This paper compiles a cognitive attribute diagnostic test paper, counts students’ mastery of cognitive attributes of college English course Civics and compares it with the ideal mastery pattern, completes the categorization of cognitive attribute mastery patterns, and constructs a hierarchical learning model. The Q-learning algorithm is used to solve the value function of different cognitive attribute learning paths to determine the learning path that has the best effect in improving the value of learning ability, and to assist teachers in improving the quality of Civics and Politics blended teaching in college English courses. The cognitive attribute mastery patterns of 200 students are 100% compatible with the 19 ideal mastery patterns, and the Civics and Politics cognition of students is in a reasonable range. Students’ cognitive attribute learning ability and Civics level.
