Do instructional strategies considering activity emotions reduce students’ boredom in a computerized open-ended learning environment?

Providing students with efficient instruction tailored to their individual characteristics in the cognitive and affective domains is an important goal in research on computer-based learning. This is especially important when seeking to enhance students' learning experience, such as by counteracting boredom, a detrimental emotion for learning. However, studies comparing instructional strategies triggered by either cognitive or emotional characteristics are surprisingly scarce. In addition, little research has examined the impact of these types of instructional strategies on performance and boredom trajectories within a lesson. In the present study, we compared the effectiveness of an intelligent tutoring system that adapted variable levels of hint details to a combination of students' dynamic, self-reported emotions and task performance (i.e., the experimental condition) to a traditional hint delivery approach consisting of a progressive, incremental supply of details following students' failures (i.e., the control condition). Linear mixed models of time-related changes in task performance and the intensity of boredom over two 1-h sessions showed that students (N = 104) in the two conditions exhibited equivalent progression in task performance and similar trajectories in boredom intensity. However, a consideration of students' achievement levels in the analyses (i.e., their final performance on the task) revealed that higher achievers in the experimental condition showed a reduction in boredom during the first session, suggesting possible benefits of using emotional information to increase the contingency of the hint delivery strategy and improve students’ learning experience.