Adding self-explanation prompts to an educational computer game

•Taught fraction arithmetic to school children in a computer game called Save Patch.•Answering focused self-explanation prompts during the game improved game play.•Answering other self-explanation prompts during the game did not improve game play.•Game play is improved by some kinds of self-explanation prompts and hurt by others. Proponents envision a role for computer games in improving student learning of academic material, including mathematics and science. Asking learners to engage in self-explanations during learning has been found to be an effective instructional method. In the present experiment, we examined the effects of adding a self-explanation prompt—asking players to answer one of three questions after completing each level of the game—within a children’s math game on addition of fractions. Middle-school participants played either a base version of the game (n=57) or the base version with a self-explanation instructional feature (n=57). Participants’ learning was measured by a fractions posttest and their learning processes measured via in-game measures of game progress and errors. When we separated the self-explanation condition into participants who used a focused self-explanation strategy versus those who did not, the focused participants had significantly fewer game level deaths, game level resets, and progressed significantly farther in the game, compared to the control group, than participants not using a focused self-explanation strategy. The major new contribution of this study is that self-explanation can help the process of playing educational games in some situations and hurt in others. In particular, the most effective self-explanation prompts were aimed at helping learners make connections between game terminology and mathematics terminology, whereas the least effective self-explanation prompts asked very simple or very abstract questions.