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.

Large language models (LLMs) have emerged as powerful tools for developing educational systems.While previous studies have explored modeling student mistakes, a critical gap remains in understanding whether LLMs can generate correct solutions that represent student responses to free-response problems.We compare the distribution of solutions from four LLMs (one proprietary, two open-sourced general, and one open-sourced math models) with various sampling and prompting techniques and those from students teaching math problems to a conversational robot.Our study reveals discrepancies between the correct solutions produced by LLMs and by students.We discuss the practical implications of these findings for the design and evaluation of LLMsupported educational systems.

The intersection of language and identity refers to the ways in which language use reflects and shapes one's sense of self and group membership. It is important to design voice-based technologies that represent and support diverse types of identities to increase accessibility. Due to constraints in the data and algorithms used to train and improve artificial intelligence (AI) systems like social robots or voice assistants, they are often unable to represent certain dialects of English and other languages. This could result in data bias, identity exclusion, and communication barriers that can impact the user's outcomes when using such dialects. Technology companies are discussing how to close this gap by improving training data; however, some researchers believe that the intersection of identity and technology calls for effective collaborative design methodologies that involve intended users. This work-in-progress paper presents preliminary findings on a participatory design method, comic-boarding, that can accomplish the above goal. Using comic-boarding, our intention is to create a space for critical reflection on language and identity in human-human and human-agent dialogue for children from marginalized communities.

Robot technologies have been introduced to computing education to engage learners. This study introduces the concept of co-creation with a robot agent into culturally-responsive computing (CRC). Co-creation with computer agents has previously focused on creating external artifacts. Our work differs by making the robot agent itself the co-created product. Through participatory design activities, we positioned adolescent girls and an agentic social robot as co-creators of the robot’s identity. Taking a thematic analysis approach, we examined how girls embody the role of creator and co-creator in this space. We identified themes surrounding who has the power to make decisions, what decisions are made, and how to maintain social relationship. Our findings suggest that co-creation with robot technology is a promising implementation vehicle for realizing CRC.

Complex texts like poetry are distinct from informative texts, requiring additional subprocesses to decode and interpret. Approaching a poem without knowledge of these cognitive strategies can result in confusion and frustration—rather than comprehension. In this work, we explore how interfaces can surface and demonstrate these cognitive processes to novice readers. We introduce Soliloquy, an interface that visualizes the thoughts of an expert as they read and interpret a poem by using animations of text and pop-up tooltips. We evaluate the interface in a five-condition Mechanical Turk study (n=254) by varying the detail of thoughts, including audio, and substituting a static text control. Our study detected a significant difference in comprehension between the detail of thoughts, but not between the Soliloquy interface and static text control. We further investigate this finding in a think-aloud study (n=13), revealing the impact individual differences, experience, and cognitive load could have on Soliloquy’s effectiveness.

Student motivation is a critical factor during collaboration because the social and environmental factors within the collaborative context may influence participation. We used students' self-reported motivational values to provide collaborative support in a cross-platform learning environment. The support used a combination of displaying prompts to facilitate digital collaboration and rewarding badges to reinforce positive aspects of student participation. A classroom study (n=52) was conducted to examine the effect of this support on participation and learning. We discuss the implications of collaborative support using the data collected from multiple digital collaborative activities.

Computing education is important for K-12 learners, but not all learners resonate with common educational practices. Culturally responsive computing initiatives center and empower learners from diverse and historically excluded backgrounds. Recently, a number of educational programs have been developed and curated for an online experience. In this paper, we describe an online synchronous culturally responsive computing (CRC) camp for middle school girls (ages 11-14 years old) and report on challenges and successes from running the camp curriculum four times over the course of a year. We also describe core iterative changes we made between our runs. We then discuss lessons learned related to building rapport and connection among learners, centering learners of different backgrounds in an online synchronous environment, and facilitating reflection on power and identity aimed at positioning learners as techno-social change agents. Lastly, we offer recommendations for running online CRC experiences.

Computer science (CS) education is an important subject for K-12 students in an increasingly computational world. However, common CS education practices may not be inclusive of all learners. Culturally responsive computing (CRC) initiatives aim to center and empower learners from diverse and historically excluded backgrounds. With a sudden shift to online learning, virtual educational experiences have been developed. We describe three main findings from running three iterations of an online synchronous CRC camp for middle school girls, which are: (1) Integration of power, identity, and CS concepts, (2) Participation in CS vs. power and identity activities based on learners' backgrounds, and (3) Adjusting instructor expectations about learner engagement to be more open-ended.

Yuya Asano, Diane Litman, Mingzhi Yu, Nikki Lobczowski, Timothy Nokes-Malach, Adriana Kovashka, Erin Walker. Proceedings of the 23rd Annual Meeting of the Special Interest Group on Discourse and Dialogue. 2022.

Culturally responsive STEM and computing initiatives aim to engage and embolden a diverse range of learners, center their identity and experiences in curriculum, and connect learners to each other and their communities. With an abrupt pivot to online learning at the beginning of 2020, more educational experiences have taken place virtually. We ran a virtual synchronous culturally responsive computing camp and saw that establishing the right environment online to support a good sense of connectedness was challenging. To investigate this further, we interviewed eight K-12 instructors of culturally responsive STEM and computing programs. Three themes emerged on defining and cultivating connectedness in learning experiences, the role of equity in supporting community online, and affordances of being online specific to culturally responsive perspectives. We support our thematic findings with vignettes from the camp data. In this study, we address K-12 culturally responsive STEM and computing instructors' beliefs, experiences, and approaches regarding cultivating connectedness online. This work fills a gap in understanding instructor perspectives on building in-program and broader community connections online from a culturally responsive STEM and computing lens.

Previous research has demonstrated the benefits of applying comparative strategies while learning from informational texts, where students identify key concepts and then attempt to establish relationships between those concepts. Concept mapping is one activity that can prompt students to use comparative strategies, but not all students benefit from this activity without support. This work presents an intelligent tutoring system for concept mapping that facilitates the development of comparative strategies through diagnostic feedback that responds to the quality of students’ concept mapping process and map correctness. The novelty of the system lies in the combination of outcome-based feedback methods typical in concept mapping with adaptive process-based evaluation. In a lab study with 46 college students, we evaluate the effect of this combined adaptive support compared to solely process-based support and no support. Results suggested that the combined feedback approach showed promise at improving students’ use of comparative strategies and learning outcomes.

Dialog with a social pedagogical robot or agent is a powerful way for kids to learn [1, 5] but may limit the formation of an agentic relationship with the technology [9]. One main purpose of conversational agents is to allow the user to have a natural interaction that reduces the need to learn artificial conventions [6], but dialog systems fall short with respect to failure recovery, vocabulary diversity, remembering conversational history, and other measures [2, 3]. Further, Hill et. al. [4] found that people adapt their model of communication to match a chatbot’s in the same way they do with a child or non-native speaker. Thus, users conversing with a pedagogical agent are implicitly trained to shape their behavior to suit the technology rather than shaping the technology. For young learners, particularly among populations that have been historically excluded from technology fields, this limits agency and reinforces marginalizing power structures [9].

Online crowds, with their large numbers and diversity, show great potential for creativity. Research has explored different ways of augmenting their creative performance, particularly during large-scale brainstorming sessions. Traditionally, this comes in the form of showing ideators some form of inspiration to get them to explore more categories or generate more and better ideas. The mechanisms used to select which inspirations are shown to ideators thus far have not taken into consideration ideators' individualities, which could hinder the effectiveness of support. In this paper, we introduce and evaluate CrowdMuse, a novel adaptive system for supporting large-scale brainstorming. The system models ideators based on their past ideas and adapts the system views and inspiration mechanism accordingly. We evaluate CrowdMuse over two iterative large online studies and discuss the implication of our findings for designing adaptive creativity support systems.

While prior research has revealed the promising impact of concept mapping on learning, few have comprehensively modeled different cognitive behaviors during concept mapping. In addition, existing concept mapping tools lack effective feedback to support better learning behaviors. This work presents MindDot, a concept map-based learning environment that facilitates the cognitive process of comparing and integrating related concepts via two forms of support. A hyperlink support and an expert template. Study results suggested that both types of support had positive impact on the development of comparative strategies and that hyperlink support enhanced learning. We further evaluated the cognitive learning progress at a fine-grained level with two forms of visualizations. We then extracted several behavioral patterns that provided insights about the cognitive progress in learning. Lastly, we derive design recommendations that we hope will inspire future intelligent tutoring systems that automatically evaluate students' learning behaviors and foster them in developing effective learning behaviors

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.

Research has explored different ways of improving crowd ideation, such as presenting examples or employing facilitators. While such support is usually generated through peripheral tasks delegated to crowd workers who are not part of the ideation, it is possible that the ideators themselves could benefit from the extra thought involved in doing them. Therefore, we iterate over an ideation system in which ideators can perform one of three peripheral tasks (rating originality and usefulness, similarity, or idea combination) on demand. In controlled experiments with workers on Mechanical Turk, we compare the effects of these secondary tasks to simple idea exposure or no support at all, examining usage of the inspirations, fluency, breadth, and depth of ideas generated. We find tasks to be as good or better than exposure, although this depends on the period of ideation and the fluency level. We also discuss implications of inspiration size, homogeneity, and frequency.

Desirable outcomes such as health are tightly linked to behaviors, thus inspiring research on technologies that support people in changing those behaviors. Many behavior-change technologies are designed by HCI experts but this approach can make it difficult to personalize support to each user's unique goals and needs. This paper reports on the iterative design of two complementary support strategies for helping users create their own personalized behavior-change plans via self-experimentation: One emphasized the use of interactive instructional materials, and the other additionally introduced context-aware computing to enable user creation of "just in time" home-based interventions. In a formative trial with 27 users, we compared these two approaches to an unstructured sleep education control. Results suggest great promise in both strategies and provide insights on how to develop personalized behavior-change technologies.

As technology is increasingly integrated into the classroom, understanding the facilitators and barriers for deployment becomes an important part of the process. While systems that employ traditional WIMP-based interfaces have a well-established body of work describing their integration into classroom environments, more novel technologies generally lack such a foundation to guide their advancement. In this paper we present Robo-Tangible Activities for Geometry (rTAG), a tangible learning environment that utilizes a teachable agent framing, together with a physical robotic agent. We describe its deployment in a school environment, qualitatively analyzing how teachers chose to orchestrate its use, the value they saw in it, and the barriers they faced while organizing the sessions with their students. Based on this analysis, we extract four recommendations that aid in designing and deploying systems that make use of affordances that are similar to those of the rTAG system.

With a growing number of applications involving social human-robot interactions, there is an increasingly important role for socially responsive speech interfaces that can effectively engage the user. For example, learning companions provide both task-related feedback and motivational support for students with the goal of improving learning. As a learning companion's ability to be socially responsive increases, so do learning outcomes. This paper presents a socially responsive speech interface for an embodied, robotic learning companion. We explore two methods of social responsiveness. The first method introduces social responses into the dialogue, while the second method augments these responses with voice-adaptation based on acoustic-prosodic entrainment. We evaluate the effect of a social, voice-adaptive robotic learning companion on social variables such as social presence and rapport, and we compare this to a companion with only social dialogue and one with neither social dialogue nor voice-adaptions. We contrast the effects against those of individual factors, such as gender. We find (1) that social presence is significantly higher with a social voice-adaptive speech interface than with purely social dialogue, and (2) that females feel significantly more rapport and are significantly more persistent in interactions with a robotic learning companion than males.

A teachable agent is a learning companion that students teach about a domain they are trying to master. While most teachable agents have been virtual, there may be advantages to having students teach an agent with a physical form (i.e., a robot). The robot may better engage students in the learning activity, and if students take embodied action in order to instruct the robot, they may develop deeper knowledge. In this paper, we investigate these two hypotheses using the rTAG system, a teachable robot for geometry learning. In a study with 37 4th-6th grade participants, we compare rTAG to two other conditions, one where students use embodied action to teach a virtual agent, and one where students teach a virtual agent on a personal computer. We find that while there are no significant learning differences between conditions, students' perceptions of the agent are influenced by condition and prior knowledge.

Observed frequently in human-human interactions, entrainment is a social phenomenon in which speakers become more like each other over the course of a conversation. Acoustic-prosodic entrainment occurs when individuals adapt their acoustic-prosodic speech features, such as pitch and intensity. Correlated with communicative success, naturalness, and conversational flow as well as social variables such as rapport, a dialogue system which automatically entrains has the potential to improve verbal interactions by increasing rapport, naturalness, and conversational flow. In an application like the learning companion, such a socially responsive dialogue system may improve learning and motivation. However, it is not clear how to produce entrainment in an automatic dialogue system in ways that produce the effects seen in human-human dialogue. In this paper, we take the first steps towards implementing a spoken dialogue system which can entrain. We propose three methods of pitch adaptation based on analysis of human entrainment, and design and implement a system which can manipulate the pitch of text-to-speech output adaptively. We find a clear relationship between perceptions of rapport and different forms of pitch adaptations. Certain adaptations are perceived as significantly more natural and rapport-like. Ultimately, adapting by shifting the pitch contour of the text-to-speech output by the mean pitch of the user results in the highest reported measures of rapport and naturalness.

End-user programming tools, if properly designed, have the potential to empower end-users to create context-aware applications tailored to their own needs and lives, in order to help them break bad habits and change their behaviors. In this work, we present GALLAG Strip, an easy to use mobile and tangible tool that allows users to create context-aware applications without the need of programming experience. It enables programming by physical demonstration of envisioned interactions with the same sensors and objects that users will later encounter in their finished application. After an initial pilot to verify the usability of GALLAG Strip, we conducted a user study to evaluate the effects of tangible programming in terms of ease of use, engagement, and facilitation of the ideation process. We found that tangibility has both benefits and drawbacks, and suggest a mixed tangible and non-tangible approach for better user experience.

Technology has the promise to transform educational prac-tices worldwide. In particular, cognitive tutoring systems are an example of educational technology that has been ex-tremely effective at improving mathematics learning over traditional classroom instruction. However, studies on the effectiveness of tutor software have been conducted mainly in the United States, Canada, and Western Europe, and little is known about how these systems might be used in other contexts with differing classroom practices and values. To understand this question, we studied the usage of mathematics tutoring software for middle school at sites in three Latin American countries: Brazil, Mexico, and Costa Rica. While cognitive tutors were designed for individual use, we found that students in these classrooms worked collaboratively, engaging in interdependently paced work and conducting work away from their own computer. In this paper we present design recommendations for how cognitive tutors might be incorporated into different classroom practices, and better adapted for student needs in these environments.

Giving effective help is an important collaborative skill that leads to improved learning for both the help-giver and help-receiver. Adding intelligent tutoring to student interaction may be one effective way of assisting students in giving and receiving better help. However, such systems have proven difficult to implement, in part due to the challenges of modeling productive dialogue in a collaborative activity. We present a theoretical model of good helping behavior in a peer tutoring context, and validate the model using student tutoring data, linking optimal and buggy behaviors to learning outcomes. We discuss the implications of the model with respect to providing intelligent tutoring for peer tutoring.