Explicit and Implicit Temporal Learning Using an Action Video Game

Abstract

Temporal training has been shown to improve temporal discrimination (Bueti & Buonomano, in press). Numerous studies have reported improved performance for the trained interval presented with different sensory characteristics (e.g., pitch, color, position, orientation; Karmarkar & Buonomano, 2003; Nagarajan et al., 1998; Westheimer, 1999; Wright, Buonomano, Mahncke, & Merzenich, 1997; Wright, Wilson, & Sabin, 2010) or in a different modality (Bratzke, Seifried, & Ulrich, 2012; Bueti, Lasaponara, Cercignani, & Macaluso, 2012; Nagarajan et al., 1998). However, temporal generalization has rarely been reported (Lapid, Ulrich, & Rammsayer, 2009; Wright, Buonomano, Mahncke, & Merzenich, 1997), making temporal perceptual learning mostly interval specific. The failure to show generalization might be due to the standard psychophysical training tasks that are typically non-engaging. Action video game training, on the other hand, has been shown to enhance performance on a wide variety of perceptual tasks that: require the flexible allocation of attentional resources, evaluate the spatial characteristics of vision, and measure the temporal characteristics of vision (Green, Li, & Bavelier, 2010). To date, studies on the effects of action video game training have been limited to the visual and spatial domain, while no such training has been applied in the temporal domain. The above-mentioned training regimes involved explicit training of a specific parameter. It has been shown, however, that implicit learning can also be achieved. For instance, implicit learning of rhythms has been previously demonstrated (Salidis, 2001), but as yet no such evidence exist for other aspects of timing such as time estimation. Drawing from evidence in the fields of temporal training, implicit learning, and video gaming, the goal of our experiment is to determine whether temporal training, explicit or implicit, can lead to improved performance on a temporal reproduction task. Participants will be assigned randomly into one of four groups (explicit – random and non random group, implicit – random and non random group) and will be pre- and post-evaluated in a vigilance, a working memory, and a temporal reproduction (visual and auditory) task. These three pre- and post-tasks have been chosen to serve a dual purpose. Firstly, participants will be evaluated in these tasks prior and after the training, instead of only in the temporal reproduction task, to ensure that they remain naive as to the purpose of the study. Secondly, as an internal clock account will be used to interpret our findings, we will measure potential enhancement in processing speed, working memory or attention due to the training regime. The training regime will be an action video game (visual) task rather than a psychophysical one, as video game experience is known to be physiologically arousing and psychologically motivating (Green, Li, & Bavelier, 2010). The game will be designed in a way that a bonus will appear on the screen randomly or at regular intervals. Knowledge of this will either be made explicit to the participants (explicit condition) or not (implicit condition). We anticipate that the group that will receive information about the regularity of the bonus and will train on the regular interval condition (explicit – non random group) will demonstrate higher accuracy and lower variability than all other experimental groups in the reproduction task at the post-training phase. We also anticipate that the group that will receive no information about the regularity of the appearance of the bonus and will train on the regular interval condition (implicit – non random group) will be more accurate and least variable as compared to the random conditions (explicit or implicit). Given that the training regime will be visual, pre- and post-testing in an auditory reproduction task will also allow for the measurement of cross modal interval enhancement.