Transitions of Affective States in an Intelligent Tutoring System

Document Type


Publication Date



This paper aims to determine the natural transitions that take place among students‘ affective states while using SQL-Tutor, a constraint-based intelligent tutor that teaches Structured Query Language (SQL). Intelligent tutoring systems, such as SQL-Tutor, have been used by various institutions to overcome the difficulties of students in learning difficult subject matter. In this paper, the researchers intend to determine how an affective state at a given time influences the affective state at a later time. Likely transitions of affective states were mined from the observation files gathered from students during interaction with the tutor. The researchers were able to identify several affective transitions that were unlikely to occur. Identifying these natural transitions is a stepping stone towards designing interventions that might help sustain desirable affective states, encourage productive affective transitions and mitigate states and transitions that may be deemed detrimental to learning.