Search for: All records

An advantage of Large Language Models (LLMs) is their contextualization capability – providing different responses based on student inputs like solution strategy or prior discussion, to potentially better engage students than standard feedback. We present a design and evaluation of a proof-of-concept LLM application to offer students dynamic and contextualized feedback. Specifically, we augment an Online Programming Exercise bot for a college-level Cloud Computing course with ChatGPT, which offers students contextualized reflection triggers during a collaborative query optimization task in database design. We demonstrate that LLMs can be used to generate highly situated reflection triggers that incorporate details of the collaborative discussion happening in context. We discuss in depth the exploration of the design space of the triggers and their correspondence with the learning objectives as well as the impact on student learning in a pilot study with 34 students. 

Abstract. Most jobs in the digital economy require 4-year university degrees, excluding many community college students. To help these students join the digital economy, our project team is developing AI-based learning technology using a novel approach. First, we employ curriculum mapping to analyze courses and identify knowledge components (KCs) that are positioned to impact student outcomes. We triangulate our results using student learning data and expert-provided qualitative assessment. We then employ the Knowledge, Learning and Instruction framework to align KCs with individual tutoring and collaborative learning. This analysis is guiding us in developing intelligent tutors and collaborative learning technology, empirically-tested forms of AI-based learning technology, to support IT students. In this paper, we describe our innovative approach and results thus far. 

Pursuing productivity, students often adopt a divide-and-conquer strategy that undercuts collaborative learning opportunities. In this study, we introduce a task structuring and role scaffolding paradigm to create opportunities for transactive exchange in such performance-oriented tasks and experimentally compare two prompting strategies -- one designed to create a focused discussion and another to intensify transactivity -- while controlling for time on task. We find significant learning gains of each strategy when used separately, but not in tandem.