A socio-cognitive model of video game usage.

Consistent with prior research on Internet usage (LaRose et al., 2003; LaRose & Eastin, 2004), the findings of the present study highlight the importance of self-regulatory mechanisms in the consumption of interactive media. That is, people tend to spend a substantial amount of time playing video games because their video gaming can provide self-reactive incentives to relieve boredom, lessen loneliness, pass the time, or provide an escape. Further, people are likely to engage in more video game playing by pursuing self-reactive incentives, which not only promote the loss of self-control over their media consumption but also trigger a repeated pattern of video game play. Consequently, people are no longer responsive to active consideration for their own video gaming and tend to play much longer than originally intended.

This study also demonstrated that video game players' flow experience instigated self-reactive outcome expectations. These in turn fostered deficient self-regulation and habit strength, the two main components of self-regulatory mechanisms. The findings also suggest that video game players' flow experience is likely to promote the loss of self-control. Thus, flow experience may not only activate self-reactive outcome expectations but also may make video game self-regulation less effective so that their video game consumption becomes repetitive and out of control. The present research also confirmed a basic tenet of the theory of flow experience: that the flow state is more likely to be experienced when perceived self-efficacy is optimally matched with perceived challenge.

[FIGURE 2 OMITTED]

The emergence of the unexpected causal path from optimal balance to deficient self-regulation in the final path diagram may also suggest that people are more likely to lose their self-control and continue their game play to an excessive extent particularly when their gaming skills optimally correspond to the level of the challenge of the video game. This significant casual link may not be surprising because optimal balance between self-efficacy and challenge can easily motivate players to continue to engage in their gaming so that they can strive to achieve self-administered reward for their performance. As such, optimally balanced skills and challenge in video game play may not only contribute to inducing flow experience, but also influence deficient self-regulation directly. Thus, this significant path should be regarded as a theoretical extension to the proposed model.

The present study also showed that video game players' flow experience was positively correlated with the amount of video game usage. However, the direct influence of flow experience on the amount of video game usage was not evident in the proposed model. The null relationship between flow experience and video game usage in the model may counter the findings of several flow studies (e.g., Ghani & Deshpande, 1994; Novak & Hoffman, 1997; Novak et al., 2000; Trevino & Webster, 1992; Webster et al., 1993) all suggesting that flow experience is directly related to the amount of time spent on media consumption. However, none of the prior studies incorporated the full range of variables found to predict media behavior here and so might have misidentified flow experience as an important predictor of usage.

Another possible explanation for the lack of direct path from flow experience to video game usage is that flow is so fleeting that it does not directly affect overall amount of time spent on video games. Instead, the flow state may only affect the duration of the play session in which it is achieved. Perhaps players reinstate their self-control over their video game play by restoring their self-observation, normative judgment, and self-reactive influence between game playing sessions. The temporary slip in self-control may be a temporary setback brought about by strong situational inducements that are eventually avoidable and alterable (Bandura, 1997).

People's ability to reinstate self-control to resist excessive video game consumption was apparent. Among people who agreed with none of the three addiction symptoms of the deficient self-regulation measure, and thus fell well below the criterion of video game addiction, there was still significant correlations between deficient self-regulation and video game usage (r = .48, p < .001, n = 283). This finding is consistent with LaRose et al.'s (2003) notion of unregulated media use. Unlike the all-or-nothing phenomenon of media addiction, deficient self-regulation was significantly related to media usage within even a normal population. Further, only 10 out of 388 video game players were possible "addicts" in this study, if one follows a criterion for media addiction suggested by Shaffer et al. (2000), in which harmful life consequences are a necessary condition. This finding suggests that concern about video game addiction may be exaggerated and that excessive video game play should be self-correctable for most people.

The present socio-cognitive view of problematic video game use also sheds light on the excessive use of traditional mass media, including television (e.g., Kubey & Csikszentmihalyi, 2002), as well as other interactive media in the Internet environment (e.g., Caplan, 2005; Diddi & LaRose, 2006; LaRose et al., 2003; Lee & Perry, 2004). As suggested in the present study, addictive television consumption behavior also may be triggered by internal cues (e.g., boredom, loneliness, or depression) and external cues (the sight of TV monitor or remote control device, or the incessant barrage of sensory cues of TV programs), which in turn promote TV viewers' self-reactive outcome expectations to relieve or lessen the unpleasant moods, habitual pattern of TV viewing, and loss of self-regulation over TV viewing. Parallel to this view, mood management theory (Bryant & Zillmann, 1984; Zillmann, 1988) also predicts that TV viewing is associated with dysphonic moods, which is indicative of problematic TV consumption behavior. Similarly, Anderson, Collins, Schmitt, and Jacobvitz (1996) found that stress is significantly linked to TV addiction. Consequently, TV viewing becomes rewarding itself by providing a means of escape from unpleasant moods and TV viewers get into downward spirals of unregulated media consumption behavior (LaRose et al., 2003).

Limitations

The present study employed a student sample that may be not generalizable to other populations who have different levels of video game usage and experience. However, the socio-cognitive mechanisms of video game consumption behavior uncovered in the current college student sample can be relevant for testing a nascent theoretical model, particularly when addressing the underlying mechanisms of the emergent media consumption behavior (Caplan, 2005; Pingree et al., 2001). As video games become integrated into the daily life of college students with new communication technologies (e.g., Jones et al., 2003), college students can be a suitable audience for understanding problematic video gaming habits that are an important conceptual issue of this study.

The self-reported recall measures of time spent playing video games also may be a limitation of the present study. In the first Middletown Studies, Papper, Holmes, and Popovich (2004) contended that self-report measures of surveys and diaries were less accurate than direct observations in reporting media use because people tend to underestimate their media use when using the self-report measures. The combination of the self-report measures and the direct observations could strengthen the validity of the result in future research.

The present study implemented cross-sectional data which provided limited support for the longitudinal assumption of the direct and indirect relationships hypothesized in the proposed model. A future study with a solid longitudinal design will be capable of demonstrating how the relationships among the involved variables develop in long-term processes of self-regulation in video game consumption behavior.

Suggestions for Future Research

The relationship between habit and deficient self-regulation should be further investigated. The two constructs may be closely related subdimensional indicators of an unobserved higher level of construct, such as self-regulation. Further research should develop alternative measures of deficient self-regulation and habit strength to achieve better discriminant validity.

It also remains uncertain whether the relationships unobserved (i.e., direct path between flow experience and usage) as well as observed in this study can be found across categories of video games (e.g., classic board game, action-adventure, role playing, puzzle, quiz, racing, strategy, etc.). For example, Sherry et al. (2006) found three underlying dimensions of video game genre preference (imagination, traditional, and physical enactment games) which corresponded to different types of gratification outcomes (e.g., social interaction, status, novel sensory, pass time, etc.).