Abstract
Object recognition can be a challenging task when identifying objects with strong similarities in appearance. Military vehicles share structural similarities that can make recognition extremely difficult (O'Kane, Biederman, Cooper, & Nystrom, 1997). That difficulty contributes to misidentification errors and leads to 'friendly fire' accidents (Regan, 1995). To investigate armored vehicle identification training methodologies, this study examined the effect of adding gamelike features to a two alternative forced choice (2AFC) combat vehicle identification training tool. Twenty-six undergraduate students were recruited and received course credit for their participation. They were trained using a 2AFC task that presented gamelike feedback for each training trial. Trials consisted of, a display that presented the name of a vehicle (e.g. M1A1) followed by two images. Participants had to choose which image was associated with the initial name presented. After which, appropriate feedback was presented based on the accuracy of their guess. That is, guessing correctly was coupled with positively valenced arousing sound effects and points as feedback for each training trial. Participants learned seven armored vehicles across six training blocks that lasted approximately two minutes each. Overall performance in the training phase was 74.9% (SD = 11.06%) correct and was significantly greater than chance (i.e. 50%). Participants' training performance significantly improved as the experiment progressed. That is, participants achieved 80.7% accuracy after four training blocks (approximately eight minutes of training). In addition, participants' correct trial response times significantly decreased. Therefore, it seems that participants became more accurate and faster as they interacted with the game-like 2AFC training. Future research will compare the learning effects of the gamelike 2AFC tank training to a control group without gaming features, explore the effects of gamification on different training modalities (i.e. augmented reality), and examine the effects of learning more vehicles.
Meeting abstract presented at VSS 2014