PRZYBYLSKI.DECI.RIGBY.AND RYAN ST-baprchmpliesrp how they influence motivation and aggression.In line with Sher ctions.For nple.in a classi 1963)ha om pas ive forms of media such as television.film t did not o ples of thwartine of ne how player skill can tically aggressio the and magnitd e of effects linking violen (1941 tha stration may instigate t exp sure to playe n it may p g from a motivational p h n the present mple.in Sud some ly ma hle and subiect ha re unable t any at the e to which a p her in testing the f th. eelings will impa find with a togethe the pre ion-facilite ns to result in ob ble shifts in s the fuller ature of the g et a 12)s th hat small when uch a (1989 d en t et al men aggression,without mplification of the ed there is a dar gress olent cue e the Links With Content-Oriented Theories of Aggression that ve play is fran The results we derived in the course of evaluating the ential of com ocial i edance hypoth speak e lively c n,and possibly aggression (e.g..Adachi Wil ically. ch highlights how Avenues for Future Research suggesting that th y used practice c fusing diffe The present research features a number of limitations that sug logical need postulated by SDT.n that Mediation nalysis provided evidence that this thwarting of com games that thwart the other two basic psych the of rel autonomy. The areful co in motivational the warting hypothesis in SDT of games that mass appealals search to date.That said.there is good SDT-based approach implies a more specific hypothesis, namely that the critical blockage is the thwarting of basic psychological need satisfactions. For example, in a classic study, Buss (1963) had college students experience one of three types of frustration: failure to win money, failure to earn a better grade, or failure on a task. All three groups showed more subsequent aggression than a control group that was not frustrated. For us, all these represented examples of thwarting of the need for competence. SDT research focused on thwarts to the needs for autonomy (Roth, Kanat￾Maymon, & Bibi, 2011) and relatedness (Weinstein, 2009) have shown a conceptually similar pattern of findings. Miller (1941) noted that frustration may instigate aggression, but this is not the only type of emotional or motivational pertur￾bation it may produce. Operating from a motivational perspective, we agree with this position for two reasons. First, competence￾impedance prompts a number of changes, including some we observed in the present research. For example, in Studies 4 –7, we found competence thwarting was also linked (negatively) to player enjoyment and motivation. Second, links from need-thwarting leads to aggression will no doubt be subject to a number of moderators. In line with Miller, SDT-based work had indicated that the degree to which a person autonomously regulates aggres￾sive thoughts and feelings will impact on observable aggressive behavior (see Legault, Green-Demers, Grant, & Chung, 2007). The present studies also interface with a more contemporary formulation of the frustration-aggression hypothesis. Berkowitz (1989) highlighted aggression-facilitating cues as necessary con￾ditions for goal frustrations to result in observable shifts in aggres￾sion. Although we did not specifically hypothesize interactions between violent cues or imagery and frustrated competence based on motivational theorizing, we did test for them in Studies 1, 2, and 5. In these studies, we did not find empirical support for the idea that violent gaming stimuli would amplify the intensity of links between undermined competence (i.e., frustration) and short-term shifts in aggressive feelings. These findings are not sufficient to disprove Berkowitz’s (1989) reformulated frustration-aggression hypothesis, but in three experiments we found competence￾impeding gaming provided the necessary and sufficient conditions for our measures of aggression, without amplification of the ag￾gression by violent cues. Links With Content-Oriented Theories of Aggression The results we derived in the course of evaluating the competence-impedance hypothesis also speak to the lively debate surrounding violent gaming content as a source of human aggres￾sion. Specifically, the present research highlights how critical it is to experimentally disentangle interactive aspects of games when evaluating postengagement experiences. In Study 1, we found evidence suggesting that the widely used practice of using differ￾ent games as prototypes of high versus low violence play is problematic from a motivational perspective. In this instance, the violent game provided a very complex interface, which impeded players developing mastery over the game’s control interface. Mediation analysis provided evidence that this thwarting of com￾petence led to aggressive feelings. Findings across all of the experiments highlighted the fact that games are inherently inter￾active and require careful consideration in motivational terms. The interactive aspects of games that drive their mass appeal also affect how they influence motivation and aggression. In line with Sherry (2007), the present work supports the idea that gaming is clearly delineated from passive forms of media such as television, films, and music. Whereas observing violence in passive activities such as watching television tends to enhance aggression in the viewers, that did not occur for the interactive activities of game playing in the current studies. The present findings specifically underline the importance of considering how game structure and player skill can and do systematically influence aggression. The present research also speaks to the wider controversy sur￾rounding the existence and magnitude of effects linking violent game content exposure to player aggression. The central aim of the present work was to investigate the robustness of the link between competence-impedance and aggression in gaming contexts. To this end, many of our studies included exposure to different levels of violent game content to evaluate the empirical rigor of our hypoth￾esis. Our designs explicitly manipulated violent game content in Studies 1, 2, and 5 and observed variation in game violence in Study 7. It was both notable and unexpected to our research team that we were unable to replicate any violent content effects that the general aggression model would predict across these studies. Fur￾ther, in testing the aggressive-cues corollary of the frustration￾aggression hypothesis, we did not find any evidence for interac￾tions between player (in)competence and violent content on aggression. Viewed together, the present findings do not provide support the GAM framing of gaming-related aggression. Instead, they suggest there is a more nuanced picture yet to be painted that describes the fuller nature of the gaming-aggression relations. The present findings are more in line with some of the more recent studies of gaming-related aggression. A recently published longitudinal study by Willoughby et al. (2012) showed that the effects of violent game content exposure and behavioral correlates of aggression were quite small when important covariates such as peer deviance and family relationships were considered. The prin￾cipal contribution of this study from a motivational perspective was the consideration given to game structure. Willoughby et al. 2012 identified the competitive aspect of some games, not violent contents per se, as a factor that accounted for part of the gaming￾aggression link. If indeed there is a dark side to electronic gaming, it could be in the alienating potential of some forms of direct player competition. Indeed, from an SDT-based perspective, the way that competitive play is framed—in terms of narrative and motivational affordances—may be key to understanding the po￾tential of competitive gaming scenarios to foster social isolation, player alienation, and possibly aggression (e.g., Adachi & Wil￾loughby, 2011). Avenues for Future Research The present research features a number of limitations that sug￾gest fruitful avenues for future work. First, the present studies had a narrow theoretical scope. Although competence is a key psycho￾logical need postulated by SDT, one that is especially salient in computer games, the present research can say little about features and experiences of games that thwart the other two basic psycho￾logical needs of relatedness and autonomy. The present work focused on the competence need because it has received the least attention in terms of the need-thwarting hypothesis in SDT re￾search to date. That said, there is good reason to think that This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. 454 PRZYBYLSKI, DECI, RIGBY, AND RYAN