The Perceptual and Cognitive Effects of Action Video Game Experience
C. SHAWN GREEN
For decades, the prevailing view in the field of cognitive neuroscience was that, upon reaching maturity, the adult brain settles into a relatively fixed, unchanging state. Consistent with such an interpretation, for instance, significant effort was put into delineating more or less plastic stages of development (e.g., critical periods; Wiesel & Hubel, 1963). More recently, however, this view has shifted substantially, with current research establishing that the brain possesses an enormous capacity for reorganization throughout the lifespan (Bavelier, Levi, Li, Dan, & Hensch, 2010). Even some critical periods that were previously believed to be quite rigid have since been shown to be flexible and in fact can often be reopened via behavioral means (e.g., through dark exposure; He, Ray, Dennis, & Quinlan, 2007). Such research on neural plasticity has spurred tremendous interest in the development of training regimens to improve brain function in domains ranging from motor skill, to vision and hearing, to broader classes of high-level cognition.
However, while we now know that the human brain retains some level of plasticity even into old age (obviously not at equivalent levels in each age range), a major obstacle remains. This obstacle has been dubbed the “curse of specificity” (Bavelier, Green, Pouget, & Schrater, 2012), and it refers to the fact that although humans show increases in performance on virtually any task given appropriate practice, the enhancements are typically limited to the exact characteristics of the trained task; little or no transfer of learning is observed to even seemingly highly similar untrained tasks (Fahle, 2005). For instance, in seminal work by Fiorentini and Berardi (1980) in the domain of perceptual learning, participants were trained to discriminate between two complex gratings. Over the course of three sessions of training, performance improved from chance levels all the way to ceiling levels. Yet, when the gratings were altered in seemingly minor ways (e.g., in orientation or spatial frequency), subject performance returned to chance levels. Similar specificity has been seen for low-level features such as retinal location, motion direction, motion speed, or even the trained eye. Furthermore, although such specificity has been perhaps most thoroughly described in the field of perceptual learning, it has been documented in essentially all fields that focus on learning, including motor learning, training of high-level cognitive skills such as working memory, and even in education (Barnett & Ceci, 2002; Redick et al., 2013; Tremblay, Houle, & Ostry, 2008). It should be intuitively obvious how significant an impediment such specificity can be for those whose goal is to construct learning paradigms for practical purposes such as rehabilitation (where success necessarily requires benefits that extend beyond the exact laboratory setup).
Interestingly, there are a variety of types of experience, which often correspond to real-world activities that have been shown to produce learning that extends beyond the specifics of the trained contexts. Music training is one such domain. In one study, for example, children who received musical training (vocal or keyboard), showed significantly larger improvements on the Wechsler Intelligence Scale for Children (which clearly bears little resemblance to vocal or keyboard training) than did children who received drama training (Schellenberg, 2004). Similarly, in the athletics domain there are myriad examples wherein individuals with extensive experience playing a given sport demonstrate enhanced abilities at basic laboratory tests (Kida, Oda, & Matsumura, 2005; Lum, Enns, & Pratt, 2002). There is the further focus of this review—playing action video games (Green & Bavelier, 2012).
Before examining these effects, we briefly discuss what makes a game an “action” video game. While there are no quantitative rules that can be applied to perfectly separate the various video game genres, there is a set of qualitative features that all action games share. In particular, action video games are those that involve exceptional speeds (both in terms of the velocity of moving items and the brevity of transient events). These games also involve extraordinary perceptual load (whereby the individual must track many objects), cognitive load (which entails considering many possible outcomes), and/or motor load (which involves engaging in multiple action plans). The games also involve temporal and spatial unpredictability and require a high degree of peripheral processing. Games that fit these criteria include so-called first-person shooter games like the Call of Duty series, third-person shooter games like the Gears of War series, and some car driving games. To one not familiar with the various video game genres, these may seem like unimportant points, but we have seen that the effects of playing various video games depends highly on the games’ content and structure (Cohen, Green, & Bavelier, 2007). Simply put, not all games produce the same types of benefits, if they provide benefits at all.
Here we review the ever-growing literature on the effects of video game experience on vision, attention, and cognitive skills. Although the paradigms that will be reviewed were designed to test processes that are thought to be relatively independent, at the conclusion we will suggest that the results of each can potentially be accounted for by a single common underlying mechanism. As the majority of the literature has compared the performance of expert action video game players (VGPs: usually defined as individuals who play more than five hours a week of action video games) against non-action video game players (NVGPs: who play no action games, though they may play other game genres), we will adopt this focus for our review. However, as simple population differences do not themselves prove a causal link, we will specifically highlight studies that have demonstrated such a link through well-controlled training paradigms.