The use of dual eye-tracking is investigated in a collaborative game setting. The automatic collection of information about partner’s gaze will eventually serve to build adaptive interfaces. Following this agenda, and in order to identify stable gaze patterns, we investigate the impact of social and task related context upon individual gaze and action during a collaborative Tetris game. Results show that experts as well as novices adapt their playing style when interacting in mixed ability pairs. We also present machine learning results about the prediction of player’s social context.