Impact of Previous Understandings on Gaze Points and Order of Cognitive and Reflective Self-Regulated Learning Procedures during Education with an Artificial Tutor System
================================================================================
A recent study has shed light on the crucial role that prior knowledge plays in the design of adaptive intelligent tutoring systems (ITS). The research, conducted using the MetaTutor system, focused on college students learning about the human circulatory system.
The study found no significant differences in fixations on single areas of interest by the prior knowledge group students were assigned to. However, it did uncover significant differences in fixations on pairs of areas of interest and sequential patterns of engaging in self-regulated learning processes.
Students with high prior knowledge were found to engage in processes containing both cognitive strategies and metacognitive strategies, while students with low prior knowledge did not. These differences were evident in the sequential patterns of engaging in cognitive and metacognitive self-regulated learning processes, as evidenced from log-file data.
The findings demonstrate the potential value of using a combination of eye-tracking and log-file data to understand the impact of prior knowledge on self-regulated learning. The study's results suggest that focusing on the interactions between different areas of interest may be more effective in promoting self-regulated learning than focusing on individual areas.
The study's key implications for the design of ITS include dynamic student modeling, monitoring fixations and learning sequences, adaptive pathways, emotionally intelligent ITS, and the use of large language models (LLMs) and AI.
Dynamic student modeling enables ITS to match instructional content within the learner’s zone of proximal development, maintaining an optimal balance of task difficulty that is neither too easy nor too frustrating, which fosters engagement and deeper learning. ITS should monitor fixations and learning sequences to infer learners’ self-regulation strategies, supporting adaptive content presentation and scaffolding for more effective cognitive and metacognitive regulation tailored to the learner’s current knowledge level.
Adaptive pathways that respond to differences in prior knowledge and observed self-regulated learning patterns can enhance motivation and autonomy by providing personalized pacing, problem complexity, and feedback styles, avoiding one-size-fits-all approaches. Emotionally intelligent ITS that integrate cognitive and affective adaptation recognize that prior knowledge affects not only cognition but also emotional states during learning. Responsive emotional support combined with cognitive adaptation improves learner persistence and academic outcomes.
The usage of large language models and AI can facilitate more natural dialogues that dynamically interpret student input, providing elaborations or revisiting concepts based on the student’s knowledge profile and learning sequence, hence supporting personalized scaffolding.
In sum, effective ITS design for college students requires integrating real-time assessments of prior knowledge with analysis of students’ fixation patterns and self-regulated learning sequences to create flexible, personalized, and emotionally supportive learning experiences that optimize both engagement and academic performance. This approach addresses the diverse backgrounds and learning habits typical of higher education learners and fosters sustained, autonomous learning.
The study's findings contribute to the growing body of evidence supporting the use of adaptive intelligent tutoring systems in promoting self-regulated learning in college students. The research involved 194 college students, with 30 participating in eye-tracking and sequence mining analyses.
- The integration of eye-tracking technology in health-and-wellness research, as demonstrated in the MetaTutor study, could significantly contribute to the science of learning by providing insights into self-regulated learning processes, especially in the context of education-and-self-development.
- In the realm of technology and science, adaptive intelligent tutoring systems (ITS) could potentially revolutionize the way health-and-wellness education is delivered, by offering personalized and adaptive learning paths based on prior knowledge and self-regulated learning patterns, thereby fostering deep understanding and autonomous learning.