Bluemke, M., & Zumbach, J. (2007). Implicit and explicit measures for analyzing the aggression of computer gamers. Emotions and aggressive behavior (pp. 38-57). Ashland, OH US: Hogrefe & Huber Publishers. Retrieved July 3, 2009, from PsycINFO database.
(from the chapter) The effects of playing computer games on children has become an important issue in psychological research in the last couple of years. The question of whether playing computer games leads to increased aggression, particularly among those playing violent games excessively, has stimulated numerous studies. Currently, the meta-analytic findings seem to imply that being exposed to violent games increases aggressive behavior, aggressive cognition, aggressive affect, and physiological arousal as well as it decreases helping behavior on average. Nevertheless, from a meta-analytical point of view, each effect size is subject to errors of unknown magnitude, and these errors can only be overcome by aggregation of similar studies, yielding more reliable estimates of effect sizes. It is not surprising that some researchers question that violent video games have detrimental effects in real life contexts–either on the basis of single-study results or on the grounds of legitimate arguments. In this chapter, we emphasize another important issue when encountering weak or null effects in aggression-related studies: the methods used when assessing aggressiveness and aggressive behaviors. More precisely, it is a well-known problem that self-reports on socially sensitive topics (such as aggressive thoughts, aggressive feelings, and aggressive behavior) can be biased and, thus, lead to inconsistent results. Within this contribution we argue that these self-reported variables, i.e., explicit measures, might meaningfully be complemented by newly developed methods, so-called implicit measures, that aim at assessing automatic affect and cognition. Although there are still many methodological problems to overcome, implicit measures have already added to our knowledge of preactivation of emotional and cognitive content in social encounters. In the following, we will (a) highlight advantages of implicit measures, (b) suggest that aggression research might profit from measuring (automatic) affective reactions and predicting (spontaneous) behaviors, (c) describe the general procedure of aggression-related Implicit Association Tests (lATs), (d) and show how these automatic cognitive-affective processes can be integrated into current models of aggressive emotions and aggressive reactions. Finally, we will report our own findings that used implicit measures to explore the consequences of playing violent computer games with regard to the automatic self-concept and attitudes toward aggressive behavior among young adults and school children. (PsycINFO Database Record (c) 2008 APA, all rights reserved)