Specificity of Transfer Effects from Video Game Training

Abstract

Although the transfer of cognitive performance from video game training is a real possibility, the current literature on the topic is mixed, and prone to a range of confounds. This dissertation developed and tested an improved method for studying transfer of skills from video game training that uses the same game as its own active control. An industrial research collaboration with Canadian video game company Telos International provided a number of commercial games that were able to be modified towards these research goals. In Chapter 2, an initial proof-of-concept study assessed specific near-transfer effects from training on “Membrain”, a 3D spatial memory game, versus training with Sudoku, a traditional number puzzle game. In Chapters 3 and 4, participants played the same commercial video game over several weeks, where the relative proportions of particular easier or harder game elements were manipulated to give different groups of participants more or less experience with particular kinds of game features. In Chapter 3, the “Paint the West” game (a speeded shooting gallery-style game) showed a range of specific transfer effects from increased distractor stimulus similarity, number and crowding within the game, to both improved performance on resisting the influence of nearby distractor stimuli in cognitive tasks, and also to changes in speed/accuracy criterion performance in other speeded tasks. In contrast, in Chapter 4, the “Orphlings” game (a problem-solving spatial puzzle game) showed no convincing transfer effects with a range of working memory and spatial tasks. I suggest that these methods allow for a better estimate of the true effect size of game-specific training improvements, and that the transfer of training effects observed in this research is more directly and reliably attributable to the particular well-controlled feature variations between training groups within the same game context.