The purpose of this study was to examine the effects of type of interactivity, prior knowledge and gender on learners' performance in learning computer programming skills through an experiential game-based learning activity. One hundred and forty six junior high school students participated in the experimental activity. Participants' levels of prior-knowledge (high vs. low) were identified according to previous computer course performance. The interactions of type-of-interactivity-prior-knowledge and type-of-interactivity- gender were significant on participants' performance. The analysis revealed that (a) for the gender effect, the male group outperformed the female group while receiving the challenge-interactivity game-play activity, but while receiving the mixed-interactivity game-play activity, the male group and the female group performed equally; (b) similarly, for the prior-knowledge effect, the high prior-knowledge group outperformed the low prior-knowledge group while receiving the challenge-interactivity game-play activity, but while receiving the mixed-interactivity game-play activity, the high prior-knowledge group and the low prior-knowledge group performed equally. The mixed-interactivity game-play was found to compensate for females' gender disadvantage and insufficient prior-knowledge.