This paper introduces an open learner model approach that enhances transparency by allowing students to engage in a negotiation process with the system, collaboratively improving the model’s accuracy, especially working within an ITS for propositional logic, particularly supporting students in theorem-proof using inference rules, deriving proofs. By enabling students to review and discuss discrepancies in their learner model, the system reduces cognitive conflicts that arise from misalignment between the student’s self-perception and the system’s assessment. We analyze student-system interactions in this logic tutoring environment and demonstrate that negotiation-driven learner modeling enhances the accuracy of the system’s understanding of student knowledge, potentially allowing metacognitive engagement and self-regulation. Our findings highlight the potential of explainable and interactive learner models to improve the effectiveness of ITSs, particularly in domains requiring logical reasoning and structured problem-solving.