The paper concerns the problem of visual support of decision-making in intelligent tutoring systems and generating natural-language explanations of these decisions. A model of interpreting a learning situation and decision making about further learning on operational, tactical, and strategic levels considered from topic, competency, and learning-goal points of view. We show a case study of learning-situation visualization of analysis and synthesis of explanatory feedback for making operational, tactical, and strategic decisions. The method was evaluated with 3 groups of graduate students; the groups that received simplified Cognitive Maps of Knowledge Diagnosis along with textual explanations demonstrated higher trust in the system’s decisions and were more likely to follow the recommendations. We conclude with recommendations for using cognitive maps of knowledge diagnosis for cross-cutting analysis of learning situations.