The Learning Code : Designing AI-Driven Adaptive Learning Systems for Social Learning
Open Access
- Author:
- Gautam, Sanjana
- Graduate Program:
- Informatics
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 25, 2024
- Committee Members:
- Dongwon Lee, Professor in Charge/Director of Graduate Studies
Jack Carroll, Major Field Member
Heather Zimmerman, Outside Unit & Field Member
Kenneth Huang, Major Field Member
Mary Beth Rosson, Chair & Dissertation Advisor - Keywords:
- education and learning
adaptive learning
social learning
personal data tracking
artificial intelligence
collaborative learning
system design - Abstract:
- Adaptive learning systems aim to emulate how skilled educators provide every student with the best possible learning experience. We investigate how these systems might be enriched by incorporating activities and indicators of social learning, which focus on the influences of learners’ social context and interactions. This doctoral research aims to explore and evaluate application of social learning theory within the context of an adaptive learning system such that the burden of initiation of social engagement falls on the system rather than the student. More generally, our work illustrates how learning theories can contribute to designing adaptive learning systems. This document describes three studies and a prototype that was completed as a part of achieving the goal described above. We begin with a pilot study exploring the inclusion of social learning in an adaptive system. Our analysis of the social learning scale demonstrates its validity and usefulness for ongoing work, while our qualitative analysis reveals how social learning varies among students. We discuss integrating rating scale results and observations of social learning into a student model to drive an adaptive system. With rapid advancements in learning technology, the field of technology-supported learning has been exploring optimal ways to support learning. Social learning theories propose that people learn through social interactions, making the initiation and support of collaboration in remote learning critical. We employed wizard of oz methods to mimic an adaptive learning platform that prompts students’ collaboration efforts based on their social learning dispositions. Using self-report rating scale data, reflections on collaboration prompts, and interviews, we demonstrate that students’ social learning dispositions can customize experiences during collaborative activities, enhancing engagement and performance in the classroom. Our findings highlight technology’s role in supporting social learning, including design principles for adaptive systems centered on social learning dispositions. This work also explores the application of generative artificial intelligence (AI) in designing intervention prompts within adaptive learning systems. Adaptive learning, tailored to individual student needs, marks a significant departure from traditional one-size-fits-all approaches. By leveraging generative AI, this research aims to automate personalized learning interventions to enhance student engagement and understanding. We detail the development of AI-driven prompts, including the algorithms and data inputs used to tailor interventions for different learning styles and competencies. Additionally, we address the ethical challenges of deploying generative AI in educational settings, examining concerns related to data privacy, bias, and the potential reinforcement of existing educational inequalities. Emphasizing transparency, fairness, and inclusiveness, the proposed ethical framework aims to guide the responsible use of AI technologies in education.