Erin Walker

Yuya Asano, Diane Litman, Mingzhi Yu et al.

Speakers build rapport in the process of aligning conversational behaviors with each other. Rapport engendered with a teachable agent while instructing domain material has been shown to promote learning. Past work on lexical alignment in the field of education suffers from limitations in both the measures used to quantify alignment and the types of interactions in which alignment with agents has been studied. In this paper, we apply alignment measures based on a data-driven notion of shared expressions (possibly composed of multiple words) and compare alignment in one-on-one human-robot (H-R) interactions with the H-R portions of collaborative human-human-robot (H-H-R) interactions. We find that students in the H-R setting align with a teachable robot more than in the H-H-R setting and that the relationship between lexical alignment and rapport is more complex than what is predicted by previous theoretical and empirical work.

Yuya Asano, Diane Litman, Mingzhi Yu et al.

While speech-enabled teachable agents have some advantages over typing-based ones, they are vulnerable to errors stemming from misrecognition by automatic speech recognition (ASR). These errors may propagate, resulting in unexpected changes in the flow of conversation. We analyzed how such changes are linked with learning gains and learners' rapport with the agents. Our results show they are not related to learning gains or rapport, regardless of the types of responses the agents should have returned given the correct input from learners without ASR errors. We also discuss the implications for optimal error-recovery policies for teachable agents that can be drawn from these findings.

Paras Sharma, YuePing Sha, Janet Shufor Bih Epse Fofang et al.

Dialogue systems have long supported learner reflections, with theoretically grounded, rule-based designs offering structured scaffolding but often struggling to respond to shifts in engagement. Large Language Models (LLMs), in contrast, can generate context-sensitive responses but are not informed by decades of research on how learning interactions should be structured, raising questions about their alignment with pedagogical theories. This paper presents a hybrid dialogue system that embeds LLM responsiveness within a theory-aligned, rule-based framework to support learner reflections in a culturally responsive robotics summer camp. The rule-based structure grounds dialogue in self-regulated learning theory, while the LLM decides when and how to prompt deeper reflections, responding to evolving conversation context. We analyze themes across dialogues to explore how our hybrid system shaped learner reflections. Our findings indicate that LLM-embedded dialogues supported richer learner reflections on goals and activities, but also introduced challenges due to repetitiveness and misalignment in prompts, reducing engagement.

Saki Imai, Lee Kezar, Laurel Aichler et al.

Most state-of-the-art sign language models are trained on interpreter or isolated vocabulary data, which overlooks the variability that characterizes natural dialogue. However, human communication dynamically adapts to contexts and interlocutors through spatiotemporal changes and articulation style. This specifically manifests itself in educational settings, where novel vocabularies are used by teachers, and students. To address this gap, we collect a motion capture dataset of American Sign Language (ASL) STEM (Science, Technology, Engineering, and Mathematics) dialogue that enables quantitative comparison between dyadic interactive signing, solo signed lecture, and interpreted articles. Using continuous kinematic features, we disentangle dialogue-specific entrainment from individual effort reduction and show spatiotemporal changes across repeated mentions of STEM terms. On average, dialogue signs are 24.6%-44.6% shorter in duration than the isolated signs, and show significant reductions absent in monologue contexts. Finally, we evaluate sign embedding models on their ability to recognize STEM signs and approximate how entrained the participants become over time. Our study bridges linguistic analysis and computational modeling to understand how pragmatics shape sign articulation and its representation in sign language technologies.

Amanda Buddemeyer, Erin Walker, Malihe Alikhani

Many educational technologies use artificial intelligence (AI) that presents generated or produced language to the learner. We contend that all language, including all AI communication, encodes information about the identity of the human or humans who contributed to crafting the language. With AI communication, however, the user may index identity information that does not match the source. This can lead to representational harms if language associated with one cultural group is presented as "standard" or "neutral", if the language advantages one group over another, or if the language reinforces negative stereotypes. In this work, we discuss a case study using a Visual Question Generation (VQG) task involving gathering crowdsourced data from targeted demographic groups. Generated questions will be presented to human evaluators to understand how they index the identity behind the language, whether and how they perceive any representational harms, and how they would ideally address any such harms caused by AI communication. We reflect on the educational applications of this work as well as the implications for equality, diversity, and inclusion (EDI).

John Frens, Erin Walker, Gary Hsieh

Prior research has examined the use of Social Question and Answer (Q&A) websites for answer and help seeking. However, the potential for these websites to support domain learning has not yet been realized. Helping users write effective answers can be beneficial for subject area learning for both answerers and the recipients of answers. In this study, we examine the utility of crowdsourced, criteria-based feedback for answerers on a student-centered Q&A website, Brainly.com. In an experiment with 55 users, we compared perceptions of the current rating system against two feedback designs with explicit criteria (Appropriate, Understandable, and Generalizable). Contrary to our hypotheses, answerers disagreed with and rejected the criteria-based feedback. Although the criteria aligned with answerers' goals, and crowdsourced ratings were found to be objectively accurate, the norms and expectations for answers on Brainly conflicted with our design. We conclude with implications for the design of feedback in social Q&A.