introduction of video games to the home market only served to further ele- vate their prevalence [Gentile, 2009]. Because of their widespread popularity, social scientists, parents, and politicians have been concerned with the potential ef- fects of video games, focusing particularly on games with violent content and their harmful effects on chil- dren. This is evident in the recent Schwarzenegger vs. Entertainment Merchants Association Supreme Court case that debated state regulation of video game sales to minors.

Although there is an extensive base of scientific lit- erature on the negative effects of violent video games [see Anderson et al., 2010 for a metaanalysis], re- search on prosocial games is much more limited. Few empirical studies have tested the effect of prosocial video game content on helpful and hurtful behaviors and results from these studies suggest that prosocial content in games can in fact increase prosocial be-

havior in the short and long terms [e.g., Gentile et al., 2009; Greitemeyer and Osswald, 2010, 2009]. Additionally, recent studies have started to explore the causal mechanisms responsible for the effects of prosocial video games on helpful and hurtful behav- iors. Results from some studies suggest that these ef- fects may be due to changes in cognitive beliefs [e.g., Gentile et al., 2009; Greitemeyer and Osswald, 2010, 2011], whereas other studies suggest these effects are due to changes in affect [Greitemeyer and Osswald, 2010; Saleem et al., 2012]. Overall, both types of ef- fects are consistent with social-cognitive models of so- cial behavior, including the General Leaning Model [GLM; Buckley and Anderson, 2006; Maier and Gen- tile, 2012].

Briefly, social-cognitive learning theories (e.g., GLM) propose that input variables (personal and situational) affect a person’s internal states (cogni- tion, affect, and arousal) and ultimately guide the person’s learning (through multiple mechanisms) and behavioral responses [Buckley and Anderson, 2006;

∗Correspondence to: Muniba Saleem, Department of Psychology, Iowa State University, W112 Lagomarcino Hall, Ames, IA 50010. E-mail: [email protected]

Received 18 July 2011; Accepted 1 February 2012

Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/ab.21428

C© 2012 Wiley Periodicals, Inc.

282 Saleem et al.

Bushman and Huesmann, 2006; Gentile and Gentile, 2008; Huesmann and Kirwil, 2007; Maier and Gen- tile, 2012; Swing et al., 2008]. Specific to the present study, these theories suggest that playing a prosocial video game should prime knowledge structures re- lated to prosocial actions, including associated cogni- tions, feelings, and physiological arousal. Of course, the efficacy of priming prosocial behavioral scripts depends on the existence of such scripts and chronic accessibility. For example, Bushman and Huesmann (2006) showed that brief experimental manipulations of media violence tend to produce somewhat larger short-term effects on older participants (mostly col- lege students) than on younger ones. This difference in short-term effects occurs because short-term ef- fects of media exposure are primarily attributed to activation of existing knowledge structures and chil- dren have less developed knowledge structures and fewer existing encoded cognitions based on their lim- ited experience. Although previous work has tested the short-term effect of violent content on children, the short-term effect of prosocial content on children remains unexplored.

In addition to short-term effects the social-cognitive learning theory predicts that repeated practice with prosocial (or antisocial) behavioral scripts can yield several long-term effects, including the develop- ment of and changes in precognitive and cogni- tive constructs (perception and expectation schemata, beliefs, scripts), cognitive-emotional constructs (at- titudes and stereotypes) and affective traits (condi- tioned emotional responses, empathy, trait hostility). Indeed, recent longitudinal studies (some as short as 3 months, others as long as 30 months) have found that children and adolescents who play a lot of vio- lent video games become more aggressive over time, even after controlling for earlier aggressiveness and other theoretically relevant variables [Anderson et al., 2007, 2008; Hopf et al., 2008; Moller and Krahe, 2009; Wallenius and Punamaki, 2008]. Aggressive be- haviors in these studies included getting into fights and delinquency. In terms of prosocial video game ef- fects, the only published longitudinal study found that prosocial video game exposure significantly predicted prosocial behavior 4–5 months later, even after statis- tically controlling for other relevant variables [Gentile et al., 2009].

In addition to situation variables, the social- cognitive learning theories suggest that person vari- ables (e.g., trait aggression) may also directly or in- directly influence prosocial and antisocial outcomes. For example, individual differences in aggressiveness might interact with the presence of violent stim- uli in the elicitation of aggression-related thoughts,

emotional states, action tendencies, and behavioral responses. More specifically, high trait aggressive in- dividuals might manifest all of these responses to aggressive cues more than individuals low on trait aggression, because they have more extensive ag- gressive cognitive-associative networks. Indeed, Bush- man (1995) found that media violence was more likely to evoke aggressive affect and behavior in high rather than low trait aggressive individuals. Similarly, Gentile et al. (2009) found trait aggression to be pos- itively related to hurting and negatively related to helping behavior. These findings are important in un- derstanding the complicated interactions between ha- bitual personality variables (e.g., trait aggression) and situational variables (e.g., media violence exposure). Thus, in the present study a measure of trait aggres- sion was included to better understand the combined influence of person and situation variables on helpful and hurtful behaviors.

In sum, previous research finds support for the short- and long-term effects of violent video games on helpful and hurtful behaviors using samples of children as well as adults [see Anderson et al., 2010 for a metaanalytic review). Similarly, prior re- search on prosocial video games has tested short- term (experimental) effects on adults, and long-term (cross-sectional, longitudinal) effects on children and adolescents [Gentile et al., 2009; Greitemeyer and Osswald, 2009, 2011]. However, one gap concerns whether brief exposure to prosocial games (relative to violent and neutral games) increases helpful and de- creases harmful behavior among children. This ques- tion is important considering short-term effects of media exposure are usually attributed to the priming of existing well-encoded scripts, schemas, and beliefs, cognitive structures that are not as well developed in children as in adults. Thus, the main focus of the cur- rent research was to examine the effects of prosocial, neutral, and violent video games on helpful and hurt- ful behaviors using a sample of 9–14 years olds. A secondary focus was to further validate the help/hurt tangram measure [Gentile et al., 2009, Study 3), which simultaneously assesses helpful and hurtful behaviors, with a different age group.

METHODS

Participants

Participants were recruited for this study by an ad- vertisement in the local paper and by contacting inter- ested parents compiled from previous studies. Each child participant (104 males, 87 females) with ages ranging from 9 to 14 (M = 11.4) completed the study individually and received 20 dollars. The study was

Aggr. Behav.

Helping and Hurting 283

approved by the Iowa State University’s Institutional Review Board and all participants were treated in ac- cordance with APA ethical guidelines.

Materials and Measures

Trait aggression. The 9-item physical aggres- sion subscale of the aggression questionnaire was ad- ministered prior to the experimental manipulation to assess trait aggression [Buss and Perry, 1992]. The Buss–Perry aggression questionnaire has been suc- cessfully used to assess trait aggression with a range of age groups including elementary school children [e.g., Reynes and Lorant, 2001, 2003, 2004; Walters et al., 2010; Zhen et al., 2011]. Participants rated their agreement with statements on a 5-point scale (1 = “Extremely uncharacteristic of me,” 5 = “Extremely characteristic of me”), alpha = .81.

Video games. Four E-rated video games were used: two violent (Ty2, n = 31, Crash Twin sanity, n = 32); and two neutral (Pure Pinball, n = 30, Su- per Monkey Ball Deluxe, n = 30). One E-10 game was used for the prosocial category (Chibi Robo, n = 671). Ty2 and Crash Twin sanity are action adventure games in which the goal is to complete various stages by defeating the enemies and bosses, while overcom- ing any obstacles on the way. The goal of Pinball is to keep the ball on the table using the left and right trig- gers. Super Monkey ball requires the player to guide a monkey through various puzzles toward the goal within the time limit. Chibi Robo lets the player con- trol a robot whose job is to make its family happy by cleaning up, helping them out in their chores, and everyday tasks. As the player cleans up throughout the house, they earn Happy Points that improve their robot’s ranking. The player can do several things to get happy points (e.g., picking up trash and throwing it in a trashcan, scrubbing stain marks with a tooth- brush). All video games were played for 30 min.

Helping or hurting behavior. Helping or hurt- ing behavior was assessed using the tangram puzzle procedure [Gentile et al., 2009]. Tangrams are based on seven differently shaped plastic pieces (e.g., small square, large triangle) used to form a specified out- lined shape. Participants chose 11 puzzles that their “partner” would attempt to complete from a set of ten easy, ten medium, and ten hard puzzles. Participants were told that their partner would win a $10 gift cer- tificate if he/she completed at least ten of the assigned puzzles within 10 min. Thus, participants could help

1A second prosocial game (as in Gentile et al., 2009) was initially tested, but its difficulty prevented many children from getting to the prosocial parts of the game.

their partner by assigning easy puzzles, or hurt their partner by assigning hard puzzles.

Postexperimental questionnaire. Participants rated their game on several dimensions using 10-point scales: action-packed, enjoyable, exciting, entertain- ing, fun, involving, hard to play, frustrating. They also rated their “ability on the video game task.” Repli- cating prior work [Anderson and Dill, 2000], these individual ratings formed a “fun” scale (alpha = .92) and a “difficulty” scale (alpha = .70). They were not significantly correlated with each other (r = .06). Par- ticipants also indicated their agreement with a state- ment that “The game involved helping other people” and a statement that “The game was violent.”

PROCEDURE

After arrival, both parent and child completed con- sent forms explaining the overall purpose2 and pro- cedures. Participants were told that the research in- volved how playing different video games affected performance on puzzles. Participants were told that they would play a video game by themselves, work together with a partner on a puzzle task, and that they would choose 11 tangrams for their partner to complete (there was no actual partner). If their part- ner completed ten of the 11 tangrams within 10 min, their partner would win a $10 gift certificate. They were told that after selecting tangrams for their part- ner they would receive 11 tangrams from their partner, and that although their performance would be scored they were not eligible to win a gift certificate. To jus- tify this unequal treatment, participants were told that one of the study’s goals was to determine whether a potential prize influences performance within this tangram task.

Participants received standardized tangram instruc- tions and a practice packet. After demonstrating un- derstanding of the tangram task, participants received video game instructions and practiced until they demonstrated mastery of the controls. Participants played the assigned game for 30 min by themselves. Then they chose 11 tangrams to assign their partner, and were encouraged to pick from multiple difficulty categories. Finally, participants completed the pos- texperimental questionnaire (including demograph- ics) and a funnel debriefing with open-ended probes for suspicion. Fourteen suspicious participants were identified and excluded from analyses, 11 males and

2The child was given a slightly different version of the consent form in which they were led to believe that there is an actual partner they will be participating with. The parent’s consent form revealed that there is no actual partner in this study.

Aggr. Behav.

284 Saleem et al.

three females. Suspicion was unrelated to experimen- tal condition, P > .20.

RESULTS

Scoring Tangram Choices

Participants chose medium difficulty tangrams most frequently (Mmedium = 4.65), followed by the hard (Mhard = 3.67), and easy tangrams (Measy = 2.70). Because the task had ten puzzles per difficulty level, participants had to pick from at least two cate- gories. It is possible for someone to pick ten medium tangrams and one easy (or hard) tangram to com- plete the 11 required. However, this individual is not necessarily intending to help (or harm) their partner, because the partner needed to complete only ten tan- grams to win the gift certificate. Thus, “helping” was operationally defined as the number of “easy” puzzles greater than one. Similarly, “hurting” was defined as the number of “hard” puzzles greater than one3.

Preliminary Analyses

Manipulation check. Participants’ perceptions of helpful and violent content in their assigned game were analyzed with a 3 (game type: prosocial, neu- tral, or violent) × 2 (rating type: helping or violent) analysis of variance, with rating type as a repeated measures factor. The predicted game type by rating type interaction was significant, F(2, 157) = 117.43, P < .001. The prosocial game (M = 7.18) was rated sig- nificantly higher on helpfulness than either the neutral (M = 2.13) or violent games (M = 1.84), Fs(1, 157) = 129.58 and 162.12, Ps < .001, respectively. The latter two conditions were not significantly different, F < 1.00. Also, the violent games (M = 4.54) were rated significantly higher on violence than the neutral (M = 1.07) or prosocial games (M = 1.28), Fs(1, 157) = 106.42 and 104.84, Ps < .001, respectively. The latter two were not significantly different, F < 1.00.

Covariates. Separate one-way ANOVAs (game type: prosocial, neutral, or violent) were conducted on the fun and difficulty game ratings. Difficulty ratings differed by game type, F(2, 169) = 20.07, P < .001. Contrasts revealed that the violent games (M = 4.68) were perceived as more difficult than the prosocial (M = 2.78) and neutral (M = 3.74) games, Fs(1, 169)

3For example, participants who chose 3 easy tangrams were assigned a “helping” score of 2. As in Gentile et al., 2009, the correlation between the raw # of easy and hard tangrams was large, r = −0.61, p < .001. Also as in that study, the “greater than one” scoring procedure reduced this correlation, r = −0.51, p < .001. Note that using the raw scores in the main analyses yielded results that were essentially the same as those reported here, also as in Gentile et al., 2009.

= 40.14 and 9.29, Ps < .001 and .01, respectively. The neutral games were rated as more difficult than the prosocial game, F(1, 169) = 9.51, P < .01. Thus, difficulty rating was used as a covariate in all analyses of the help/hurt tangram choices.

Fun ratings did not differ by game type, nor were any specific game contrasts significant, Fs < 2.00, Ps > .15. Furthermore, preliminary analyses found no significant effects of fun ratings or participant sex on helping and hurting behavior. Therefore, these pre- dictors were dropped.

Main Analyses: Tangram Choices

A 3 (game type: prosocial, neutral, or violent) × 2 (behavior type: helping or hurting) analysis of covari- ance was conducted on helping and hurting tangram behavior, with behavior type as a repeated factor, and trait aggression and game difficulty as covariates. The game type by trait aggression interaction was not sig- nificant and thus was not included in the final model. Game difficulty yielded no significant effects, Ps > .50, and is not discussed further.

Video game effects. The main prediction was that playing a prosocial game would lead to increased helpful and decreased hurtful behavior, relative to vi- olent games, with neutral games yielding intermediate behaviors. The means (shown in Fig. 1) fit this model quite well, and the specific contrast testing this model (decreasing linear contrast from prosocial to neutral to violent games for helpful behavior, increasing lin- ear contrast for hurtful behavior) accounted for 97% of the predicted interaction variance, F(1, 152) = 9.02, P < .001. Furthermore, deviations from the predicted pattern were nonsignificant, F < 1.

Fig. 1. Helpful and hurtful puzzle choices as a function of video game played.

Aggr. Behav.

Helping and Hurting 285

For both helpful and hurtful behavior, the proso- cial and violent game conditions were significantly different. Those who played a prosocial game were significantly more helpful (M = 2.25; SD = 2.23) than those who had played a violent game (M = 1.43; SD = 1.48), F(1, 152) = 4.65, P < .05, d = 0.35. Those who played a violent game were significantly more hurtful (M = 3.23; SD = 1.86) than those who played a prosocial game (M = 2.01; SD = 2.03), F(1, 152) = 9.20, P < .01, d = 0.49. The neutral condition means for helpful (M = 1.77; SD = 1.47) and hurtful (M = 2.82; SD = 1.89) behaviors fell between the other two game conditions.

Trait aggression effects. The behavior type by trait aggression interaction also was significant, F(1, 152) = 5.82, P < .05, d = .39. As expected, trait aggression was positively related to hurtful behavior and negatively related to helpful behavior, bhurtful = 0.15, P < .05, bhelpful = −0.16, P < .05, respectively. This further validates the tangram task as a measure of both prosocial and aggressive behavior.

DISCUSSION

The main goal of the current study was to test if short-term exposure to prosocial video games can in- crease helpful and decrease hurtful behaviors in chil- dren compared to neutral and violent games. Results revealed that video games with prosocial content in- creased helpful and decreased hurtful behaviors in a short-term experimental context with children. In contrast, children’s games with violent content in- creased hurtful and decreased helpful behavior. This study adds to the existing literature in several ways: (1) it is the first to test experimental prosocial video game effects on children; (2) it provides additional validity tests of the Tangram procedure for assessing aspects of helpful and hurtful behavior; (3) it is one of a handful of experimental studies that have used vio- lent children’s video games rather than more graphic and realistic violent games.

Sestir and Bartholow (2010) recently demonstrated that some nonviolent games can decrease aggression, despite having no prosocial content. Whether a given game or game-type appears to increase or decrease aggression or prosocial behavior depends, of course, on the conditions with which it is being compared. No-game comparison conditions have been problem- atic in this domain, because they differ in so many ways from the key violent or prosocial game condi- tions. Among other things, they tend to be boring or even frustrating to participants, especially to those who expected to be playing a video game or who

believe that other participants are playing. There- fore, most video game experiments have all partici- pants play some type of game, and the best ones as- sess potential confounding variables (such as fun and frustration) and control for them statistically when necessary. In the present experiment, even after sev- eral theoretically relevant factors were accounted for, results revealed that prosocial game content increased helpful and decreased hurtful behavior relative to both violent and neutral games.

The idea of prosocial games increasing helpful be- haviors in children in the short term is encouraging. Of course, more significant is the potential for these short-term effects to produce long-term changes. In- deed, social-cognitve learning theories suggests that the processes that produce such short-term effects can, with repeated exposure, lead to long-term in- creases in the accessibility and use of prosocial knowledge structures (including behavioral scripts), as demonstrated in cross-sectional and longitudinal studies [Gentile et al., 2009]. Although results from the present study are encouraging in finding be- havioral effects of prosocial games, it is important to study the underlying mechanisms responsible for these behavioral effects. Previous studies on proso- cial video game effects suggest that these behavioral effects might be due to changes in cognition [e.g., Gen- tile et al., 2009; Greitemeyer and Osswald, 2010, 2011] as well as affect [Greitemeyer and Osswald, 2010; Saleem et al., 2012]. Note that the studies testing the effects of prosocial games on affect have used an adult sample. Future research should explore this link us- ing a sample of children. It would be ideal to test the effects of prosocial video game content on prosocial and antisocial cognitions, affect, and behaviors within the same study in order to do meditational analyses exploring the underlying mechanisms responsible for prosocial effects on prosocial behaviors.

Similar to the behavioral study by Gentile et al. (2009), the present study assessed helpful and hurt- ful behaviors using a single help/hurt tangram mea- sure. This method has an important advantage in that helpful and hurtful behaviors can be assessed simulta- neously. Although helpful and hurtful behaviors are conceptually distinct, they often are inversely related especially in the short-term real world contexts. For example, when people engage in a hurtful behavior toward a target person, they seldom simultaneously engage in helpful behavior toward that same target. By allowing participants the option to help, be neu- tral, or hurt another individual, this tangram measure has the potential to assess a range of interpersonal be- haviors related to conflict.

Aggr. Behav.

286 Saleem et al.

However, an important limitation of this design is that individuals who score high on helpfulness by se- lecting a greater number of easy puzzles will tend to score low on hurtfulness, and vice versa. Indeed, even after using our “greater than one” scoring proce- dure, the correlation between the helpful and hurtful scores remained at r = −.51. Thus, there is an is- sue of nonindependence in the way hurtfulness and helpfulness is assessed using the tangram measure. In previous studies [Gentile et al., 2009] and in the current study, this concern has been addressed in at least three ways: (1) ignoring the medium cate- gory for the analyses, thus reducing interdependence; (2) using the number of easy and difficult puzzles greater than one instead of raw scores so partici- pants can obtain a score of 0 on both helpfulness and hurtfulness; and (3) entering both helpful and hurtful scores as a within-subject factor. An alterna- tive scoring procedure is to derive one overall score by assigning equidistant weights to hard, medium, and easy choices, such −1, 0, +1.4 In the present study, this overall score yielded significant main ef- fects of game type, F(2, 152) = 4.43, P < .05, and trait aggression, F(1, 152) = 5.31, P < .05. The pat- tern of means was as expected, with the most positive scores by participants who had played the prosocial game and the most negative scores by those who had played a violent game. This contrast pattern was sig- nificant, F(1, 152) = 8.40, P < .01, and accounted for 95% of the between groups variance. Of course, none of these solutions completely solve the nonin- dependence question, either with the tangram task or with the broader conceptual questions of whether people can (or do) try to help and hurt others at the same time. Nonetheless, the present study found that prosocial and violent video games with unreal- istic cartoonish characters significantly affected chil- dren’s behavior in a task where they could help or hurt another child. We believe that there is a need for continued research using novel measures, such as the tangram task, in order to fully understand video game effects on children. Furthermore, additional work with the tangram task is needed to explore its conceptual and methodological advantages and lim- itations within the prosocial and aggressive behavior literatures.

4Note that statistically this is identical to assigning 1, 2, & 3 to the number of hard, medium, and easy tangram choices. We prefer −1, 0, & +1 because negative and positive scores indicate a preponderance of hard vs. easy choices.

REFERENCES

Anderson CA, Gentile DA, Buckley K. 2007. Violent video game effects on children and adolescents. New York: Oxford University Press.

Anderson CA, Sakamoto A, Gentile DA, Ihori N, Shibuya A, Yukawa S, et al. 2008. Longitudinal effects of violent video games aggres- sion in Japan and the United States. Pediatrics 122:1067–1072.

Anderson CA, Dill KE. 2000. Video games and aggressive thoughts, feelings, and behavior in the laboratory and in life. J Pers Soc Psychol 78:772–790.

Anderson CA, Shibuya A, Ihori N, Swing EL, Bushman BJ, Sakamoto A, et al. 2010. Violent video game effects on aggression, empathy, and prosocial behavior in Eastern and Western countries. Psychol Bull 136:151–173.

Buckley KE, Anderson CA. 2006. A theoretical model of the effects and consequences of playing video games. In: Vorderer P, Bryant J, editors. Playing video games—motives, responses, and conse- quences. Mahwah, NJ: LEA. pp 363–378.

Bushman BJ, Huesmann LR. 2006. Short-term and long-term effects of violent media on aggression in children and adults. Arch Pediat Adol Med 160:348–352.

Buss AH, Perry M. 1992. The aggression questionnaire. J Pers Soc Psychol 63:452–459.

Gentile DA. 2009. Pathological video game use among youth 8 to 18: A national study. Psychol Sci 20:594–602.

Gentile DA, Anderson CA, Yukawa S, Ihori N, Saleem M, Ming LK, et al. 2009. The effects of prosocial video games on prosocial be- haviors: International evidence from correlational, experimental, and longitudinal Studies. Pers Soc Psychol B 35:752–763.

Gentile DA, Gentile JR. 2008. Violent video games as exemplary teach- ers: A conceptual analysis. J Youth Adolesc 9:127–141.

Greitemeyer T, Osswald S. 2009. Prosocial video games reduce aggres- sive cognitions. J Exp Soc Psychol 45:896–900.

Greitemeyer T, Osswald S. 2010. Effects of prosocial video games on prosocial behavior. J Pers Soc Psychol 98:211–221.

Greitemeyer T, Osswald S. 2011. Playing prosocial video games in- creases the accessibility of prosocial thoughts. J Soc Psychol 151:121–128.

Huesmann LR, Kirwil L. 2007. Why observing violence increases the risk of violent behavior in the observer. In: Flannery DJ, Vazsonyi AT, Waldman ID, editors. The Cambridge handbook of violent behavior and aggression. Cambridge, UK: Cambridge University Press. pp 545–570.

Hopf WH, Huber GL, Weis RH. 2008. Media violence and youth violence: A 2-year longitudinal study. J Media Psychol 20:79–96.

Maier JA, Gentile DA. 2012. Learning aggression through the me- dia: Comparing psychological and communication approaches. In: Shrum LJ, editor. The psychology of entertainment media: Blur- ring the lines between entertainment and persuasion, 2nd edition. New York: Taylor & Francis. pp 267–299.

Moller I, Krahe B. 2009. Exposure to violent video games and aggres- sion in German adolescents: A longitudinal analysis. Aggr Behav 35:75–89.

Reynes E, Lorant J. 2001. Do competitive martial arts attract aggres- sive children? Percept Motor Skills 93:382–386.

Reynes E, Lorant J. 2002. Effect of traditional judo training on ag- gressiveness among young boys. Percept Motor Skills 94:21–25.

Reynes E, Lorant J. 2004. Competitive martial arts and aggressiveness: A 2-yr. longitudinal study among young boys. Percept Motor Skills 98:103–115.

Saleem M, Anderson CA, Gentile DA. 2012. Effects of prosocial, neutral, and violent video games on college students’ affect. Aggr Behav 38:263–271.

Aggr. Behav.

Helping and Hurting 287

Sestir MA, Bartholow BD. 2010. Violent and nonviolent video games produce opposing effects on aggressive and prosocial outcomes. J Exp Soc Psychol 46:934–942.

Swing EL, Gentile DA, Anderson CA. 2008. Violent video games: Learning processes and outcomes. In: Ferdig RE, editor. Hand- book of research on effective electronic gaming in educa- tion. Hershey, PA: Information Science Reference. pp 876– 892.

Wallenius M, Punamaki R. 2008. Digitial game violence and direct aggression in adolescence: A longitudinal study of the roles of sex,

age, and parent-child communication. J Appl Dev Psychol 29:286– 294.

Walters GD, Ronen T, Rosenbaum M. 2010. The latent structure of childhood aggression: A taxometric analysis of self-reported and teacher-rated aggression in Israeli schoolchildren. Psychol Assess 22:628–637.

Zhen S, Xie H, Zhang W, Wang S, Li D. 2011. Exposure to violent video games and Chinese adolescents’ physical aggression: Gen- der and developmental differences. Comput Hum Behav 27:1675– 1687.

Aggr. Behav.

Copyright of Aggressive Behavior is the property of John Wiley & Sons, Inc. and its content may not be copied

or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.

However, users may print, download, or email articles for individual use.