Abstract
Playing first-person shooter (FPS) video games has been shown to improve performance in a number of visual tasks (e.g., Green & Bavelier, 2003). Here we ask whether a specific aspect of FPS games (friend vs. foe discrimination) might be responsible for some of these improvements. Non video game players were identified using a covert assessment strategy and invited to the study. On Day 1, all participants completed a pre-test consisting of the filter, visual working memory, and flanker tasks and then were assigned to one of three groups: no training, friends and foes, or foes only. The no training group returned on Day 4 to take the post-test consisting of the same tasks as the pre-test. The friends and foes and foes only groups played custom designed video game levels for two hours (30 minutes on Day 1, and 45 minutes on Days 2 and 3) before taking the post-test on Day 4. In the friends and foes condition, half of the characters in the game world attacked the player and the other half ran up to and attempted to follow the player. In the foes only condition, all characters attacked the player. The player’s task was to eliminate all enemies (without shooting friends) and then progress to the next room. The game world consisted of 16 rooms that progressed in difficulty. Preliminary results indicate that filter task performance for the friends and foes condition was significantly better than the no training condition (p <.01) and marginally better than the foes only condition (p = .083) though training condition did not interact with time. Additionally, participants in the friends and foes condition showed a decrease in flanker interference that was marginally larger than the decreases in the foes only (p = .104) and no training (p = .054) conditions.
Meeting abstract presented at VSS 2012