Enhancing the Effectiveness of Digital Game-Based Learning with Adaptive Instructional Support

Abstract

In an ideal world, education would offer every student an equal opportunity to fulfill their aspirations. The influences of individual differences would be drastically reduced if the instruction could provide individualized learning experiences that meet the unique needs and preferences of each student. Nonetheless, incorporating such instructional practices can cause extensive resource demands and practical challenges in traditional classroom teaching. Consequently, scholars have proposed the implementation of adaptation in digital game-based learning as a potential solution to provide instruction tailored to individual needs.

The overall aim of this dissertation was to investigate the impacts of adaptive instructional support on the effectiveness of digital game-based learning. Three interventions were conducted. Data were gathered from over 400 Finnish students in grades 4–7 who were studying rational numbers using the Number Trace learning game. Value-added study designs were utilized to examine the specific motivational and learning effects of emotionally designed scaffolding and the motivational effects of task difficulty adaptation. The collected quantitative data consisted of pre-and posttest, repeated single-item in-game self-reported measurements of motivational outcomes, in-game metrics, and motivational trait measurements. In-game measurements of situational interest, self-efficacy, and perceived difficulty were employed multiple times during the learning activity.

The results revealed that the employed in-game measurements of situational interest and self-efficacy were positively related to learning outcomes. It was found that the substantial downward adaptation of task difficulty resulted in significant increases in situational interest, and substantial upward adaptation of task difficulty significantly decreased situational interest. Only significant changes in task difficulty affected situational interest and perceived difficulty, and these changes occurred within a relatively short time frame. Additionally, it was discovered that the effectiveness of adaptive instructional support can be enhanced with emotional design. That is, emotionally designed scaffolding elicited significantly higher levels of situational interest and self-efficacy compared to scaffolding presented with emotionally neutral mathematical hatch marks. Furthermore, it was found that, a game-based learning environment incorporating scaffolding yielded substantial learning effects. However, the visualization of scaffolding had no impact on learning outcomes.

This dissertation demonstrated that the effectiveness of game-based learning can be enhanced with adaptive instructional support. Overall, these findings advance our understanding of the psychological and instructional mechanics affecting human learning. These advances contribute to motivational theory development and to the development of models of adaptive and game-based learning. In addition, the findings of this dissertation advance our methodological understanding by indicating that a comprehensive evaluation of the effectiveness of adaptive instructional support requires an examination of the strength and direction of difficulty adaptation and the use of repeated measurements of motivational outcomes during the learning activity. This work demonstrated that single-item in-game self-reported measurements could be used to accomplish this.