PhD Dissertation Proposal: Andre Kenneth Chase Randall, Adaptive System Behavior in Technology Enhanced Learning Environments: A Comparative Study across Indonesia, Thailand and Vietnam
Content
Speaker:
Abstract:
Adaptive and technology enhanced learning systems are increasingly used to support learning, but many are designed within Western instructional contexts and assume that feedback, participation, and reasoning transfer across classrooms in the same way. As institutions deploy these systems across different educational environments, differences in language, curriculum structure, classroom practice, and institutional expectations raise a central question: do adaptive learning systems remain meaningful and aligned when deployed across instructional contexts? In this dissertation, I examine this problem as a question of contextual validity, focusing on how instructional assumptions baked into adaptive and curriculum aware systems influence learner interaction, trust, reasoning, and interpretation. This dissertation examines the phenomenon of contextual mismatch in adaptive learning systems and the mechanisms through which instructional context influences system behavior, interpretation, trust, and alignment. This dissertation evaluates curriculum aware adaptive tutoring systems and computational learning platforms identified through institutional collaboration and field deployment opportunities across Indonesia, Thailand, and Vietnam.
This dissertation asks four related questions. First, how do students and teachers interpret adaptive feedback, explanations, and recommendations across instructional settings in Indonesia, Thailand, and Vietnam? Second, how do cultural, curricular, instructional, and language differences influence adaptive system behavior, trust, and alignment with classroom practice? Third, when institutions deploy adaptive systems across instructional contexts, do failures reflect limitations in the system design, or instructional assumptions that fail to generalize across environments? Fourth, how can these findings guide the design of curriculum aware adaptive systems across different educational contexts? Together, these questions position the dissertation as a study of adaptive system behavior in real instructional settings rather than only a study of classroom practice.
Before international fieldwork begins, I incorporate customer discovery methods through a forthcoming application to the U.S. National Science Foundation Innovation Corps (I-Corps™) program to support early validation of stakeholder needs, instructional challenges, and assumptions about adaptive and technology enhanced learning systems across institutions in the United States. The U.S. National Science Foundation Innovation Corps (I-Corps™) program emphasizes customer discovery and stakeholder validation to assess how technologies operate beyond controlled laboratory settings. In this dissertation, these methods support the early identification of instructional challenges, deployment assumptions, and institutional needs before international fieldwork begins. This work builds on prior customer discovery training completed through the NSF I Corps Hub for the Interior Northeast led by Cornell University and supports a forthcoming national application through the Massachusetts Institute of Technology. This phase includes over 100 customer discovery interviews with students, instructors, administrators, and education technology stakeholders to identify system level challenges, deployment assumptions, and instructional needs across institutional environments.
The dissertation then transitions to comparative international fieldwork in Indonesia and Thailand, supported by dissertation fieldwork funding awarded by the UMass Graduate School. These comparative fieldwork phases examine how adaptive learning systems operate across different instructional and institutional environments. Building on this comparative analysis, the dissertation extends into a deeper fieldwork phase in Vietnam supported through the Fulbright U.S. Student Program. This extended phase supports deeper analysis of adaptive system behavior, instructional alignment, and contextual validity within a single instructional environment over time.
Using interviews, classroom observations, curriculum reviews, and small group discussions, I examine how students and teachers interpret adaptive feedback, recommendations, and instructional support provided through computational learning systems. I analyze how cultural, curricular, instructional, and institutional factors influence system use, including when feedback is trusted, ignored, misunderstood, or misaligned with classroom expectations.
I also examine how instructional context influences adaptive system behavior across Indonesia, Thailand, and Vietnam. I compare patterns across sites to distinguish between failures caused by system design limitations and failures caused by instructional assumptions that do not align with local classroom practice. This analysis helps identify when adaptive systems fail because of technical limitations and when they fail because their assumptions do not generalize across instructional environments.
This dissertation contributes to AI in Education and technology enhanced learning by providing empirical evidence on how AI tools and adaptive learning systems operate across different cultural and instructional settings. It also contributes a framework for interpreting adaptive system behavior by distinguishing correctness from alignment with the instructional context. This dissertation further argues that curriculum learning is not a monolithic property of instruction or training, but a set of interacting structural and contextual dimensions shaped by the learning environment. Overall, this dissertation uses a system, task, environment, and phenomenon framework to study how instructional context influences the behavior, interpretation, alignment, and deployment of adaptive computational systems in educational settings. These findings help guide the design of adaptive learning systems that support learning while keeping teachers and learners engaged in the instructional process.
Advisor:
Beverly Woolf