We construct a student model in the algebra field. For gathering the data, we have used the Aplusix learning environment that allows students to make freely calculation steps and records all the students’actions. In a problem-solving environment, one way to build and update the student model is to precisely follow what the student is doing, by means of a detailed representation of cognitive skills. The challenge is therefore to reconstruct high-level behaviors from this low-level data. We try to discover patterns of student behaviors. For that, we use the statistical implicative analysis which makes it possible to seek the stability of the actions and to determine the contexts where they appear. Implicative analysis allows us to build implicative connections between contexts towards actions.