AI technologies have long-term societal implications that impact youth, prompting a need for critical AI literacy for students. While current K-12 AI curricula have increasingly integrated societal impact and ethics concepts in AI curricula, there is a need to center youth’s agency in decision-making around AI systems that impact them. In this work, we engaged 94 middle and high school art students in a Policy Design learning activity as a part of an Art and AI learning workshop. Students worked in groups to create policies around AI's use in art, considering stakeholders like artists, AI companies, and consumers. Findings revealed that students developed nuanced, actionable policies that reflected a deep understanding of AI's impact on the art ecosystem, including issues of copyright, artist compensation, and transparency. The activity empowered students to think critically about AI’s ethical implications on various systems in the AI and art ecosystem and fostered a sense of agency in shaping its future. This work demonstrates the value of integrating policy design into K-12 AI curricula, providing youth with the skills and perspectives to become informed, ethical citizens in an AI-driven world.

AI literacy initiatives in K-12 education frequently emphasize the importance of AI ethics. However, many existing AI ethics curricula offer limited opportunities for students to examine ethical dilemmas or practice ethical decision-making with real-world scenarios. Prior work suggests that games can support ethics education by providing low-stakes, playful environments for exploration and reflection. In this paper, we present Orbiting the VirtueVerse, a scenario-based ethics game designed to introduce middle-school students to virtue ethics in the context of artificial intelligence. We report findings from a research workshop with 22 students aged 13–14. Participants learned about virtue ethics, played the game, and co-designed new scenarios for a future iteration of the game. Our findings suggest that game-based ethical scenarios are a promising approach for supporting students’ ethical understanding of AI. We conclude by offering design guidelines for effectively integrating ethical scenarios into educational games for AI literacy.

The use of generative AI in K-12 schools is surging despite a lack of appropriate guidance. Although prior technical AI literacy has incorporated more ethical considerations, there is little work on teaching students a concrete, practical, actionable AI use plan for specific learning tasks. This paper presents an AI-literacy curriculum integrated in the English Language Arts (ELA) Argumentation unit that centers on "use". Using the debate topic “Does Using AI in School Cause Cognitive Debt?” for deliberation and policymaking for resolution, the curriculum provides a learning trajectory that gives students insights into appropriate AI use through research and personal experience, eventually helping them develop their own informed plan. Early pilot evidence with 15 students suggests a significant increase in developing an informed plan for AI use.

AI is increasingly framed as a solution to societal challenges, particularly educational inequities. Yet, AI systems often reproduce the very exclusionary structures they claim to address. Despite this, there is limited research on empowering minoritized youth to shape AI in ways that align with their communities’ needs. This systemic issue is especially significant for Child–Computer Interaction research, which must examine how young people critically engage with AI technologies. Drawing on data science projects, surveys, and semi-structured interviews, we document how youth develop anti-AI solutionist perspectives alongside data science practices, and how they re-imagine more justice-oriented roles for AI. Our primary contribution shows how our critical AI fluency curricula enable youth to determine when AI should be adopted or refused because they understand racial inequities as sociotechnical rather than technological problems. We position this capacity for AI refusal as essential for redistributing power under algorithmic governance.

Immigrant-origin adolescents frequently encounter identity-related stressors that undermine school belonging, yet existing identity interventions require intensive facilitation that under-resourced schools cannot sustain. This study asks what challenges shape ethnic-racial identity (ERI) support under institutional constraints, and what design features a school-ready digital intervention should provide in response. We conducted six participatory workshops with 20 K-12 school counselors via a statewide association, using reflexive thematic analysis to identify three dimensions of ERI challenge-interpersonal, internal, and structural/contextual-and a consistent gap between what evidence-based interventions require and what schools can provide. These findings directly informed SelfQuest, a nine-chapter gamified identity-support prototype. This WiP contributes a practitioner-derived challenge codebook and a traceable derivation from counselor insights to prototype features. Classroom evaluation with immigrant-origin adolescents is the next step.

Artificial intelligence (AI) is increasingly present in children’s lives, yet few tools support developmentally appropriate AI literacy for grades K-3. This work examines the role of narrative in early AI literacy by directly comparing two versions of interactive game-based digital storybooks for children ages 6-9. The “Book+” condition combined an overarching story and characters with mini-games and scaffolded AI interactions, designed to be enjoyable, provide narrative context, and to give hands-on AI experience. We compared this with a “Game” condition that included the same learning goals, mini-games, and AI interactions but replaced the narrative with primarily instructional text. Across 57 participants, both conditions elicited high engagement, but “Book+” participants showed significantly greater learning gains and higher perceived knowledge. Qualitative findings revealed that while both groups enjoyed the creative AI mini-games, “Book+” participants more frequently used AI vocabulary in responses, connected concepts to the learning context, and expressed stronger emotional connection.

Children regularly negotiate questions of authority and control in home and school life, but little is known about how they believe robots should fit into these dynamics. We conducted a 75-minute design session with 17 children (ages 6-9) to examine when robots should take, share, or defer control, and how expectations shift when robots are framed as teachers, classmates, or mentees. Children resisted robot control, particularly in adult-regulated domains and areas tied to personal skill or self-expression. They were more open to robot control in domains where they felt less competent, or where robots, perceived as less legitimate authorities than humans, could substitute for adult control. Role framing further shaped expectations: teacher robots were granted autonomy, classmate robots were expected to act as peers, and mentee robots were expected to defer. These findings show that children apply context- and role-sensitive rules when negotiating control with robots. We conclude with design considerations for robots in children's everyday lives that respect children's agency, calibrate autonomy by domain, and align behavior with children's context-sensitive expectations.

The growing impact of AI on various fields, including art, highlights the importance of integrating AI learning into art education. This work investigates whether traditional art lessons can be adapted to meaningfully incorporate AI, focusing on its application to art-making practices. We adapted a character design activity to incorporate AI at different stages, such as using AI for creating references, getting feedback, visual design, animation, and personality design. We developed a character design learning activity which was supplemented by a code notebook and a front-end character design tool. 39 middle and high school students participated in this activity during two in-person Art and AI workshops. Analysis of creative outputs, knowledge surveys, and classroom discussions showed that students showed significant shifts in their understanding of AI as a creative collaborator, their art making practice, and their confidence with using AI tools. Learners demonstrated different creative styles while adopting AI into their character design. This approach demonstrates the potential for integrating AI into art lessons and offers a scalable framework for other non-CS subjects.

Is it better to perform tennis training in a pristine indoor environment or a noisy outdoor one? To model this problem, here we investigate whether shifts in the transition probabilities between the training and testing environments in reinforcement learning problems can lead to better performance under certain conditions. We generate new Markov Decision Processes (MDPs) starting from a given MDP, by adding quantifiable, parametric noise into the transition function. We refer to this process as Noise Injection and the resulting environments as δ-environments. This process allows us to create variations of the same environment with quantitative control over noise serving as a metric of distance between environments. Conventional wisdom suggests that training and testing on the same MDP should yield the best results. In stark contrast, we observe that agents can perform better when trained on the noise-free environment and tested on the noisy δ-environments, compared to training and testing on the same δ-environments. We confirm that this finding extends beyond noise variations: it is possible to showcase the same phenomenon in ATARI game variations including varying Ghost behavior in PacMan, and Paddle behavior in Pong. We demonstrate this intriguing behavior in 60 different variations of ATARI games, including PacMan, Pong, and Breakout. We refer to this phenomenon as the Indoor-Training Effect. Code to reproduce our experiments and to implement Noise Injection.

Part of a university initiative supporting responsible AI for social empowerment and education, the project-based RAICA (Responsible AI for Computational Action) curriculum supports middle/high school learners and novice AI literacy teachers use AI creatively for good. This paper offers a rare example of design-based implementation research (DBIR) in AI education across widely varied contexts, provides fine grain implementation data that contributes to a foundation for evaluating effectiveness and expanding access. We present a novel approach to analyzing fidelity of implementation data from RAICA’s computer vision module beta-test. Twelve educators working with ~282 students across nine pilot sites in four countries used a bespoke fidelity of implementation data collection tool (pre-made comment prompts in a Google Docs version of the teacher guide) to provide 236 qualitative responses about AI literacy and responsible design activities, plus 111 ordinal ratings of embedded teacher supports. Analyses revealed that while the curriculum was generally implemented as designed, educators frequently made modifications. Although most changes produced practical insights for improved curriculum design, others helped the design team anticipate and prevent changes that could obscure learning objectives and hinder outcomes. We discuss the pedagogical, design, and research implications of these findings for effective AI teaching/learning in diverse settings.

Initial discussion of AI literacy assessment has focused on competency frameworks and learning standards rather than materials for classroom use. Responsible AI for Computational Action (RAICA), a constructionist AI curriculum for middle and high school students, includes assessment materials to support teachers with the evaluation of student AI literacy competencies in their classrooms. These materials include exit tickets used as formative assessments at the end of each lesson and both teacher and student-facing rubrics. After beta-testing a module of the curriculum with nine teachers and 282 students, we reviewed teacher usage data and feedback as well as student responses. The review process surfaced a number of improvements to the materials to better align them with classroom teaching practice. These included clarifying language and adding visual scaffolds. We present the assessment materials and iterative design process used to bridge the gap between the theoretical AI literacy competencies and their practical implementation in classrooms.

The Model AI Assignments session seeks to gather and disseminate the best assignment designs of the Artificial Intelligence (AI) Education community. Recognizing that assignments form the core of student learning experience, we here present abstracts of thirteen AI assignments from the 2025 session that are easily adoptable, playfully engaging, and flexible for a variety of instructor needs. Assignment specifications and supporting resources may be found at http://modelai.gettysburg.edu

As generative AI (GenAI) emerges as a transformative force, clear understanding of high school students' perspectives is essential for GenAI's meaningful integration in high school environments. In this work, we draw insights from a participatory design workshop where we engaged 17 high school students -- a group rarely involved in prior research in this area -- through the design of novel GenAI tools and school policies addressing their key concerns. Students identified challenges and developed solutions outlining their ideal features in GenAI tools, appropriate school use, and regulations. These centered around the problem spaces of combating bias & misinformation, tackling crime & plagiarism, preventing over-reliance on AI, and handling false accusations of academic dishonesty. Building on our participants' underrepresented perspectives, we propose new guidelines targeted at educational technology designers for development of GenAI technologies in high schools. We also argue for further incorporation of student voices in development of AI policies in their schools.

CHI ’25, Yokohama, Japan

Conversational breakdowns in close relationships are deeply shaped by personal histories and emotional context, yet most NLP research treats conflict detection as a general task, overlooking the relational dynamics that influence how messages are perceived.In this work, we leverage nonviolent communication (NVC) theory to evaluate LLMs in detecting conversational breakdowns and assessing how relationship backstory influences both human and model perception of conflicts.Given the sensitivity and scarcity of real-world datasets featuring conflict between familiar social partners with rich personal backstories, we contribute the PERSONACONFLICTS CORPUS 1 , a dataset of N = 5, 772 naturalistic simulated dialogues spanning diverse conflict scenarios between friends, family members, and romantic partners.Through a controlled human study, we annotate a subset of dialogues and obtain finegrained labels of communication breakdown types on individual turns, and assess the impact of backstory on human and model perception of conflict in conversation.We find that the polarity of relationship backstories significantly shifted human perception of communication breakdowns and impressions of the social partners, yet models struggle to meaningfully leverage those backstories in the detection task.Additionally, we find that models consistently overestimate how positively a message will make a listener feel.Our findings underscore the critical role of personalization to relationship contexts in enabling LLMs to serve as effective mediators in human communication for authentic connection.

This research designs and applies inclusive child-robot interactions for collaborative creativity, where elementary school children and a social robot collaboratively create and publish picture stories. The robot offers creativity scaffolding during parts of the creative process of storytelling through social interactions such as feedback, question asking, divergent thinking, and positive reinforcement. The collaborative tasks and robot interactions are personalized for neurodivergent children's unique needs. Through a five-session user study with 32 children (ages 5–9) over 8 months, we investigate the impact of the social robot on children's exhibited creativity in storytelling over time, their creative interactions with the robot, and their perceptions of the robot as a creative collaborator. Our research revealed that inclusive design practices eliminated creative barriers for children with neurodevelopmental disorders. The robot's creativity scaffolding interactions positively influenced children's verbal creativity in storytelling, and had an influence on their storytelling creative process. After multiple sessions interacting with the robot, we observed the emergence of diverse creator styles among neurodivergent learners. We propose Inclusive Co-creative Child-robot Interaction (ICCRI) guidelines for fostering creativity in children, and accommodating diverse creator styles in complex, open-ended creative tasks.

Power dynamics through the lens of autonomy in human-robot interaction (HRI) has largely been considered through the robot's persuasion and its impact on users. Power found in hierarchies and relationships between users and robots has been less investigated. Through a co-design workshop with industry employees, we investigate how employees design autonomous agents (AA) for a risk management and safety task, focusing on the designed agent interactions and personalities (robots and chatbots). The qualitative analysis of the storyboards and personalities revealed how employees designed autonomy and power dynamics between the workers and AAs, mirroring power dynamics in social systems. These results revealed how power dynamics are interconnected with anthropomorphization of AAs, demonstrating what has been previously theorized by HRI scholars. Overall, this work considers autonomy and power in design representations and unifies varying conceptualizations of power in HRI to support a holistic perspective of how power is explored through HRI design.

Despite living in an increasingly connected world, social isolation is a prevalent issue today. While social robots have been explored as tools to enhance social connection through companionship, their potential as asynchronous social platforms for fostering connection towards humanity has received less attention. In this work, we introduce the design of a social support companion that facilitates the exchange of emotionally relevant stories and scaffolds reflection to enhance feelings of connection via five design dimensions. We investigate how social robots can serve as “social proxies” facilitating human stories, passing stories from other human narrators to the user. To this end, we conduct a real-world deployment of 40 robot stations in users' homes over the course of two weeks. Through thematic analysis of user interviews, we find that social proxy robots can foster connection towards other people's experiences via mechanisms such as identifying connections across stories or offering diverse perspectives. We present design guidelines from our study insights on the use of social robot systems that serve as social platforms to enhance human empathy and connection.

Vocal health plays a crucial role in peoples' lives, significantly impacting their communicative abilities and interactions. However, despite the global prevalence of voice disorders, many lack access to convenient diagnosis and treatment. This paper introduces VocalAgent, an audio large language model (LLM) to address these challenges through vocal health diagnosis. We leverage Qwen-Audio-Chat fine-tuned on three datasets collected in-situ from hospital patients, and present a multifaceted evaluation framework encompassing a safety assessment to mitigate diagnostic biases, cross-lingual performance analysis, and modality ablation studies. VocalAgent demonstrates superior accuracy on voice disorder classification compared to state-of-the-art baselines. Its LLM-based method offers a scalable solution for broader adoption of health diagnostics, while underscoring the importance of ethical and technical validation.

This paper reports on the creation and composition of a new corpus of children's speech, the Ohio Child Speech Corpus, which is publicly available on the Talkbank-CHILDES website. The audio corpus contains speech samples from 303 children ranging in age from 4 – 9 years old, all of whom participated in a seven-task elicitation protocol conducted in a science museum lab. In addition, an interactive social robot controlled by the researchers joined the sessions for approximately 60% of the children, and the corpus itself was collected in the peri‑pandemic period. Two analyses are reported that highlighted these last two features. One set of analyses found that the children spoke significantly more in the presence of the robot relative to its absence, but no effects of speech complexity (as measured by MLU) were found for the robot's presence. Another set of analyses compared children tested immediately post-pandemic to children tested a year later on two school-readiness tasks, an Alphabet task and a Reading Passages task. This analysis showed no negative impact on these tasks for our highly-educated sample of children just coming off of the pandemic relative to those tested later. These analyses demonstrate just two possible types of questions that this corpus could be used to investigate.

Text-to-image generation (TTIG) technologies are Artificial Intelligence (AI) algorithms that use natural language algorithms in combination with visual generative algorithms. TTIG tools have gained popularity in recent months, garnering interest from non-AI experts, including educators and K-12 students. While they have exciting creative potential when used by K-12 learners and educators for creative learning, they are also accompanied by serious ethical implications, such as data privacy, spreading misinformation, and algorithmic bias. Given the potential learning applications, social implications, and ethical concerns, we designed 6-hour learning materials to teach K-12 teachers from diverse subject expertise about the technical implementation, classroom applications, and ethical implications of TTIG algorithms. We piloted the learning materials titled “Demystify text-to-image generative tools for K-12 educators" with 30 teachers across two workshops with the goal of preparing them to teach about and use TTIG tools in their classrooms. We found that teachers demonstrated a technical, applied and ethical understanding of TTIG algorithms and successfully designed prototypes of teaching materials for their classrooms.

Generative AI tools introduce new and accessible forms of media creation for youth. They also raise ethical concerns about the generation of fake media, data protection, privacy and ownership of AI-generated art. Since generative AI is already being used in products used by youth, it is critical that they understand how these tools work and how they can be used or misused. In this work, we facilitated students’ generative AI learning through expression of their imagined future identities. We designed a learning workshop - Dreaming with AI - where students learned about the inner workings of generative AI tools, used text-to-image generation algorithms to create their imaged future dreams, reflected on the potential benefits and harms of generative AI tools and voiced their opinions about policies for the use of these tools in classrooms. In this paper, we present the learning activities and experiences of 34 high school students who engaged in our workshops. Students reached creative learning objectives by using prompt engineering to create their future dreams, gained technical knowledge by learning the abilities, limitations, text-visual mappings and applications of generative AI, and identified most potential societal benefits and harms of generative AI.

Artificial Intelligence (AI) is permeating almost every area of society, reshaping how many people, including youth, navigate the world. Despite the increased presence of AI, most people lack a baseline knowledge of how AI works. Moreover, social barriers often hinder equal access to AI courses, perpetuating disparities in participation in the field. To address this, it is crucial to design AI curricula that are effective, inclusive, and relevant, especially to learners from backgrounds that are historically excluded from working in tech. In this paper, we present AI for Wellbeing, a curriculum where students explore conversational AI and the ethical considerations around using it to promote wellbeing. We specifically designed content, educator materials, and educational technologies to meet the interests and needs of students and educators from diverse backgrounds. We piloted AI for Wellbeing in a 5-day virtual workshop with middle school teachers and students. Then, using a mixed-methods approach, we analyzed students' work and teachers' feedback. Our results suggest that the curriculum content and design effectively engaged students, enabling them to implement meaningful AI projects for wellbeing. We hope that the design of this curriculum and insights from our evaluation will inspire future efforts to create culturally relevant K-12 AI curricula.

We describe an approach for aligning an LLMbased dialogue agent for long-term social dialogue, where there is only a single global score given by the user at the end of the session.In this paper, we propose the usage of denser naturally-occurring multimodal communicative signals as local implicit feedback to improve the turn-level utterance generation.Therefore, our approach (dubbed GELI) learns a local, turn-level reward model by decomposing the human-provided Global Explicit (GE) sessionlevel reward, using Local Implicit (LI) multimodal reward signals to crossmodally shape the reward decomposition step.This decomposed reward model is then used as part of the RLHF pipeline to improve an LLM-based dialog agent.We run quantitative and qualitative human studies on two large-scale datasets to evaluate the performance of our GELI approach, and find that it shows consistent improvements across various conversational metrics compared to baseline methods.

Empathy serves as a cornerstone in enabling prosocial behaviors, and can be evoked through sharing of personal experiences in stories.While empathy is influenced by narrative content, intuitively, people respond to the way a story is told as well, through narrative style.Yet the relationship between empathy and narrative style is not fully understood.In this work, we empirically examine and quantify this relationship between style and empathy using LLMs and large-scale crowdsourcing studies.We introduce a novel, theory-based taxonomy, HEART (Human Empathy and Narrative Taxonomy) that delineates elements of narrative style that can lead to empathy with the narrator of a story.We establish the performance of LLMs in extracting narrative elements from HEART, showing that prompting with our taxonomy leads to reasonable, human-level annotations beyond what prior lexicon-based methods can do.To show empirical use of our taxonomy, we collect a dataset of empathy judgments of stories via a large-scale crowdsourcing study with N = 2, 624 participants.1 We show that narrative elements extracted via LLMs, in particular, vividness of emotions and plot volume, can elucidate the pathways by which narrative style cultivates empathy towards personal stories.Our work suggests that such models can be used for narrative analyses that lead to human-centered social and behavioral insights.1. Flatness/roundness (Keen, 2006) of the charac-

In this research-to-practice full paper, we present the third iteration of our educational framework that advances AI-informed leadership - a much-needed competency in this era of rapid AI transformation. Our study aimed to evaluate our proposed content and pedagogy and whether it can be made widely accessible to non-technical leaders. We focused on modifying our existing curriculum and research protocol based on the feedback and learnings from our previous workshops. Our current workshop is part of an AI-Education program developed by MIT aiming to provide AI education to the U.S Air Force, which is one of the largest organizations in the US. Previous two iterations of the workshop aimed to offer an introduction to AI to U.S Air Force leaders, while the current iteration's goals are twofold: offering the updated content to U.S Air Force leaders, and supporting their capacity building by preparing a cohort of workshop trainers, who will lead the future iterations of the program inhouse. We conducted the workshop with over 50 participants, including 10 facilitators who were trained to run future workshops internally. Our comprehensive educational materials, including fa-cilitator guides, student-facing documents, and digital resources, support this scalable model. The results demonstrate significant improvements in AI knowledge and leadership competencies, including AI mindset, culture, and ethos. Our findings affirm the effectiveness of experiential learning methodologies and underscore the potential for scalable, sustainable AI education within large organizations through the facilitators' feedback.

This innovative research paper presents a program that introduces entrepreneurship, innovation, and biomedical product development to engineering, computer science, and other STEM and non-STEM professionals to engage them into startup creation with the goals of preventing, diagnosing or treating substance use disorder (SUD), one of the US most pressing health and social challenges. For more than two decades, SUD has been affecting people from all ages, demographic and socio-economic groups in the US. The National Institute on Drug Abuse (NIDA), as the lead federal agency supporting scientific research on drug use, has identified that current SUD research is not properly translating into commercial solutions for SUD. Seeking alternatives, NIDA is fostering the entrepreneurial spirit of SUD researchers, so they are the ones directly offering SUD-focused technologies into the market. This paper presents the MIT SUD Ventures program, a NIDA-funded project focused on training multidisciplinary teams of SUD researchers, engineers, healthcare, and management experts on how to commercialize their discoveries with the support of government funding sources. We introduce the learner profile, the content and skills deemed necessary to support their entrepreneurial efforts, the curriculum implemented and the program structure. We share results related to the learner expectations, experiences, and outcomes of this innovative practice, as well as future steps for the 2024 cohort and impact evaluation. Finally, guided by our results, we discuss the positive value of teaching entrepreneurial skills to academics and researchers to bridge the gaps between basic and translational research. We also make a call for action for all STEM professionals, especially engineers, computer scientists and technologists, to employ their scholarship and research capacity to have a positive impact on solving the SUD epidemic, where they are most needed.

In this paper, we introduce a novel conceptual model for a robot's behavioral adaptation in its long-term interaction with humans, integrating dynamic robot role adaptation with principles of flow experience from psychology. This conceptualization introduces a hierarchical interaction objective grounded in the flow experience, serving as the overarching adaptation goal for the robot. This objective intertwines both cognitive and affective sub-objectives and incorporates individual and group-level human factors. The dynamic role adaptation approach is a cornerstone of our model, highlighting the robot's ability to fluidly adapt its support roles-from leader to follower-with the aim of maintaining equilibrium between activity challenge and user skill, thereby fostering the user's optimal flow experiences. Moreover, this work delves into a comprehensive exploration of the limitations and potential applications of our proposed conceptualization. Our model places a particular emphasis on the multi-person HRI paradigm, a dimension of HRI that is both under-explored and challenging. In doing so, we aspire to extend the applicability and relevance of our conceptualization within the HRI field, contributing to the future development of adaptive social robots capable of sustaining long-term interactions with humans.

Today, Artificial Intelligence (AI) is prevalent in everyday life, with emerging technologies like AI companions, autonomous vehicles, and AI art tools poised to significantly transform the future. The development of AI curricula that shows people how AI works and what they can do with it is a powerful way to prepare everyone, and especially young learners, for an increasingly AI-driven world. Educators often employ robotic toolkits in the classroom to boost engagement and learning. However, these platforms are generally unsuitable for young learners and learners without programming expertise. Moreover, these platforms often serve as either programmable artifacts or pedagogical agents, rarely capitalizing on the opportunity to support students in both capacities. We designed Doodlebot, a mobile social robot for hands-on AI education to address these gaps. Doodlebot is an effective tool for exploring AI with grade school (K-12) students, promoting their understanding of AI concepts such as perception, representation, reasoning and generation. We begin by elaborating Doodlebot's design, highlighting its reliability, user-friendliness, and versatility. Then, we demonstrate Doodlebot's versatility through example curricula about AI character design, autonomous robotics, and generative AI accessible to young learners. Finally, we share the results of a preliminary user study with elementary school youth where we found that the physical Doodlebot platform was as effective and user-friendly as the virtual version. This work offers insights into designing interactive educational robots that can inform future AI curricula and tools.

Foundation models are becoming valuable tools in medicine. Yet despite their promise, the best way to leverage Large Language Models (LLMs) in complex medical tasks remains an open question. We introduce a novel multi-agent framework, named **M**edical **D**ecision-making **Agents** (**MDAgents**) that helps to address this gap by automatically assigning a collaboration structure to a team of LLMs. The assigned solo or group collaboration structure is tailored to the medical task at hand, a simple emulation inspired by the way real-world medical decision-making processes are adapted to tasks of different complexities. We evaluate our framework and baseline methods using state-of-the-art LLMs across a suite of real-world medical knowledge and clinical diagnosis benchmarks, including a comparison of LLMs’ medical complexity classification against human physicians. MDAgents achieved the **best performance in seven out of ten** benchmarks on tasks requiring an understanding of medical knowledge and multi-modal reasoning, showing a significant **improvement of up to 4.2\%** ($p$ < 0.05) compared to previous methods' best performances. Ablation studies reveal that MDAgents effectively determines medical complexity to optimize for efficiency and accuracy across diverse medical tasks. Notably, the combination of moderator review and external medical knowledge in group collaboration resulted in an average accuracy **improvement of 11.8\%**. Our code can be found at https://github.com/mitmedialab/MDAgents.

Social-emotional learning (SEL) skills are essential for children to develop to provide a foundation for future relational and academic success. Using art as a medium for creation or as a topic to provoke conversation is a well-known method of SEL learning. Similarly, social robots have been used to teach SEL competencies like empathy, but the combination of art and social robotics has been minimally explored. In this paper, we present a novel child-robot interaction designed to foster empathy and promote SEL competencies via a conversation about art scaffolded by a social robot. Participants (N=11, age range: 7-11) conversed with a social robot about emotional and neutral art. Analysis of video and speech data demonstrated that this interaction design successfully engaged children in the practice of SEL skills, like emotion recognition and self-awareness, and greater rates of empathetic reasoning were observed when children engaged with the robot about emotional art. This study demonstrated that art-based reflection with a social robot, particularly on emotional art, can foster empathy in children, and interactions with a social robot help alleviate discomfort when sharing deep or vulnerable emotions.

An essential element of K-12 AI literacy is educating learners about the ethical and societal implications of AI systems. Previous work in AI ethics literacy have developed curriculum and classroom activities that engage learners in reflecting on the ethical implications of AI systems and developing responsible AI. There is little work in using game-based learning methods in AI literacy. Games are known to be compelling media to teach children about complex STEM concepts. In this work, we developed a competitive card game for middle and high school students called “AI Audit” where they play as AI start-up founders building novel AI-powered technology. Players can challenge other players with potential harms of their technology or defend their own businesses by features that mitigate these harms. The game mechanics reward systems that are ethically developed or that take steps to mitigate potential harms. In this paper, we present the game design, teacher resources for classroom deployment and early playtesting results. We discuss our reflections about using games as teaching tools for AI literacy in K-12 classrooms.

The Feature Detection tool is a web-based activity that allows students to detect features in images and build their own rule-based classification algorithms. In this paper, we introduce the tool and share how it is incorporated into two, 45-minute lessons. The objective of the first lesson is to introduce students to the concept of feature detection, or how a computer can break down visual input into lower-level features. The second lesson aims to show students how these lower-level features can be incorporated into rule-based models to classify higher-order objects. We discuss how this tool can be used as a "first step" to the more complex concept ideas of data representation and neural networks.

The Model AI Assignments session seeks to gather and disseminate the best assignment designs of the Artificial Intelligence (AI) Education community. Recognizing that assignments form the core of student learning experience, we here present abstracts of six AI assignments from the 2023 session that are easily adoptable, playfully engaging, and flexible for a variety of instructor needs. Assignment specifications and supporting resources may be found at http://modelai.gettysburg.edu .

Teaching young people about artificial intelligence (A.I.) is recognized globally as an important education effort by organizations and programs such as UNICEF, OECD, Elements of A.I., and AI4K12. A common theme among K-12 A.I. education programs is teaching how A.I. can impact society in both positive and negative ways. We present an effective tool that teaches young people about the societal impact of A.I. that goes one step further: empowering K-12 students to use tools and frameworks to create socially responsible A.I. The computational action process is a curriculum and toolkit that gives students the lessons and tools to evaluate positive and negative impacts of A.I. and consider how they can create beneficial solutions that involve A.I. and computing technology. In a human-subject research study, 101 U.S. and international students between ages 9 and 18 participated in a one-day workshop to learn and practice the computational action process. Pre-post questionnaires measured on the Likert scale students’ perception of A.I. in society and students' desire to use A.I. in their projects. Analysis of the results shows that students who identified as female agreed more strongly with having a concern about the impacts of A.I. than those who identified as male. Students also wrote open-ended responses to questions about what socially responsible technology means to them pre- and post-study. Analysis shows that post-intervention, students were more aware of ethical considerations and what tools they can use to code A.I. responsibly. In addition, students engaged actively with tools in the computational action toolkit, specifically the novel impact matrix, to describe the positive and negative impacts of A.I. technologies like facial recognition. Students demonstrated breadth and depth of discussion of various A.I. technologies' far-reaching positive and negative impacts. These promising results indicate that the computational action process can be a helpful addition to A.I. education programs in furnishing tools for students to analyze the effects of A.I. on society and plan how they can create and use socially responsible A.I.

As artificial intelligence (AI) becomes a prominent part of modern life, AI literacy is becoming important for all citizens, not just those in technology careers. Previous research in AI education materials has largely focused on the introduction of terminology as well as AI use cases and ethics, but few allow students to learn by creating their own machine learning models. Therefore, there is a need for enriching AI educational tools with more adaptable and flexible platforms for interested educators with any level of technical experience to utilize within their teaching material. As such, we propose the development of an open-source tool (Build-A-Bot) for students and teachers to not only create their own transformer-based chatbots based on their own course material but also learn the fundamentals of AI through the model creation process. The primary concern of this paper is the creation of an interface for students to learn the principles of artificial intelligence by using a natural language pipeline to train a customized model to answer questions based on their own school curriculums. The model uses contexts given by their instructor, such as chapters of a textbook, to answer questions and is deployed on an interactive chatbot/voice agent. The pipeline teaches students data collection, data augmentation, intent recognition, and question answering by having them work through each of these processes while creating their AI agent, diverging from previous chatbot work where students and teachers use the bots as black-boxes with no abilities for customization or the bots lack AI capabilities, with the majority of dialogue scripts being rule-based. In addition, our tool is designed to make each step of this pipeline intuitive for students at a middle-school level. Further work primarily lies in providing our tool to schools and seeking student and teacher evaluations.

Given the widespread prevalence of depression and its consequential impact on individuals and society, it is crucial to obtain objective measures for early diagnosis and intervention. As a multidisciplinary topic, these objective measures should be interpretable and accessible to health care professionals, ensuring effective collaboration and treatment planning in the realm of mental health care. Even though current automated depression diagnosis approaches improved over the last decade, a critical gap exists as they often lack affect-specificity and interpretability, limiting their practical application and potential impact on mental health care. In particular, interpretability from temporal activities from videos when deep models are used is not fully explored. In this study, we present a novel framework for analyzing Deep Neural Networks’ decisions when trained on facial videos, specifically focusing on automatic depression severity diagnosis. By fine-tuning Deep Convolutional Neural Networks (DCNN) pre-trained on Action Recognition datasets on depression severity facial videos from AVEC depression dataset, our framework is able to interpret the model’s saliency maps by examining face regions and temporal expression semantics. Our approach generates both visual and quantitative explanations for the model’s decisions, providing greater insight into its reasoning. In addition to this interpretability, our video-based modeling has improved upon previous single-face benchmarks for visual depression diagnosis, resulting in enhanced predictive performance. Overall, our work demonstrates the successful development of a framework capable of generating hypotheses from a facial model’s decisions while simultaneously improving depression’s predictive capabilities.

Escape rooms have become increasingly popular as a form of entertainment, in addition to being adopted by educators for their effectiveness in improving student engagement and learning. While they have been introduced in various educational contexts, from nursing to mathematics, and for different age groups, including K-12 and university students, little research has been conducted on the benefits of escape rooms for adult learning of artificial intelligence (AI). Furthermore, most escape room implementations lack relevance to real-world situations and challenges with using AI systems in the wild. This study explores the effectiveness of an escape room, DRONEscape, as a tool for teaching AI concepts to Air Force participants. The results suggest that escape rooms can most effectively facilitate engagement and collaboration, and have positive effects on learning AI concepts. This paper also provides considerations for improvements to future iterations of AI-themed escape rooms to enhance learning, collaboration, engagement, and enjoyment.

The most meaningful connections between people are often fostered through expression of shared vulnerability and emotional experiences in personal narratives.We introduce a new task of identifying similarity in personal stories based on empathic resonance, i.e., the extent to which two people empathize with each others' experiences, as opposed to raw semantic or lexical similarity, as has predominantly been studied in NLP.Using insights from social psychology, we craft a framework that operationalizes empathic similarity in terms of three key features of stories: main events, emotional trajectories, and overall morals or takeaways.We create EM-PATHICSTORIES, a dataset of 1,500 personal stories annotated with our empathic similarity features, and 2,000 pairs of stories annotated with empathic similarity scores.Using our dataset, we finetune a model to compute empathic similarity of story pairs, and show that this outperforms semantic similarity models on automated correlation and retrieval metrics.Through a user study with 150 participants, we also assess the effect our model has on retrieving stories that users empathize with, compared to naive semantic similarity-based retrieval, and find that participants empathized significantly more with stories retrieved by our model.Our work has strong implications for the use of empathy-aware models to foster human connection and empathy between people.

This research to practice full paper explores a new educational framework for AI-informed leadership and evaluates its curriculum and pedagogical approach through a novel, tailored, research instrument. Artificial Intelligence continues to rapidly transform many aspects of markets, solutions, and organizational culture across companies, agencies, and institutions in the public and private sectors. Within complex organizations, AI tools, technologies, and applications inform how leaders engage in strategy-making, management, operations, human resources, and professional education. Non-technical managers and executives are increasingly expected to lead teams to implement responsible AI solutions with the promise to improve efficiency, effectiveness, productivity, profitability, and more. AI is rapidly transforming organizational culture, requiring non-technical leaders to develop AI literacy and essential skills to lead teams in implementing responsible AI solutions. In the face of AI-driven change, business leaders need to be AI literate and develop their own essential skills, knowledge, procedures, and perspectives to successfully set vision and strategy to lead teams that can leverage AI to achieve inward-facing and outward-facing business goals. This presents challenges and opportunities to develop new pedagogical approaches and measures to prepare and assess business leaders' AI leadership skills - including understanding human-AI systems in the workplace and their responsible development and ethical use. There are also cultural and organizational behavior challenges in successfully adopting these new capabilities into a global and diverse human-AI workforce at scale. To advance these, we present an innovative hands-on AI leadership curriculum, where participants learn by making and team problem-solving, for United States Air Force (USAF) leaders to learn about AI and its responsible use in human-robot teaming with autonomous robots. We contribute new measures to assess their attitudinal shifts in AI leadership with respect to culture, mindsets, and ethics. We present a pilot study to evaluate our curriculum design and pedagogical approach to foster positive shifts in our AI leadership measures.

This innovative practice full paper presents a novel educational program that aims to improve work readiness of emerging talent around the world through remote, paid, global apprenticeships, and human skills training for both apprentices and managers. In the last decade, multiple technical and socioeconomic factors, along with the COVID-19 pandemic, have radically changed the job market, the way companies interact with their employees and customers, and how universities train their students. In response to these changes, we have identified three core aspects of the modern workforce that need attention from academic institutions and industry to promote a more diverse, inclusive, and stable working environment for entry-level and emerging talent, across the world: 1) remote work readiness, 2) real-life mentored learning (apprenticeships), and 3) manager and supervisor preparedness. Remote work, when properly implemented, has presented advantages and opportunities to students, workers, and companies: it improves performance and facilitates innovation through cross-pollination of ideas between diverse groups and gives opportunities to emerging talent globally. Internships and apprenticeships are mechanisms implemented to promote a smoother transition from academia to the workplace. Both internships and apprenticeships promote real-life work experience but differ in that the latter includes a predesigned learning experience guided by a mentor (a seasoned manager). Senior undergraduates are frequently ill-prepared to face the difficulties of work because of a disconnect between their academic training and the needs of a job. This is especially critical in engineering students, who focus mainly on technical skills, leaving behind human and professional skills necessary to thrive in the workplace. Internships and apprenticeships offer opportunities to bridge this gap, though several evaluation criteria must be defined and met to consider them successful. To enhance remote work readiness, as well as success in internship and apprenticeship programs, manager and supervisor preparedness is critical for properly guiding engineering students, apprentices, and entry-level employees in their first job experience. With these core concepts in mind and using the Agile Continuous Education (ACE) framework, The Intern Group (TIC) and MIT Open Learning (MIT OL) created the Global Apprenticeship Program (GAP), an initiative focused on bridging the gap between talent and opportunities around the world. The program aims to 1) increase apprenticeship performance and facilitate full-time employment for students and diverse emerging talent, at a global scale; and 2) support how companies successfully recruit, onboard, and retain emerging talent. Innovation in this approach lies in the particular focus placed on the apprentice-mentor (intern-manager) dyad, including tailored training for managers and supervisors. In this paper we present in detail the different programmatic components of the learning tracks. These consist of a variety of individual self-paced asynchronous learning activities, group learning synchronous workshops, community building and cultural exchange events, and a real-life mentored learning experience (apprenticeship). We conclude with implementation challenges, opportunities for improvement, and lessons learned regarding content, pedagogies, and technologies used throughout the program, and effect on participation and engagement.

Social support plays a crucial role in managing and enhancing one's mental health and well-being. In order to explore the role of a robot's companion-like behavior on its therapeutic interventions, we conducted an eight-week-long deployment study with seventy participants to compare the impact of (1) acontrol robot with only assistant-like skills, (2) acoach-like robot with additional instructive positive psychology interventions, and (3) acompanion-like robot that delivered the same interventions in a peer-like and supportive manner. The companion-like robot was shown to be the most effective in building a positive therapeutic alliance with people, enhancing participants' well-being and readiness for change. Our work offers valuable insights into how companion AI agents could further enhance the efficacy of the mental health interventions by strengthening their therapeutic alliance with people for long-term mental health support.

Pedagogical agent research has yielded fruitful results in both academic skill learning and meta-cognitive skill acquisition, often studied in instructional or peer-to-peer paradigms. In the past decades, child-centric pedagogical research, which emphasizes the learner's active participation in learning with self-motivation, curiosity, and exploration, has attracted scholarly attention. Studies show that combining child-driven pedagogy with appropriate adult guidance leads to efficient learning and a strengthened feeling of self-efficacy. However, research on using social robots for guidance in child-driven learning still remains open and under-explored. In our study, we focus on children's exploration as the vehicle in literacy learning and develop a social robot companion that provides guidance to encourage and motivate children to explore during a storybook reading interaction. To investigate the effect of the robot's explorative guidance, we compare it against a control condition in which children have full autonomy to explore and read the storybooks. We conduct a between-subjects study with 31 children aged 4 to 6, and the result shows that children who receive explorative guidance from the social robot exhibit a growing trend of self-exploration. Further, children's self-exploration in the explorative guidance condition is found correlated to their learning outcome. We conclude the study with recommendations for designing social agents to guide children's exploration and future research directions in child-centric AI-assisted pedagogy.

As we move closer to real-world social AI systems, AI agents must be able to deal with multiparty (group) conversations. Recognizing and interpreting multiparty behaviors is challenging, as the system must recognize individual behavioral cues, deal with the complexity of multiple streams of data from multiple people, and recognize the subtle contingent social exchanges that take place amongst group members. To tackle this challenge, we propose the Multiparty-Transformer (Multipar- T), a transformer model for multiparty behavior modeling. The core component of our proposed approach is Crossperson Attention, which is specifically designed to detect contingent behavior between pairs of people. We verify the effectiveness of Multipar-T on a publicly available video-based group engagement detection benchmark, where it outperforms state-of-the-art approaches in average F-1 scores by 5.2% and individual class F-1 scores by up to 10.0%. Through qualitative analysis, we show that our Crossperson Attention module is able to discover contingent behaviors.

ADM systems can be used to perform a task as inconsequential as recommending a song on Spotify, to making a decision that is instrumental to someone's life, such as determining their candidacy for college. If an algorithm is trained on biased data, it can propagate prejudice. Thus, it is pertinent to find methods to decrease ADM bias. This paper presents a way to potentially mitigate ADM bias by teaching high school students a intersectional data analysis activity that incorporates the second pillar of the liberatory computing framework, critical consciousness. This activity is designed to enable high school students to understand the bias and history behind the college admission process, which allows students to develop a critical consciousness. Establishing a critical consciousness will diversify the computing field and the data incorporated into ADM systems by encouraging minoritized high school students to get a degree in computer science. The National Institute of Standards and Technology (NIST) suggests that diversifying the computing field has the potential to reduce bias in ADM systems. Thus, the activity is focused on students developing a critical consciousness. This paper discusses the preliminary findings from teaching a two-day computing activity to high school students.

As artificial intelligence involves and shapes personal and professional lives, there is a critical need to nurture and prepare AI-enabled problem-solvers. FutureMakers is designed as a six-week program that introduces foundational knowledge and essential skills to develop innovative solutions with AI responsibly. Our study utilized a convergent mixed-method design to evaluate the impact of the FutureMakers program on students' learning outcomes and shifting perspectives on AI. Quantitative data showed a shift in students' AI literacy with a large effect size. Qualitative data, based on student interviews, showed an awareness of an ethical engineering design process in applying technical skills to solve real-world problems. The program showed the impact of the computational action approach to tackle authentic challenges.

Biased data is increasingly becoming a part of algorithms that deter- mine people's livelihood, such as predictive policing or recidivism predictors. One of the most effective ways of understanding how such algorithms work starts by examining the systems of oppression that lead to biased data. The lesson, "Intersectional Data Analysis: Examining Shootings in Boston", begins with examining the connection between racism, housing, and policing. Then, students use their data science skills to analyze how gun violence disproportionately harms African Americans. As a result, students examine the direct effects of historical bias embedded in data. The results show the student's ability to use data science and their knowledge of gun violence being a racial justice issue to create unbiased datasets, which may lead to fair algorithms.

High-quality parent-child conversational interactions are crucial for children's social, emotional, and cognitive development. However, many children have limited exposure to these interactions at home. As increasingly accessible and scalable interventions in child development, interactive technologies, such as social robots, have great potential for facilitating parent-child interactions. However, such technology-based interventions are still underexplored, as the technologies' limited ability to understand the social-emotional dynamics of human dyadic interactions impedes their effective delivery of timely, adaptive interventions. To advance research on resolving this roadblock, we present a “dyadic affect in multimodal interaction-parent to child” (DAMI-P2C) dataset collected during a study of 34 parent-child pairs, where parents and children (3-7 years old) engaged in reading storybooks together. In contrast to existing public datasets for social-emotional behaviors in dyadic interactions, each instance for both participants in our dataset was annotated for affect by three labelers. Additionally, the dataset contains audiovisual recordings as well as each dyad's sociodemographic profiles, co-reading behaviors, affect labels, and body joints. We describe the dataset's main characteristics and provide a preliminary analysis of the interrelations between sociodemographic profiles, co-reading behaviors, and affect labels. The dataset provides us with useful insights into the computing and social science fields.

Despite the increase in awareness and support for mental health, college students' mental health is reported to decline every year in many countries. Several interactive technologies for mental health have been proposed and are aiming to make therapeutic service more accessible, but most of them only provide one-way passive contents for their users, such as psycho-education, health monitoring, and clinical assessment. We present a robotic coach that not only delivers interactive positive psychology interventions but also provides other useful skills to build rapport with college students. Results from our on-campus housing deployment feasibility study showed that the robotic intervention showed significant association with increases in students' psychological well-being, mood, and motivation to change. We further found that students' personality traits were associated with the intervention outcomes as well as their working alliance with the robot and their satisfaction with the interventions. Also, students' working alliance with the robot was shown to be associated with their pre-to-post change in motivation for better well-being. Analyses on students' behavioral cues showed that several verbal and nonverbal behaviors were associated with the change in self-reported intervention outcomes. The qualitative analyses on the post-study interview suggest that the robotic coach's companionship made a positive impression on students, but also revealed areas for improvement in the design of the robotic coach. Results from our feasibility study give insight into how learning users' traits and recognizing behavioral cues can help an AI agent provide personalized intervention experiences for better mental health outcomes.

Generative Artificial Intelligence (AI) models are a compelling way to introduce K-12 students to AI education using an artistic medium, and hence have drawn attention from K-12 AI educators. Previous Creative AI curricula mainly focus on Generative Adversarial Networks (GANs) while paying less attention to Autoregressive Models, Variational Autoencoders (VAEs), or other generative models, which have since become common in the field of generative AI. VAEs' latent-space structure and interpolation ability could effectively ground the interdisciplinary learning of AI, creative arts, and philosophy. Thus, we designed a lesson to teach high school students about VAEs. We developed a web-based game and used Plato's cave, a philosophical metaphor, to introduce how VAEs work. We used a Google Colab notebook for students to re-train VAEs with their hand-written digits to consolidate their understandings. Finally, we guided the exploration of creative VAE tools such as SketchRNN and MusicVAE to draw the connection between what they learned and real-world applications. This paper describes the lesson design and shares insights from the pilot studies with 22 students. We found that our approach was effective in teaching students about a novel AI concept.

Special education teachers’ perception and attitudes towards technology and design play a critical role in pedagogical practices. The study aims to explore changes in special education teachers’ attitudes and design beliefs through a co-design process with children. The initial pilot study focused on preparing special education teachers for effective integration of pedagogical agents into teaching and learning. The initial pilot study followed the mixed-method design and was guided by the following research question: In what ways the co-design process influenced teachers’ attitudes and design beliefs towards pedagogical agents through the co-design process with children? The preliminary results indicated by the end of the program that teachers’ attitudes towards pedagogical agents increased significantly with moderate effect sizes, which might contribute to the co-design process and interactions with pedagogical agents. Qualitative analysis based on teacher interviews, lesson projects, and field notes also suggested that the shifts in participant's attitudes and design beliefs are influenced by a variety of personal and contextual factors including i) The didactic use of pedagogical agents; ii) the usefulness of pedagogical agents for inclusive education; (iii) teachers’ attitudes about the role of the teacher as a designer, and (iv) leadership support to facilitate the immersive learning experience created through the interaction between the human and pedagogical agents.

In this work, we explore the effect of a social robot’s embodiment and creativity scaffolding on children’s creative problem solving skills in the context of a digital creative problem-solving game called Escape!Bot. Children aged 5-11 years played the video game, which involved assembling contraptions to escape a digital world, and the robot Jibo acted as a collaborative peer that offered questions, reflective prompts, challenges, and ideas. In order to evaluate the role of the robot’s co-presence and creativity scaffolding, we ran a 2x2 experiment to determine the factorial efficacy of the robot’s embodiment and creativity scaffolding behaviors. We observed mixed results, with the robot’s creativity scaffolding having a positive influence on the time taken to complete the game, but not on the overall use of novel objects or reuse of objects. We present the system design, user study and findings from Escape!Bot to investigate the feasibility of designing social robots to support creative problem solving.

This Research to Practice Full Paper presents an AI Education program. In January 2021 MIT entered into an agreement with the United States Air Force (USAF) and the Department of Defense (DoD) to design and offer a new educational research program focusing on Artificial Intelligence (AI) training. The goal of this collaboration is to design and advance educational research activities that promote maximum learning outcomes at scale for learners with diverse roles and educational backgrounds, ranging from Air Force and DoD personnel to the general public. This program is expected to offer different learning tracks addressing different groups of USAF employees based on their unique professional needs and backgrounds. The first pilot is currently underway and will provide the research team with data and insights that will inform the next iteration of the program, with the ultimate goal of formulating recommendations for the USAF and general public on how to reach large numbers of learners at scale in an optimum way. Currently, the program offers three different learning journeys for each of three different cohorts of USAF employees (i.e., leaders, developers, and users). These learning journeys span from online asynchronous and synchronous courses to in-person activities. Our research goals focus on exploring and understanding the learner experience via the study and analysis of AI content and curriculum, pedagogical approaches, learning modalities, and technological innovations to deliver learning experiences at scale. Key research activities involve evaluating a range of existing digital AI courses, mapping out the landscape of educational needs and competencies, and developing and piloting experiential learning experiences (to advance innovative technology-enabled training and learning technologies and methods). This paper discusses how preliminary research findings from this first pilot are informing the design and implementation of the next program iteration. The research provides insights that will benefit AI learners across the US while supporting the DoD’s objective to develop elite and world-class AI-ready services.

Children are beginning to interact and develop rapport with social robots in their homes. These devices pose new concerns around marketing to children. These include questions of how advertisements can and should be embedded in a robot and the robot's persona and which methods of conveying advertisements to the user are deceptive. In this paper, we engage with 62 children ages 9–12 in an activity to design future robot advertising policies. Results demonstrate that children prefer robots to advertise to them through casual conversations, citing a more positive user experience and the benefit of personalized and conversationally relevant advertising. These findings illuminate a tension between child preferences and more deceptive advertising policies. Overall, the work presented in this paper prompts new design and legal policy questions for how and if robots should advertise to children.

Technology, such as robots, can entrench existing inequities and create new ones. Leveraging design justice, par-ticipatory design, and design fictions, we propose new ways for designing robots and policies around social robots to incorporate more voices and values into design processes. This work critically examines design justice in the context of human-robot interaction (HRI) and suggests a framework to engage multiple stakeholders in participatory design of robots and policies grounded in design justice. Overall, we promote discussion around how we can design more equitable robot technology design and policy design in HRI.

In the past two decades, human-robot interaction (HRI) researchers have increasingly deployed autonomous and reliable robots long-term in various social contexts including the home. Our work provides a mixed-method approach for analyzing older adults' long-term robot usage data patterns combining quantitative data of robot usage logs with qualitative descriptions from participants' own experience. Overall, this provides a fuller picture to how older adults use and experience social robots in their homes. Our work involves a robot hosting period for at least a month (up to 12 months) in older adults' homes with an experience debrief session held a month into the robot hosting time period. We propose reflections on the novelty effect with respect to older adults' usage data and highlight feelings of guilt, the robot's proactivity and movement, meeting (or not meeting) user expectations, and the robot's persona as key aspects of the hosting experience that promoted usage or non-usage. Finally, we provide design guidelines for structuring future mixed-method long-term robot usage studies being mindful of ethical considerations in this space.

Social agents have been mostly designed to engage with children one-on-one as tutors or learning peers. Besides this child-robot dyadic interaction paradigm, they have the potential to empower parents to more actively interact with their children. Robot-assisted parent-child interaction could be a sustainable future approach for promoting children’s in-home learning. Motivated by this new design direction, this work takes an iterative design approach to explore how we design triadic interactions through "lived technology experiences" and interviews. For 3-6 weeks, we deployed and remotely teleoperated a social robot in the homes of 12 families with 3-7-year-old children to engage in a triadic story reading activity with both parent and child for six 25-min sessions. Before and after the deployment, we conducted a semi-structured interview with participants on their triadic interaction experience and desired robot design features. The results of our qualitative analysis show that social robots can improve various aspects of parent-child interaction. We propose design guidelines for robot-assisted parent-child interactions at home, the considerations of participants’ values around technology design, and promotion of their long-term lived technology experiences as critical sources for design knowledge.

As social robots rapidly become mainstream technologies, it is critical for HRI researchers and practitioners to consider their societal and ethical impacts as well as their ability to perpetuate or mitigate intersectional social inequities and hierarchies relating to race, class, gender, disability, and other social axes. Through an equity, ethics, and justice-centered audit of human-robot interaction (HRI) scholarship, we reveal how the HRI community has engaged with these topics over the past two decades. We use the five senses ethical framework that has been proposed specifically for use in HRI contexts to perform the review paired with an analysis of equity and justice. We then expand the Design Justice framework (a framework for analyzing how design impacts society and distributes benefits and burdens to society through the lenses of equity, values, scope, ownership, and accountability) to HRI contexts through the inclusion of HRI-specific topics such as autonomy, transparency, deception, and policies. We invite researchers and practitioners to explore the HRI Equitable Design framework to work towards designing equitable and inclusive HRI research studies and technologies.

Although artificial intelligence (AI) is increasingly involved in everyday technologies, AI literacy amongst the general public remains low. Thus many AI education curricula for people without prior AI experience have emerged, often utilizing graphical programming languages for hands-on projects. However, there are no tools that assist educators in evaluating learners’ AI projects or provide learners with contemporaneous feedback on their work. We developed LevelUp, an automatic code analysis tool to support these educators and learners. LevelUp is built into a block-based programming platform and gives users continuous feedback on their text classification projects. We evaluated the tool with a crossover user study where participants completed two text classification projects, once where they could access LevelUp and once when they could not. To measure the tool’s impact on participants’ understanding of text classification, we used pre-post assessments and graded both of their projects against LevelUp’s rubric. We saw a significant improvement in the quality of participants’ projects after they used the tool. We also used questionnaires to solicit participants’ feedback. Overall, participants said that LevelUp was useful and intuitive. Our investigation of this novel automatic assessment tool can inform the design of future code analysis tools for AI education.

Applications of Generative Machine Learning techniques such as Generative Adversarial Networks (GANs) are used to generate new instances of images, music, text, and videos. While GANs have now become commonplace on social media, a part of children’s lives, and have considerable ethical implications, existing K-12 AI education curricula do not include generative AI. We present a new module, “What are GANs?”, that teaches middle school students how GANs work and how they can create media using GANs. We developed an online, team-based game to simulate how GANs work. Students also interacted with up to four web tools that apply GANs to generate media. This module was piloted with 72 middle school students in a series of online workshops. We provide insight into student usage, understanding, and attitudes towards this lesson. Finally, we give suggestions for integrating this lesson into AI education curricula.

Body-tracking artificial intelligence (AI) systems like Kinect games, Snapchat Augmented Reality (AR) Lenses, and Instagram AR Filters are some of the most engaging ways students experience AI in their everyday lives. Additionally, many students have existing interests in physical hobbies like sports and dance. In this paper, we present PoseBlocks; a suite of block-based programming tools which enable students to build compelling body-interactive AI projects in any web browser, integrating camera/microphone inputs and body-sensing user interactions. To accomplish this, we provide a custom block-based programming environment building on the open source Scratch project, introducing new AI-model-powered blocks supporting body, hand, and face tracking, emotion recognition, and the ability to integrate custom image/pose/audio models from the online transfer learning tool Teachable Machine. We introduce editor functionality such as a project video recorder, pre-computed video loops, and integration with curriculum materials. We discuss deploying this toolkit with an accompanying curriculum in a series of synchronous online pilots with 46 students, aged 9-14. In analyzing class projects and discussions, we find that students learned to design, train, and integrate machine learning models in projects of their own devising while exploring ethical considerations such as stakeholder values and algorithmic bias in their interactive AI systems.

To enable a diverse citizenry to fully participate in future society, we must prepare all students to construct and critique emerging technologies like Artificial Intelligence (AI). Classrooms are important spaces to teach students these skills, however there are few AI curricula that have been developed for and used by K-12 teachers. We developed the \textit{How to Train Your Robot: AI and Ethics Curriculum} for middle school teachers who want to introduce AI to their students. This paper describes the curriculum and professional development we used to prepare teachers to run a five-day AI course. Before and after they ran the curriculum, we interviewed teachers to understand their opinions on pedagogical approaches to teaching AI, meeting students' needs, and the feasibility of doing the activities in the classroom. Our results indicate that, with appropriate training, even teachers who were new to computer science felt prepared and successfully engaged their students in the topic. We hope our insights will inform future efforts to realize AI education in primary and secondary classrooms.

Applications of generative models such as Generative Adversarial Networks (GANs) have made their way to social media platforms that children frequently interact with. While GANs are associated with ethical implications pertaining to children, such as the generation of Deepfakes, there are negligible efforts to educate middle school children about generative AI. In this work, we present a generative models learning trajectory (LT), educational materials, and interactive activities for young learners with a focus on GANs, creation and application of machine-generated media, and its ethical implications. The activities were deployed in four online workshops with 72 students (grades 5-9). We found that these materials enabled children to gain an understanding of what generative models are, their technical components and potential applications, and benefits and harms, while reflecting on their ethical implications. Learning from our findings, we propose an improved learning trajectory for complex socio-technical systems.

Recent state-of-the-art approaches in opendomain dialogue include training end-to-end deep-learning models to learn various conversational features like emotional content of response, symbolic transitions of dialogue contexts in a knowledge graph and persona of the agent and the user, among others.While neural models have shown reasonable results, modelling the cognitive processes that humans use when conversing with each other may improve the agent's quality of responses.A key element of natural conversation is to tailor one's response such that it accounts for concepts that the speaker and listener may or may not know and the contextual relevance of all prior concepts used in conversation.We show that a rich representation and explicit modeling of these psychological processes can improve predictions made by existing neural network models.In this work, we propose a novel probabilistic approach using Markov Random Fields (MRF) to augment existing deep-learning methods for improved next utterance prediction.Using human and automatic evaluations, we show that our augmentation approach significantly improves the performance of existing state-ofthe-art retrieval models for open-domain conversational agents.

Parent-child nonverbal communication plays a crucial role in understanding their relationships and assessing their interaction styles. However, prior works have seldom studied the exchange of these nonverbal cues between the dyad and focused on isolated cues from one person at a time. In contrast, this work analyzes both parents' and children's individual and dyadic nonverbal behaviors in relation to their four relationship characteristics, i.e., child temperament, parenting style, parenting stress, and home literacy environment. We utilize a state-of-the-art feature selection framework on a dataset of 31 parent-child interactions to automatically extract and select a set of temporal nonverbal behaviors as key indicators of the dyad's relationship characteristics. The results show that relationship characteristics were associated with both individuals' and dyads' nonverbal behaviors. This finding highlights the importance of accounting for both individual- and dyad-scale nonverbal behaviors when predicting dyadic relationship characteristics as well as the potential limitations of utilizing single persons' nonverbal data in isolation. It therefore motivates future work on this topic to take a holistic and relational approach. The dataset and extracted nonverbal data are made public to aid the development of automated detection tools for parent-child relationship characteristics that trains on visual recordings of their dyadic interactions.

Self-disclosure is an important part of mental health treatment process. As interactive technologies are becoming more widely available, many AI agents for mental health prompt their users to self-disclose as part of the intervention activities. However, most existing works focus on linguistic features to classify self-disclosure behavior, and do not utilize other multi-modal behavioral cues. We present analyses of people's non-verbal cues (vocal acoustic features, head orientation and body gestures/movements) exhibited during self-disclosure tasks based on the human-robot interaction data collected in our previous work. Results from the classification experiments suggest that prosody, head pose, and body postures can be independently used to detect self-disclosure behavior with high accuracy (up to 81%). Moreover, positive emotions, high engagement, self-soothing and positive attitudes behavioral cues were found to be positively correlated to self-disclosure. Insights from our work can help build a self-disclosure detection model that can be used in real time during multi-modal interactions between humans and AI agents.

Intent recognition models, which match a written or spoken input's class in order to guide an interaction, are an essential part of modern voice user interfaces, chatbots, and social robots. However, getting enough data to train these models can be very expensive and challenging, especially when designing novel applications such as real-world human-robot interactions. In this work, we first investigate how much training data is needed for high performance in an intent classification task. We train and evaluate BiLSTM and BERT models on various subsets of the ATIS and Snips datasets. We find that only 25 training examples per intent are required for our BERT model to achieve 94% intent accuracy compared to 98% with the entire datasets, challenging the belief that large amounts of labeled data are required for high performance in intent recognition. We apply this knowledge to train models for a real-world HRI application, character strength recognition during a positive psychology interaction with a social robot, and evaluate against the Character Strength dataset collected in our previous HRI study. Our real-world HRI application results also confirm that our model can produce 76% intent accuracy with 25 examples per intent compared to 80% with 100 examples. In a real-world scenario, the difference is only one additional error per 25 classifications. Finally, we investigate the limitations of our minimal data models and offer suggestions on developing high quality datasets. We conclude with practical guidelines for training BERT intent recognition models with minimal training data and make our code and evaluation framework available for others to replicate our results and easily develop models for their own applications.

More and more voice-user interfaces (VUIs), such as smart speakers like Amazon Alexa or social robots like Jibo or Cozmo, are entering multi-user environments including homes. VUIs can utilize multi-modal cues such as graphics, expressive sounds, and movement to convey social engagement, affecting how users perceive agents as social others. Reciprocal relationships with VUIs, i.e., relationships with give-and-take between the VUI and user, are of key interest as they are more likely to foster rapport and emotional engagement, and lead to successful collaboration. Through an elicitation study with three commercially available VUIs, we explore small group interactions (n = 33 participants) focused on the behaviors participants display to various VUIs to understand (1) reciprocal interactions between VUIs and participants and among small groups and (2) how participants engage with VUIs as the interface's embodiment becomes more socially capable. The discussion explores (1) theories of sociability applied to the users' behaviors seen with the VUIs, and (2) the group contexts where VUIs that build reciprocal relationships with users can become a powerful persuasive technology and a collaborative companion. We conclude the discussion with recommendations for promoting reciprocity from participants and, therefore, fostering rapport and emotional engagement in VUI interactions.

Users can provide valuable insights for designing new technologies like social robots, with the right tools and methodologies. Challenges in inviting users as co-designers of social robots is due to lack of guidelines or methodologies to (1) organize co-design processes and/or (2) engage with people long-term to develop technologies together. The main contribution of this work is to provide guidelines for long- term co-design for how other researchers can adopt long- term co-design, informed by a 12-month co-design with older adults designing a social social robot. We leveraged human- centered, tactile and experiential design activities, including participatory design, based upon the following design principles: scenario specific exploration, long-term lived experiences, supporting multiple design activities, cultivating relationships, and employing divergent and convergent processes. We present seven different sessions across three stages as examples of this methodology that build on each other to engage users as co- designers, successfully deployed in a co-design project of home social robots with 28 older adults. Lastly, we detail 10 long- term divergent-convergent co-design guidelines for designing social robots. We demonstrate the value of leveraging people's lived technology experiences and co-design activities to generate actionable social robot design guidelines, advocating for more applications of the methodology in broader contexts as well.

In this experience report, we describe an AI summer workshop designed to prepare middle school students to become informed citizens and critical consumers of AI technology and to develop their foundational knowledge and skills to support future endeavors as AI-empowered workers. The workshop featured the 30-hour "Developing AI Literacy" or DAILy curriculum that is grounded in literature on child development, ethics education, and career development. The participants in the workshop were students between the ages of 10 and 14; 87% were from underrepresented groups in STEM and Computing. In this paper we describe the online curriculum, its implementation during synchronous online workshop sessions in summer of 2020, and preliminary findings on student outcomes. We reflect on the successes and lessons we learned in terms of supporting students' engagement and conceptual learning of AI, shifting attitudes toward AI, and fostering conceptions of future selves as AI-enabled workers. We conclude with discussions of the affordances and barriers to bringing AI education to students from underrepresented groups in STEM and Computing.

In recent years, Artificial Intelligence (AI) has become increasingly prevalent in our lives. Because of this, individuals of all ages need to be aware of how AI works. To introduce middle school students to AI concepts, we built a text classifier extension into a block-based programming interface that allows students to train custom machine learning models. To make the extension more accessible, a translator was incorporated where the language of each input is automatically detected and translated to English. Our classifier's accuracy was comparable to similar classifiers such as Machine Learning for Kids' text classifier and Uclassify's text classifier, and its effectiveness was tested against these classifiers with two test datasets. We piloted the classifier with middle school students in an online AI course. The students first learned the concepts behind the classifier which consisted of word embeddings, K-Nearest-Neighbors, and classification bias. They were then able to use the text classifier to create their own projects. Some of the projects were a snake identifier, a TV show suggester, a chat robot, and a healthcare robot. With this extension, students were able to engage in project-based learning to become more knowledgeable about the ever-growing field of AI and raise their awareness about applications of AI within their own lives.

Artificial curiosity, based on developmental psychology concepts wherein an agent attempts to maximize its learning progress, has gained much attention in recent years. Similarly, social robots are slowly integrating into our daily lives, in schools, factories, and in our homes. In this contribution, we integrate recent advances in artificial curiosity and social robots into a single expressive cognitive architecture. It is composed of artificial curiosity and social expressivity modules and their unique link, i.e., the robot verbally and non-verbally communicates its internally estimated learning progress, or learnability, to its human companion. We implemented this architecture in an interaction where a fully autonomous robot took turns with a child trying to select and solve tangram puzzles on a tablet. During the curious robot’s turn, it selected its estimated most learnable tangram to play, communicated its selection to the child, and then attempted at solving it. We validated the implemented architecture and showed that the robot learned, estimated its learnability, and improved when its selection was based on its learnability estimation. Moreover, we ran a comparison study between curious and non-curious robots, and showed that the robot’s curiosity-based behavior influenced the child’s selections. Based on the artificial curiosity module of the robot, we have formulated an equation that estimates each child’s moment-by-moment curiosity based on their selections. This analysis revealed an overall significant decrease in estimated curiosity during the interaction. However, this drop in estimated curiosity was significantly larger with the non-curious robot, compared to the curious one. These results suggest that the new architecture is a promising new approach to integrate state-of-the-art curiosity-based algorithms to the growing field of social robots.

Understanding how machines learn is critical for children to develop useful mental models for exploring artificial intelligence (AI) and smart devices that they now frequently interact with. Although children are very familiar with having conversations with conversational agents like Siri and Alexa, children often have limited knowledge about AI and machine learning. We leverage their existing familiarity and present Zhorai, a conversational platform and curriculum designed to help young children understand how machines learn. Children ages eight to eleven train an agent through conversation and understand how the knowledge is represented using visualizations. This paper describes how we designed the curriculum and evaluated its effectiveness with 14 children in small groups. We found that the conversational aspect of the platform increased engagement during learning and the novel visualizations helped make machine knowledge understandable. As a result, we make recommendations for future iterations of Zhorai and approaches for teaching AI to children.

If we expect our children to be driving technology design agendas in the future, we must first help them recognize that opinions and beliefs are baked into the technologies that we create and that these opinions may serve some groups of people more than others. In this paper, we discuss an ethical design activity completed by 19 middle school-aged children. The activity encourages students to see technical systems as socio-technical systems, to engage them in stakeholder analysis, and to apply ethical design tools in order to redesign YouTube. Results indicate students are capable of transforming into critical users and ethical designers of technology. They are able to recognize the underlying design agendas for popular technologies such as YouTube, identify stakeholders who shape those design agendas, and apply an array of tools to reimagine technologies in a more inclusive manner.

Prior work in affect-aware educational robots has often relied on a common belief that the relationship between student affect and learning is independent of agent behaviors (child's/robot's) or unidirectional (positive/negative but not both) throughout the entire student-robot interaction. We argue that the student affect-learning relationship should be interpreted in two contexts: (1) social learning paradigm and (2) sub-events within child-robot interaction. In our paper, we examine two different social learning paradigms where children interact with a robot that acts either as a tutor or a tutee. Sub-events within child-robot interaction are defined as task-related events occurring in specific phases of an interaction (e.g., when the child/robot gets a wrong answer). We examine sub-events at a macro level (entire interaction) and a micro level (within specific sub-events). In this paper, we provide an in-depth correlation analysis of children's facial affect and vocabulary learning. We found that children's affective displays became more predictive of their vocabulary learning when children interacted with a tutee robot who did not scaffold their learning. Additionally, children's affect displayed during micro-level events was more predictive of their learning than during macro-level events. Last, we found that the affect-learning relationship is not unidirectional, but rather is modulated by context, i.e., several affective states facilitated student learning when displayed in some sub-events but inhibited learning when displayed in others. These findings indicate that both social learning paradigm and sub-events within interaction modulate student affect-learning relationship.

Automatic speech-based affect recognition of individuals in dyadic conversation is a challenging task, in part because of its heavy reliance on manual pre-processing. Traditional approaches frequently require hand-crafted speech features and segmentation of speaker turns. In this work, we design end-to-end deep learning methods to recognize each person's affective expression in an audio stream with two speakers, automatically discovering features and time regions relevant to the target speaker's affect. We integrate a local attention mechanism into the end-to-end architecture and compare the performance of three attention implementations - one mean pooling and two weighted pooling methods. Our results show that the proposed weighted-pooling attention solutions are able to learn to focus on the regions containing target speaker's affective information and successfully extract the individual's valence and arousal intensity. Here we introduce and use a "dyadic affect in multimodal interaction - parent to child" (DAMI-P2C) dataset collected in a study of 34 families, where a parent and a child (3-7 years old) engage in reading storybooks together. In contrast to existing public datasets for affect recognition, each instance for both speakers in the DAMI-P2C dataset is annotated for the perceived affect by three labelers. To encourage more research on the challenging task of multi-speaker affect sensing, we make the annotated DAMI-P2C dataset publicly available, including acoustic features of the dyads' raw audios, affect annotations, and a diverse set of developmental, social, and demographic profiles of each dyad.

A significant number of college students suffer from mental health issues that impact their physical, social, and occupational outcomes. Various scalable technologies have been proposed in order to mitigate the negative impact of mental health disorders. However, the evaluation for these technologies, if done at all, often reports mixed results on improving users' mental health. We need to better understand the factors that align a user's attributes and needs with technology-based interventions for positive outcomes. In psychotherapy theory, therapeutic alliance and rapport between a therapist and a client is regarded as the basis for therapeutic success. In prior works, social robots have shown the potential to build rapport and a working alliance with users in various settings. In this work, we explore the use of a social robot coach to deliver positive psychology interventions to college students living in on-campus dormitories. We recruited 35 college students to participate in our study and deployed a social robot coach in their room. The robot delivered daily positive psychology sessions among other useful skills like delivering the weather forecast, scheduling reminders, etc. We found a statistically significant improvement in participants' psychological wellbeing, mood, and readiness to change behavior for improved wellbeing after they completed the study. Furthermore, students' personality traits were found to have a significant association with intervention efficacy. Analysis of the post-study interview revealed students' appreciation of the robot's companionship and their concerns for privacy.

Conversational AI agents are proliferating, embodying a range of devices such as smart speakers, smart displays, robots, cars, and more. We can envision a future where a personal conversational agent could migrate across different form factors and environments to always accompany and assist its user to support a far more continuous, personalized and collaborative experience. This opens the question of what properties of a conversational AI agent migrates across forms, and how it would impact user perception. To explore this, we developed a Migratable AI system where a user's information and/or the agent's identity can be preserved as it migrates across form factors to help its user with a task. We validated the system by designing a 2x2 between-subjects study to explore the effects of information migration and identity migration on user perceptions of trust, competence, likeability and social presence. Our results suggest that identity migration had a positive effect on trust, competence and social presence, while information migration had a positive effect on trust, competence and likeability. Overall, users report highest trust, competence, likeability and social presence towards the conversational agent when both identity and information were migrated across embodiments.

Personalized education technologies capable of delivering adaptive interventions could play an important role in addressing the needs of diverse young learners at a critical time of school readiness. We present an innovative personalized social robot learning companion system that utilizes children’s verbal and nonverbal affective cues to modulate their engagement and maximize their long-term learning gains. We propose an affective reinforcement learning approach to train a personalized policy for each student during an educational activity where a child and a robot tell stories to each other. Using the personalized policy, the robot selects stories that are optimized for each child’s engagement and linguistic skill progression. We recruited 67 bilingual and English language learners between the ages of 4–6 years old to participate in a between-subjects study to evaluate our system. Over a three-month deployment in schools, a unique storytelling policy was trained to deliver a personalized story curriculum for each child in the Personalized group. We compared their engagement and learning outcomes to a Non-personalized group with a fixed curriculum robot, and a baseline group that had no robot intervention. In the Personalization condition, our results show that the affective policy successfully personalized to each child to boost their engagement and outcomes with respect to learning and retaining more target words as well as using more target syntax structures as compared to children in the other groups.

PopBots is a hands-on toolkit and curriculum designed to help young children learn about artificial intelligence (AI) by building, programming, training, and interacting with a social robot. Today’s children encounter AI in the forms of smart toys and computationally curated educational and entertainment content. However, children have not yet been empowered to understand or create with this technology. Existing computational thinking platforms have made ideas like sequencing and conditionals accessible to young learners. Going beyond this, we seek to make AI concepts accessible. We designed PopBots to address the specific learning needs of children ages four to seven by adapting constructionist ideas into an AI curriculum. This paper describes how we designed the curriculum and evaluated its effectiveness with 80 Pre-K and Kindergarten children. We found that the use of a social robot as a learning companion and programmable artifact was effective in helping young children grasp AI concepts. We also identified teaching approaches that had the greatest impact on student’s learning. Based on these, we make recommendations for future modules and iterations for the PopBots platform.

In recent years, researchers have developed technology to analyze human facial expressions and other affective data at very high time resolution. This technology is enabling researchers to develop and study interactive robots that are increasingly sensitive to their human interaction partners' affective states. However, typical interaction planning models and algorithms operate on timescales that are frequently orders of magnitude larger than the timescales at which real-time affect data is sensed. To bridge this gap between the scales of sensor data collection and interaction modeling, affective data must be aggregated and interpreted over longer timescales. In this paper we clarify and formalize the computational task of affect interpretation in the context of an interactive educational game played by a human and a robot, during which facial expression data is sensed, interpreted, and used to predict the interaction partner's gameplay behavior. We compare different techniques for affect interpretation, used to generate sets of affective labels for an interactive modeling and inference task, and evaluate how the labels generated by each interpretation technique impact model training and inference. We show that incorporating a simple method of personalization into the affect interpretation process - dynamically calculating and applying a personalized threshold for determining affect feature labels over time - leads to a significant improvement in the quality of inference, comparable to performance gains from other data pre-processing steps such as smoothing data via median filter. We discuss the implications of these findings for future development of affect-aware interactive robots and propose guidelines for the use of affect interpretation methods in interactive scenarios.

Children's interactions with social robots and other technologies are increasingly longitudinal, especially in areas such as healthcare, therapy, and education. As such, we need to understand how children perceive social robots over time and the kinds of relationships they develop. Relatively few validated assessments exist that measure young children's relationships or their perception and acceptance of social robots. Thus, we present pilot tests of two assessments created for use with children aged 4--7: the Picture Sorting Task and the Social Acceptance Questionnaire. Through a single-session study and also a long-term study, we found that children responded appropriately to the assessments and that the assessments could capture changes in children's perception and relationship over multiple encounters.

Children's creativity contributes to their learning outcomes and personal growth. Standardized measures of creative thinking reveal that as children enter elementary school, their creativity drops. In this work, we evaluated whether a social robotic peer can help 6-10-year-old children think creatively by demonstrating creative behavior. We designed verbal and non-verbal behaviors of the social robot that constitute interaction patterns for artificial creativity. 51 participants played the Droodle Creativity Game with the robot to generate creative titles for ambiguous images. One group of participants interacted with the creative robot, and one group interacted with the non-creative robot. Participants that interacted with the creative robot generated significantly higher number of Droodle titles, expressed greater variety in titles, and scored higher on the Droodles' creativity. We observe that children can model a social robotic peer's creativity, and hence inform robot interaction patterns for artificial creativity that can foster creativity in children.

We developed a novel early childhood artificial intelligence (AI) platform, PopBots, where preschool children train and interact with social robots to learn three AI concepts: knowledge-based systems, supervised machine learning, and generative AI. We evaluated how much children learned by using AI assessments we developed for each activity. The median score on the cumulative assessment was 70% and children understood knowledge-based systems the best. Then, we analyzed the impact of the activities on children's perceptions of robots. Younger children came to see robots as toys that were smarter than them, but their older counterparts saw them more as people that were not as smart as them. Children who performed worse on the AI assessments believed that robots were like toys that were not as smart as them, however children who did better on the assessments saw robots as people who were smarter than them. We believe early AI education can empower children to understand the AI devices that are increasingly in their lives.

This paper defines a dual computational framework to nonverbal communication for human-robot interactions. We use a Bayesian Theory of Mind approach to model dyadic storytelling interactions where the storyteller and the listener have distinct roles. The role of storytellers is to influence and infer the attentive state of listeners using speaker cues, and we computationally model this as a POMDP planning problem. The role of listeners is to convey attentiveness by influencing perceptions through listener responses, which we computational model as a DBN with a myopic policy. Through a comparison of state estimators trained on human-human interaction data, we validate our storyteller model by demonstrating how it outperforms current approaches to attention recognition. Then through a human-subjects experiment where children told stories to robots, we demonstrate that a social robot using our listener model more effectively communicates attention compared to alternative approaches based on signaling.

In this paper we present additional results from a prior study of speech-based games to promote early literacy skills through child-robot interaction [6]. The additional data and results support our original conclusion, that pronunciation analysis software can be an effective enabler of speech child-robot interactions. We also include a comparison of other pronunciation services, an updated version of the SpeechAce API and a new technology from Soapbox Labs. We reflect on some lessons learned and introduce a redesigned version of the game interaction called `RhymeRacer' based on the results and observations from both data collections.

Intelligent tutoring systems (ITS) provide educational benefits through one-on-one tutoring by assessing children's existing knowledge and providing tailored educational content. In the domain of language acquisition, several studies have shown that children often learn new words by forming semantic relationships with words they already know. In this paper, we present a model that uses word semantics (semantics-based model) to make inferences about a child's vocabulary from partial information about their existing vocabulary knowledge. We show that the proposed semantics-based model outperforms models that do not use word semantics (semantics-free models) on average. A subject-level analysis of results reveals that different models perform well for different children, thus motivating the need to combine predictions. To this end, we use two methods to combine predictions from semantics-based and semantics-free models and show that these methods yield better predictions of a child's vocabulary knowledge. Our results motivate the use of semantics-based models to assess children's vocabulary knowledge and build ITS that maximizes children's semantic understanding of words.

In May 2018, the Association for the Advancement of Artificial Intelligence (AAAI) and the Computer Science Teachers Association (CSTA) formed a joint working group to develop national guidelines for teaching K-12 students about artificial intelligence. Inspired by CSTA's national standards for K-12 computing education, the - AI for K-12 guidelines (ai4k12.org) will define what students in each grade band should know about artificial intelligence, machine learning, and robotics. The working group is also creating an online resource directory where teachers can find AI-related videos, demo software, and activity descriptions they can incorporate into their lesson plans. The goal of this session is to raise the SIGCSE community's awareness of the initiative, its deliverables, and outcomes, and to foster a community-wide conversation about AI education in K-12. This initiative parallels other recent initiatives in K-12 AI education and community-wide initiatives and discussions around CS For All and CS in K-12. This Special Session is aimed toward K-12 CS educators, researchers, and curriculum and tool designers.

Intelligent toys and smart devices are becoming ubiquitous in children's homes. As such, it is imperative to understand how these computational objects impact children's development. Children's attribution of intelligence relates to how they perceive the behavior of these agents [6]. However, their underlying reasoning is not well understood. To explore this, we invited 30 pairs of children (4--10 years old) and their parents to assess the intelligence of mice, robots, and themselves in a maze-solving activity. Participants watched videos of mice and robots solving a maze. Then, they solved the maze by remotely navigating a robot. Solving the maze enabled participants to gain insight into the agent's mind by referencing their own experience. Children and their parents gave similar answers for whether the mouse or the robot was more intelligent and used a wide variety of explanations. We also observed developmental differences in childrens' references to agents' social-emotional attributes, strategies and performance.

Social robots are increasingly being developed for long-term interactions with children in domains such as healthcare, education, therapy, and entertainment. As such, we need to deeply understand how children's relationships with robots develop through time. However, there are few validated assessments for measuring young children's long-term relationships. In this paper, we present a pilot test of four assessments that we have adapted or created for use in this context with children aged 5--6: the Inclusion of Other in Self task, the Social-Relational Interview, the Narrative Description, and the Self-disclosure Task. We show that children can appropriately respond to these assessments with reasonably high internal reliability, and that the proposed assessments are able to capture child-robot relationship adjustments over a long-term interaction. Furthermore, we discuss gender and population differences in children's responses.

Today's children are growing up with smart toys, Internet-connected devices that use artificial intelligence to drive interactive play. In a prior research study, we found that children ages 4--10 perceive these toys as worthy of trust [5]. This leads us to inquire if children in this age range could be directly influenced by these devices. In this work, we used a conformity test and a disobedience task to study how children are influenced by a talking doll. We found that the doll could influence children to change their judgments about moral transgressions, however it was unsuccessful in persuading children to disobey an instruction. Finally, we analyzed children's perceptions of the smart toy and discusses implications of this work for future child-agent interaction.

Most hospitals make efforts to provide socio-emotional support for patients and their families during care. In order to expand the service provided by certified child life specialists, we created a social robot and a virtual avatar that augment part of the care CCLS offers to patients by engaging pediatric patients in playful interactions and promoting their socio-emotional wellbeing. We ran a randomized controlled trial in a form of a Wizard-of-Oz study at a local pediatric hospital to study how three different interactive media (a plush teddy bear, a virtual agent on a screen, and a social robot) influence the pediatric patient's affect, joyful play, and social interactions with others. Behavioral analyses of verbal utterance transcriptions and children's physical behavior revealed that the social robot is most effective in producing socially energetic conversations as well as increasing positivity and promoting multi-party interactions. The virtual avatar was socially engaging but children tended to attend more exclusively to a virtual avatar and were less responsive to others. The plush toy was least engaging of the three interventions, but children touched it the most. Based on these findings, we recommend use cases for each agent appropriate for individual pediatric patients' health conditions and needs. These analyses of behavioral data suggest the benefit of deploying a physically embodied social robot in pediatric inpatient-care contexts on young patients'; social and emotional wellbeing.

Understanding social-emotional behaviors in storytelling interactions plays a critical role in the development of interactive educational technologies for children. A challenge when designing for such interactions using technology like social robots, virtual agents, and tablets is understanding the social-emotional behaviors pertinent to storytelling-especially when emulating a natural peer-to-peer relation between the child and the technology. We present P2PSTORY, a dataset of young children (5-6 years old) engaging in natural peer-to-peer storytelling interactions with fellow classmates. The dataset consists of rich social behaviors of children without adult supervision, with each participant demonstrating being a storyteller and a listener. The dataset contains 58 video recorded sessions along with a diverse set of behavioral annotations as well as developmental and demographic profiles of each child participant. We describe the main characteristics of the dataset in addition to findings that reveal perceptual differences between adults and children when evaluating the attentiveness of listeners.

Autonomous technology is becoming more prevalent in our daily lives. We investigated how children perceive this technology by studying how 26 participants (3-10 years old) interact with Amazon Alexa, Google Home, Cozmo, and Julie Chatbot. We refer to them as "agents" in the context of this paper. After playing with the agents, children answered questions about trust, intelligence, social entity, personality, and engagement. We identify four themes in child-agent interaction: perceived intelligence, identity attribution, playfulness and understanding. Our findings show how different modalities of interaction may change the way children perceive their intelligence in comparison to the agents'. We also propose a series of design considerations for future child-agent interaction around voice and prosody, interactive engagement and facilitating understanding.

The fields of Social Robotics and Human-Robot Interaction are undergoing rapid growth, motivated by important societal challenges facing the general public such as aging in place, healthcare, education, manufacturing, transportation, and much more. Such applications motivate the development of ever more intelligent and capable autonomous robots and technologies that can work collaboratively with people in human environments. In the consumer marketplace, intelligent conversational technologies are entering the home at a surprising rate as Internet of Things devices are enabled by AI-based SDKs. Affordable social robots are also poised to enter the market as mass consumer products and 3rd party developer platforms. The importance of developing and understanding the beneficial and longitudinal impact of such technologies on daily life never been more relevant. This keynote presentation highlights a number of provocative research findings from the Personal Robots Group at the MIT Media Lab along these themes. We develop social robots and apply them as a new kind of scientific tool to understand human behavior. We then use these insights to design and develop social robots that can longitudinally engage people to enhance quality of life outcomes. In this presentation, we highlight relevant work in healthcare and early childhood education.

While there has been a growing body of work in child-robot interaction, we still have very little knowledge regarding young children's speaking and listening dynamics and how a robot companion should decode these behaviors and encode its own in a way children can understand. In developing a backchannel prediction model based on observed nonverbal behaviors of 4-6 year-old children, we investigate the effects of an attentive listening robot on a child's storytelling. We provide an extensive analysis of young children's nonverbal behavior with respect to how they encode and decode listener responses and speaker cues. Through a collected video corpus of peer-to-peer storytelling interactions, we identify attention-related listener behaviors as well as speaker cues that prompt opportunities for listener backchannels. Based on our findings, we developed a backchannel opportunity prediction (BOP) model that detects four main speaker cue events based on prosodic features in a child's speech. This rule-based model is capable of accurately predicting backchanneling opportunities in our corpora. We further evaluate this model in a human-subjects experiment where children told stories to an audience of two robots, each with a different backchanneling strategy. We find that our BOP model produces contingent backchannel responses that conveys an increased perception of an attentive listener, and children prefer telling stories to the BOP model robot.

Mindset has been shown to have a large impact on people's academic, social, and work achievements. A growth mindset, i.e., the belief that success comes from effort and perseverance, is a better indicator of higher achievements as compared to a fixed mindset, i.e., the belief that things are set and cannot be changed. Interventions aimed at promoting a growth mindset in children range from teaching about the brain's ability to learn and change, to playing computer games that grant brain points for effort rather than success. This work explores a novel paradigm to foster a growth mindset in young children where they play a puzzle solving game with a peer-like social robot. The social robot is fully autonomous and programmed with behaviors suggestive of it having either a growth mindset or a neutral mindset as it plays puzzle games with the child. We measure the mindset of children before and after interacting with the peer-like robot, in addition to measuring their problem solving behavior when faced with a challenging puzzle. We found that children who played with a growth-mindset robot 1) self-reported having a stronger growth mindset and 2) tried harder during a challenging task, as compared to children who played with the neutral-mindset robot. These results suggest that interacting with peer-like social robot with a growth mindset can promote the same mindset in children.

This paper investigates how a robot that can produce contingent listener response, i.e., backchannel, can deeply engage children as a storyteller. We propose a backchannel opportunity prediction (BOP) model trained from a dataset of children's dyad storytelling and listening activities. Using this dataset, we gain better understanding of what speaker cues children can decode to find backchannel timing, and what type of nonverbal behaviors they produce to indicate engagement status as a listener. Applying our BOP model, we conducted two studies, within- and between-subjects, using our social robot platform, Tega. Behavioral and self-reported analyses from the two studies consistently suggest that children are more engaged with a contingent backchanneling robot listener. Children perceived the contingent robot as more attentive and more interested in their story compared to a non-contingent robot. We find that children significantly gaze more at the contingent robot while storytelling and speak more with higher energy to a contingent robot.

Children and their parents may undergo challenging experiences when admitted for in-patient care at pediatric hospitals. While most pediatric hospitals make an effort to provide socio-emotional support for patients and their families during care, such as with child life services, gaps still exist between professional resource supply and patient demand. There is an opportunity to apply interactive companion-like technologies as a way to augment and extend professional care teams. To explore the opportunity of social robots to augment child life services, we performed a randomized clinical trial at a local pediatric hospital to investigate how three different companion-like interventions (a plush toy, a virtual character on a screen, and a social robot) affected child-patients physical activity and social engagement - both linked to positive patient outcomes. We recorded video of patients, families and a certified child life specialist with each intervention to gather behavioral data. Our results suggest that children are the most physically and verbally engaged when interacting with the physically co-present social robot over time than the other two interventions. A post-study interview with child life specialists reveals their perspective on potential opportunities for social robots (and other companion-like interventions) to assist them with providing education, diversion, and companionship in the pediatric inpatient care context.

Though substantial research has been dedicated towards using technology to improve education, no current methods are as effective as one-on-one tutoring. A critical, though relatively understudied, aspect of effective tutoring is modulating the student's affective state throughout the tutoring session in order to maximize long-term learning gains. We developed an integrated experimental paradigm in which children play a second-language learning game on a tablet, in collaboration with a fully autonomous social robotic learning companion. As part of the system, we measured children's valence and engagement via an automatic facial expression analysis system. These signals were combined into a reward signal that fed into the robot's affective reinforcement learning algorithm. Over several sessions, the robot played the game and personalized its motivational strategies (using verbal and non-verbal actions) to each student. We evaluated this system with 34 children in preschool classrooms for a duration of two months. We saw that (1) children learned new words from the repeated tutoring sessions, (2) the affective policy personalized to students over the duration of the study, and (3) students who interacted with a robot that personalized its affective feedback strategy showed a significant increase in valence, as compared to students who interacted with a non-personalizing robot. This integrated system of tablet-based educational content, affective sensing, affective policy learning, and an autonomous social robot holds great promise for a more comprehensive approach to personalized tutoring.

We developed a smartphone application that detects users' affect and provides personalized positive psychology interventions in order to enhance users' psychological wellbeing. Users' emotional states were measured by analyzing facial expressions and the sentiment of SMS messages. A virtual character in the application prompted users to verbally journal about their day by providing three positive psychology interventions. The system used a Markov Decision Process (MDP) model and a State-Action-Reward-State-Action (SARSA) algorithm to learn users' preferences about the positive psychology interventions. Nine participants were recruited for an experimental study to test the application. They used it daily for three weeks. The interactive journaling activity increased participants' arousal and valence levels immediately following each interaction, and we saw a trend toward improved self-acceptance levels over the three week period. The interaction duration increased significantly throughout the study as well. The qualitative analysis on journal entries showed that the application users explored and reflected on various aspects of themselves by looking at daily events, and found novel appreciation for and meanings in their daily routine.

Teleoperation or Wizard-of-Oz control of social robots is commonly used in human-robot interaction (HRI) research. This is especially true for child-robot interactions, where technologies like speech recognition (which can help create autonomous interactions for adults) work less well. We propose to study young children's understanding teleoperation, how they conceptualize social robots in a learning context, and how this affects their interactions. Children will be told about the teleoperator's presence either before or after an interaction with a social robot. We will assess children's behavior, learning, and emotions before, during, and after the interaction. Our goal is to learn whether children's knowledge about the teleoperator matters (e.g., for their trust and for learning outcomes), and if so, how and when it matters most (e.g. at what age).

We deployed an autonomous social robotic learning companion in three preschool classrooms at an American public school for two months. Before and after this deployment, we asked the teachers and teaching assistants who worked in the classrooms about their views on the use of social robots in preschool education. We found that teachers' expectations about the experience of having a robot in their classrooms often did not match up with their actual experience. These teachers generally expected the robot to be disruptive, but found that it was not, and furthermore, had numerous positive ideas about the robot's potential as a new educational tool for their classrooms. Based on these interviews, we provide a summary of lessons we learned about running child-robot interaction studies in preschools. We share some advice for future researchers who may wish to engage teachers and schools in the course of their own human-robot interaction work. Understanding the teachers, the classroom environment, and the constraints involved is especially important for microgenetic and longitudinal studies, which require more of the school's time-as well as more of the researchers' time-and is a greater opportunity investment for everyone involved.

Tega is a new expressive “squash and stretch”, Android-based social robot platform, designed to enable long-term interactions with children.

Framing or priming a situation can subtly influence how a person reacts to or thinks about the situation. In this paper, we describe a recent study and some preliminary results in which the framing of a robot is manipulated such that it is presented as a social agent or as a machine-like entity. We ask whether framing the robot in these ways influences young children's social behavior during an interaction with the robot, independent of any changes in the robot itself. Following the framing manipulation, children play a fifteen-minute game with the robot. Their behavior, such as the amount of conversation, mimicry of the robot, and various courteous, prosocial actions will be coded and compared across conditions.

Extensive work focuses on the uses of technology at scale for post-literate populations (e.g., MOOC, learning games, Learning Management Systems). Little attention is afforded to non-literate populations, particularly in the developing world. This paper presents an approach using mobile devices with the ultimate goal to reach 770 million people. We developed a novel platform with a cloud backend to deliver educational content to over a thousand marginalized children in different countries: specifically, in remote villages without schools, urban slums with overcrowded schools, and at-risk, rural schools. Here we describe the theoretical basis of our system and results from case studies in three educational contexts. This model will help researchers and designers understand how mobile devices can help children acquire basic skills and aid each other's learning when the benefit of teachers is limited or non-existent.

The presentation or framing of a situation-such as how something or someone is introduced-can influence people's subsequent behavior. In this paper, we describe a study in which we manipulated how a robot was introduced, framing it as either a social agent or as a machine-like being. We asked whether framing the robot in these ways would influence young children's social behavior while playing a ten-minute game with the robot. We coded children's behavior during the robot interaction, including their speech, gaze, and various courteous, prosocial actions. We found several subtle differences in children's gaze behavior between conditions that may reflect children's perceptions of the robot's status as more, or less, of a social actor. In addition, more parents of children in the Social condition reported that their children acted less shy and more talkative with the robot that parents of children in the Machine condition. This study gives us insight into how the interaction context can influence how children think about and respond to social robots.

Effective tutoring requires personalization of the interaction to each student.Continuous and efficient assessment of the student's skills are a prerequisite for such personalization.We developed a Bayesian active-learning algorithm that continuously and efficiently assesses a child's word-reading skills and implemented it in a social robot.We then developed an integrated experimental paradigm in which a child plays a novel story-creation tablet game with the robot.The robot is portrayed as a younger peer who wishes to learn to read, framing the assessment of the child's word-reading skills as well as empowering the child.We show that our algorithm results in an accurate representation of the child's word-reading skills for a large age range, 4-8 year old children, and large initial reading skill range.We also show that employing child-specific assessment-based tutoring results in an age- and initial reading skill-independent learning, compared to random tutoring.Finally, our integrated system enables us to show that implementing the same learning algorithm on the robot's reading skills results in knowledge that is comparable to what the child thinks the robot has learned.The child's perception of the robot's knowledge is age-dependent and may facilitate an indirect assessment of the development of theory-of-mind.

Driving is an attention-demanding task, especially with children in the back seat. While most recommendations prefer to reduce children's screen time in common entertainment systems, e.g. DVD players and tablets, parents often rely on these systems to entertain the children during car trips. These systems often lack key components that are important for modern parents, namely, sociability and educational content. In this contribution we introduce PANDA, a parental affective natural driving assistant. PANDA is a virtual in-car entertainment agent that can migrate around the car to interact with the parent-driver or with children in the back seat. PANDA supports the parent-driver via speech interface, helps to mediate her interaction with children in the back seat, and works to reduce distractions for the driver while also engaging, entertaining and educating children. We present the design of PANDA system and preliminary tests of the prototype system in a car setting.

Designing toolkits for teaching programming concepts to children using robots has received growing attention in recent years. However, teaching preschool children computational concepts, such as non-determinism and event-based programming, presents particular challenges. We have developed a programming toolkit that is embedded in an interpersonal interaction context with a social robot. The toolkit enables young children to program social robots by "teaching" them to interact. In order to "teach", the children show the robot rules designed with reusable vinyl stickers. In doing so, children can experiment with computational concepts while having a playful interaction with the social robot. We present the purpose, context, and design of the social robot toolkit (SoRo Toolkit), and an evaluation performed with 22 preschool children. We show that children have an engaging experience designing and "teaching" social interaction rules to the robot, orchestrating give-and-take exchanges, and delighting in how the robot engages with them as they explore computational ideas.

We present the design of the Huggable robot that can playfully interact with children and provide socio-emotional support for them in pediatric care context. Our design takes into consideration that many young patients are nervous, intimidated, and are socio-emotionally vulnerable at hospitals. The Huggable robot has a childish and furry look be perceived friendly and can perform swift and smooth motions. It uses a smart phone device for its computational power and internal sensors. The robot's haptic sensors perceive physical touch and can use the information in meaningful ways. The modular arm component allows easy sensor replacement and increases the usability of the Huggable robot for various pediatric care services. From a preliminary pilot user study with two healthy and two ill children, all participants enjoyed playing with the robot but the two children with medical conditions showed caring and empathetic behaviors than the two health children. We learned various types of physical touch occurred during the child-robot interaction, and will continue to develop more intelligent haptic sensory system for the Huggable robot to better assist and support child patients' socio-emotional needs.

Curiosity is key to learning, yet school children show wide variability in their eagerness to acquire information. Recent research suggests that other people have a strong influence on children's exploratory behavior. Would a curious robot elicit children's exploration and the desire to find out new things? In order to answer this question we designed a novel experimental paradigm in which a child plays an education tablet app with an autonomous social robot, which is portrayed as a younger peer. We manipulated the robot's behavior to be either curiosity-driven or not and measured the child's curiosity after the interaction. We show that some of the child's curiosity measures are significantly higher after interacting with a curious robot, compared to a non-curious one, while others do not. These results suggest that interacting with an autonomous social curious robot can selectively guide and promote children's curiosity.

People have been shown to project lifelike attributes onto robots and to display behavior indicative of empathy in human-robot interaction. Our work explores the role of empathy by examining how humans respond to a simple robotic object when asked to strike it. We measure the effects of lifelike movement and stories on people's hesitation to strike the robot, and we evaluate the relationship between hesitation and people's trait empathy. Our results show that people with a certain type of high trait empathy (empathic concern) hesitate to strike the robots. We also find that high empathic concern and hesitation are more strongly related for robots with stories. This suggests that high trait empathy increases people's hesitation to strike a robot, and that stories may positively influence their empathic responses.

Children's oral language skills in preschool can predict their academic success later in life. As such, increasing children's skills early on could improve their success in middle and high school. To this end, we propose that a robotic learning companion could supplement children's early language education. The robot targets both the social nature of language learning and the adaptation necessary to help individual children. The robot is designed as a social character that interacts with children as a peer, not as a tutor or teacher. It will play a storytelling game, during which it will introduce new vocabulary words, and model good story narration skills, such as including a beginning, middle, and end; varying sentence structure; and keeping cohesion across the story. We will evaluate whether adapting the robot's level of language to the child's - so that, as children improve their storytelling skills, so does the robot - influences (i) whether children learn new words from the robot, (ii) the complexity and style of stories children tell, (iii) the similarity of children's stories to the robot's stories. We expect children will learn more from a robot that adapts to maintain an equal or greater ability than the children, and that they will copy its stories and narration style more than they would with a robot that does not adapt (a robot of lesser ability). However, we also expect that playing with a robot of lesser ability could prompt teaching or mentoring behavior from children, which could also be beneficial to language learning.

We report observations on how children in a developing country respond to a literacy tablet that is designed to initiate and scaffold literacy learning toward self-sufficiency. This paper describes our first lessons from developing an educational system for enabling children who have no access to schooling to read with minimal outside intervention. We share lessons learned from challenges in the design process, discuss implementation considerations for deploying in remote developing areas, and discuss observations of how children use the technology. We reflect how this experience impacts ongoing work on developing countries regarding collaboration and literacy learning.

People are increasingly working with robots in teams and recent research has focused on how human-robot teams function, but little attention has yet been paid to the role of social signaling behavior in human-robot teams. In a controlled experiment, we examined the role of backchanneling and task complexity on team functioning and perceptions of the robots' engagement and competence. Based on results from 73 participants interacting with autonomous humanoid robots as part of a human-robot team (one participant, one confederate, and three robots), we found that when robots used backchanneling team functioning improved and the robots were seen as more engaged. Ironically, the robots using backchanneling were perceived as less competent than those that did not. Our results suggest that backchanneling plays an important role in human-robot teams and that the design and implementation of robots for human-robot teams may be more effective if backchanneling capability is provided.

Research from the past two decades indicates that preschool is a critical time for children's oral language and vocabulary development, which in turn is a primary predictor of later academic success. However, given the inherently social nature of language learning, it is difficult to develop scalable interventions for young children. Here, we present one solution in the form of robotic learning companions, using the DragonBot platform. Designed as interactive, social characters, these robots combine the flexibility and personalization afforded by educational software with a crucial social context, as peers and conversation partners. They can supplement teachers and caregivers, allowing remote operation as well as the potential for autonomously participating with children in language learning activities. Our aim is to demonstrate the efficacy of the DragonBot platform as an engaging, social, learning companion.

This paper presents AIDA (Affective Intelligent Driving Agent), a social robot that acts as a friendly, in-car companion. AIDA is designed to use the driver's mobile device as its face. The phone displays facial expressions and is the main computational unit to manage information presented to the driver. We conducted an experiment in which participants were placed in a mock in-car environment and completed driving tasks while stress-inducing phone and vehicle notifications occurred throughout the interaction. Users performed the task with the help of: 1) a smartphone, 2) the AIDA persona with the phone mounted on a static dock, or 3) the AIDA persona attached to a robot. Results revealed that AIDA users felt less stressed throughout the interaction, performed vehicle safety precautions more often, and felt more companionship with AIDA as compared to smartphone users. Further, participants developed a deeper bond with AIDA as a social robot compared to AIDA as a static, expressive agent.

Supporting a wide variety of interaction styles across a diverse set of people is a significant challenge in human-robot interaction (HRI). In this work, we explore a data-driven approach that relies on crowdsourcing as a rich source of interactions that cover a wide repertoire of human behavior. We first develop an online game that requires two players to collaborate to solve a task. One player takes the role of a robot avatar and the other a human avatar, each with a different set of capabilities that must be coordinated to overcome challenges and complete the task. Leveraging the interaction data recorded in the online game, we present a novel technique for data-driven behavior generation using case-based planning for a real robot. We compare the resulting autonomous robot behavior against a Wizard of Oz base case condition in a real-world reproduction of the online game that was conducted at the Boston Museum of Science. Results of a post-study survey of participants indicate that the autonomous robot behavior matched the performance of the human-operated robot in several important measures. We examined video recordings of the real-world game to draw additional insights as to how the novice participants attempted to interact with the robot in a loosely structured collaborative task. We discovered that many of the collaborative interactions were generated in the moment and were driven by interpersonal dynamics, not necessarily by the task design. We explored using bids analysis as a meaningful construct to tap into affective qualities of HRI. An important lesson from this work is that in loosely structured collaborative tasks, robots need to be skillful in handling these in-the-moment interpersonal dynamics, as these dynamics have an important impact on the affective quality of the interaction for people. How such interactions dovetail with more task-oriented policies is an important area for future work, as we anticipate such interactions becoming commonplace in situations where personal robots perform loosely structured tasks in interaction with people in human living spaces.

We present the idea and formative design of a blended reality character, a new class of character able to maintain visual and kinetic continuity between the fully physical and fully virtual. The interactive character's embodiment fluidly transitions from an animated character on-screen to a small, alphabet block-shaped mobile robot designed as a platform for informal learning through play. We present the design and results of our study with thirty-four children aged three and a half to seven conducted using non-reactive, unobtrusive observational methods and a validated evaluation instrument. Our claim is that young children have accepted the idea, persistence and continuity of blended reality characters. Furthermore, we found that children are more deeply engaged with blended reality characters and are more fully immersed in blended reality play as co-protagonists in the experience, in comparison to interactions with strictly screen-based representations. As substantiated through the use of quantitative and qualitative analysis of drawings and verbal utterances, the study shows that young children produce longer, detailed and more imaginative descriptions of their experiences following blended reality play. The desire to continue engaging in blended reality play as expressed by children's verbal requests to revisit and extend their play time with the character positively affirms the potential for the development of an informal learning platform with sustained appeal to young children.

This work outlines the development of a reasoning architecture that uses physics-, social-, and agent capability-based knowledge to generate manipulation strategies for a dexterous robot. The architecture learns object affordances through human observations, imposed constraints, and hardcoded physics logic. Human observations are also used to develop a unique manipulation repertoire suitable for the robot. Bayesian Networks are then used to probabilistically determine manipulation strategies for the robot to execute. The robot leverages this knowledge during experimental trials where manipulation strategies suggested by the reasoning architecture are shown to perform well during new manipulation tasks.

Today, the way children learn to read is very different from the way they learn from playing with toys. Books present static images and text on the page, whereas toys allow for manipulation and interactive exploration of cause-effect relations. What if books were tinkerable? What if children could actively explore and modify a story, through voice and touch, to dynamically explore meaning as conveyed by the relationship of text to illustrated concept? How might this change how books are experienced, explored, and shared between parent and child? How might interactivity support and enhance existing shared reading practices?We report the development of interaction design techniques for encouraging storytelling behavior during shared book reading. The design of our storytelling platform, the TinkRBook, encourages active exploration when parents read to very young children (ages 2--5 years old). Our approach uses findings from in-situ parent-child ethnographies and advice from 24 participatory design interviews with researchers, designers and professionals from relevant domains. We believe that our approach addresses the environmental conditions in which interactive storytelling with preschoolers is most likely to be adopted, and is compatible with current shared reading practices.

We present I/O Stickers, adhesive sensors and actuators that children can use to handcraft personalized remote communication interfaces. By attaching I/O Stickers to special wirelessly connected greeting cards, children can invent ways to communicate with long-distance loved ones. Children decorate these cards with their choice of craft materials, creatively expressing themselves while making a functioning interface. The low-bandwidth connections -- simple actuators that change as the sensor stickers are manipulated -- leave room not only to design the look and function of the card, but also to decide how to interpret the information transmitted. We aim to empower children to implement ideas that would otherwise require advanced electronics knowledge. In addition, we hope to support creative learning about communication and to make keeping in touch playful and meaningful. In this paper, we describe the design of the I/O Stickers, analyze a variety of artifacts children have created, and explore future directions for the toolkit.

TofuDraw combines an expressive semi-autonomous robot character (called Tofu) with a new mixed reality DigitalPaint interface whereby children can draw a "program" on the floor that governs the robot character's behavior. Initial evaluations of the TofuDraw system with children ages 3--8 suggest that children can successfully use this interface to choreograph the expressive robot's behavior. Our ultimate goal for this tool is to enable young children to engage in STEM learning experiences in new contexts such as creating interactive robot theatre performances.

In interactive storytelling systems, we see common challenges of artistic expression that pertains to presentation, standing apart from narrative structure. We believe this expression can be achieved computationally, which is a core challenge in using procedurally-generated worlds in interactive storytelling. This computational expression so is what we call dynamic dramaturgy. We intend dynamic dramaturgy as a complement to interactive narrative systems, particularly drama management, and as a fundamentally distinct task from plot-level narrative construction, yet it is still a basic medium for artistic expression by an author. It is, in effect, an art of presentation in interactive storytelling.

We describe the design and evaluation of Chaski, a robot plan execution system that uses insights from human-human teaming to make human-robot teaming more natural and fluid. Chaski is a task-level executive that enables a robot to collaboratively execute a shared plan with a person. The system chooses and schedules the robot's actions, adapts to the human partner, and acts to minimize the human's idle time.

The ability for robots to engage in interactive behavior with a broad range of people is critical for future development of social robotic applications. In this paper, we propose the use of online games as a means of generating large-scale data corpora for human-robot interaction research in order to create robust and diverse interaction models. We describe a data collection approach based on a multiplayer game that was used to collect movement, action and dialog data from hundreds of online users. We then study how these records of human-human interaction collected in a virtual world can be used to generate contextually correct social and task-oriented behaviors for a robot collaborating with a human in a similar real-world environment. We evaluate the resulting behavior model using a physical robot in the Boston Museum of Science, and show that the robot successfully performs the collaborative task and that its behavior is strongly influenced by patterns in the crowdsourced dataset.

We describe an interactive, mixed reality (MR) robot gaming platform in which the user controls a tangible, physically embodied character. Miso, an expressive tele-operated robot plays with its virtual peers by passing a graphical object back and forth seamlessly through an integrated physical and virtual environment. Special emphasis is placed on the importance of maintaining perceptual continuity by closely coupling the simulated world's physical laws to our material reality. We present our implemented MR robot gaming environment and describe the design of an interreality portal at the boundary of the physical and virtual realities.

Telepresence refers to a set of technologies that allow users to feel present at a distant location; telerobotics is a subfield of telepresence. This paper presents the design and evaluation of a telepresence robot which allows for social expression. Our hypothesis is that a telerobot that communicates more than simply audio or video but also expressive gestures, body pose and proxemics, will allow for a more engaging and enjoyable interaction. An iterative design process of the MeBot platform is described in detail, as well as the design of supporting systems and various control interfaces. We conducted a human subject study where the effects of expressivity were measured. Our results show that a socially expressive robot was found to be more engaging and likable than a static one. It was also found that expressiveness contributes to more psychological involvement and better cooperation.

Robotic companions offer a unique combination of embodiment and computation which open many new interesting opportunities in the field of pediatric care. As these new technologies are developed, we must consider the central research questions of how such systems should be designed and what the appropriate applications for such systems are. In this paper we present the Huggable, a robotic companion in the form factor of a teddy bear and outline a series of studies we are planning to run using the Huggable in a pediatric care unit.

The TOFU project introduces a robotic platform for enabling new opportunities in robot based learning with emphasis on storytelling and artistic expression. This project introduces a socially expressive robot character designed to mimic the expressive abilities of animated characters. This demonstration proposal describes the expressive abilities and operator interface to the TOFU project. In this demonstration session, participants will have the opportunity to physically interact with the TOFU project and puppeteer the behavior of the robotic character through a simple joystick interface.

The emerging field of Human-Robot Interaction is undergoing rapid growth, motivated by important societal challenges and new applications for personal robotic technologies for the general public. In this talk, I highlight several projects from my research group to illustrate recent research trends to develop socially interactive robots that work and learn with people as partners. An important goal of this work is to use interactive robots as a scientific tool to understand human behavior, to explore the role of physical embodiment in interactive technology, and to use these insights to design robotic technologies that can enhance human performance and quality of life. Throughout the talk I will highlight synergies with HCI and connect HRI research goals to specific applications in healthcare, education, and communication.

This paper describes the hardware and algorithms for a realtime social touch gesture recognition system. Early experiments involve a sensate bear test-rig with full body touch sensing, sensor visualization and gesture recognition capabilities. Algorithms are based on real humans interacting with a plush bear. In developing a preliminary gesture library with thirteen symbolic gestures and eight touch subtypes, we have taken the first steps toward a robotic touch API, showing that the huggable robot behavior system will be able to stream currently active sensors to detect regional social gestures and local sub-gestures in realtime. The system demonstrates the infrastructure to detect three types of touching: social touch, local touch, and sensor-level touch.

Persuasive Robotics is the study of persuasion as it applies to human-robot interaction (HRI). Persuasion can be generally defined as an attempt to change another's beliefs or behavior. The act of influencing others is fundamental to nearly every type of social interaction. Any agent desiring to seamlessly operate in a social manner will need to incorporate this type of core human behavior. As in human interaction, myriad aspects of a humanoid robot's appearance and behavior can significantly alter its persuasiveness - this work will focus on one particular factor: gender. In the current study, run at the Museum of Science in Boston, subjects interacted with a humanoid robot whose gender was varied. After a short interaction and persuasive appeal, subjects responded to a donation request made by the robot, and subsequently completed a post-study questionnaire. Findings showed that men were more likely to donate money to the female robot, while women showed little preference. Subjects also tended to rate the robot of the opposite sex as more credible, trustworthy, and engaging. In the case of trust and engagement the effect was much stronger between male subjects and the female robot. These results demonstrate the importance of considering robot and human gender in the design of HRI.

We have developed a cognitive architecture for robotic teammates based on the neuro-psychological principles of perceptual symbols and simulation, with the aim of attaining increased fluency in human-robot teams. An instantiation of this architecture was implemented on a robotic desk lamp, performing in a human-robot collaborative task. This paper describes initial results from a human-subject study measuring team efficiency and team fluency, in which the robot works on a joint task with untrained subjects. We find significant differences in a number of efficiency and fluency metrics, when comparing our architecture to a purely reactive robot with similar capabilities.

We present a learning mechanism, socially guided exploration, in which a robot learns new tasks through a combination of self-exploration and social interaction. The system's motivational drives (novelty, mastery), along with social scaffolding from a human partner, bias behavior to create learning opportunities for a reinforcement learning mechanism. The system is able to learn on its own, but can flexibly use the guidance of a human partner to improve performance. An experiment with non-expert human subjects shows a human is able to shape the learning process through suggesting actions and drawing attention to goal states. Human guidance results in a task set that is significantly more focused and efficient, while self exploration results in a broader set.

Spatial scaffolding is a naturally occurring human teaching behavior, in which teachers use their bodies to spatially structure the learning environment to direct the attention of the learner. Robotic systems can take advantage of simple, highly reliable spatial scaffolding cues to learn from human teachers. We present an integrated robotic architecture that combines social attention and machine learning components to learn tasks effectively from natural spatial scaffolding interactions with human teachers. We evaluate the performance of this architecture via a human subjects experiment which examines our humanoid robotpsilas ability to learn from live interactions with human teachers in a secret-constraint task domain. This evaluation provides quantitative evidence for the utility of spatial scaffolding cues to systems that learn from natural human teaching behavior.

Human-robot interaction (HRI) is now well enough understood to allow us to build useful systems that can function outside of the laboratory. We are studying long-term interaction in natural user environments and describe the implementation of a robot designed to help individuals effect behavior change while dieting. Our robotic weight loss coach is compared to a standalone computer and a paper log in a controlled study. We describe the software model used to create successful long-term HRI. We summarize the experimental design, analysis, and results of our study, the first where a sociable robot interacts with a user to achieve behavior change. Results show that participants track their calorie consumption and exercise for nearly twice as long when using the robot than with the other methods and develop a closer relationship with the robot. Both are indicators of longer-term success at weight loss and maintenance and show the effectiveness of sociable robots for long-term HRI.

This paper describes a robotic puppeteering system used in a theatrical production involving one robot and two human performers on stage. We draw from acting theory and human-robot interaction to develop a hybrid-control puppeteering interface which combines reactive expressive gestures and parametric behaviors with a point-of-view eye contact module. Our design addresses two core considerations: allowing a single operator to puppeteer the robotpsilas full range of behaviors, and allowing for gradual replacement of human-controlled modules by autonomous subsystems. We wrote a play specifically for a performance between two humans and one of our research robots, a robotic lamp which embodied a lead role in the play. We staged three performances with the robot as part of a local festival of new plays. Though we have yet to perform a formal statistical evaluation of the system, we interviewed the actors and director and present their feedback about working with the system.

Robots as an embodied, multi-modal technology have great potential to be used as a new type of communication device. In this paper we outline our development of the Huggable robot as a semi-autonomous robot avatar for two specific types of remote interaction — family communication and education. Through our discussion we highlight how we have applied six important elements in our system to allow for the robot to function as a richly embodied communication channel.

The emerging field of Human-Robot Interaction is undergoing rapid growth, motivated by important societal challenges and new applications for personal robotic technologies for the general public. In this talk, I highlight several projects from my research group to illustrate recent research trends to develop socially interactive robots that work and learn with people as partners. An important goal of this work is to use interactive robots as a scientific tool to understand human behavior, to explore the role of physical embodiment in interactive technology, and to use these insights to design robotic technologies that can enhance human performance and quality of life. Throughout the talk I will highlight synergies with HCI and connect HRI research goals to specific applications in healthcare, education, and communication.

We present a learning system, socially guided exploration, in which a social robot learns new tasks through a combination of self-exploration and social interaction. The system's motivational drives, along with social scaffolding from a human partner, bias behaviour to create learning opportunities for a hierarchical reinforcement learning mechanism. The robot is able to learn on its own, but can flexibly take advantage of the guidance of a human teacher. We report the results of an experiment that analyses what the robot learns on its own as compared to being taught by human subjects. We also analyse the video of these interactions to understand human teaching behaviour and the social dynamics of the human-teacher/robot-learner system. With respect to learning performance, human guidance results in a task set that is significantly more focused and efficient at the tasks the human was trying to teach, whereas self-exploration results in a more diverse set. Analysis of human teaching behaviour reveals insights of social coupling between the human teacher and robot learner, different teaching styles, strong consistency in the kinds and frequency of scaffolding acts across teachers and nuances in the communicative intent behind positive and negative feedback.

We present RoCo, the first robotic computer designed with the ability to move its monitor in subtly expressive ways that respond to and encourage its user's own postural movement. We use RoCo in a novel user study to explore whether a computer's "posture" can in fluence its use''s subsequent posture, and if the interaction of the user's body state with their affective state during a task leads to improved task measures such as persistence in problem solving. We believe this is possible in light of new theories that link physical posture and its in uence on affect and cognition. Initial results with 71 subjects support the hypothesis that RoCo's posture not only manipulates the user's posture, but also is associated with hypothesized posture-affect interactions. Specifically, we found effects on increased persistence on a subsequent cognitive task, and effects on perceived level of comfort.

A crucial skill for fluent action meshing in human team activity is a learned and calculated selection of anticipatory actions. We believe that the same holds for robotic teammates, if they are to perform in a similarly fluent manner with their human counterparts.In this work, we propose an adaptive action selection mechanism for a robotic teammate, making anticipatory decisions based on the confidence of their validity and their relative risk. We predict an improvement in task efficiency and fluency compared to a purely reactive process.We then present results from a study involving untrained human subjects working with a simulated version of a robot using our system. We show a significant improvement in best-case task efficiency when compared to a group of users working with a reactive agent, as well as a significant difference in the perceived commitment of the robot to the team and its contribution to the team's uency and success. By way of explanation, we propose a number of fluency metrics that differ significantly between the two study groups.

When a human learns a new motor skill from a teacher, they learn using multiple channels: They receive high level information aurally about the skill, visual information about how another performs the skill, and at times, tactile information, from a teacher's physical guidance of the student. This research proposes a novel approach, the application of this tactile feedback through a robotic wearable system, while a student tries to learn from a teacher. Initial tests on a 5-DOF robotic suit show a decrease in motion errors of over 20%, and an accelerated learning rate of 7%, both conservative given the system setup and statistically very significant (p les 0.01). This research is intended in use of sports training, motor rehabilitation after neurological damage, dance, postural retraining for health, and many other contexts.

The ability for people to interact with robots and teach them new skills will be crucial to the successful application of robots in everyday human environments. In order to design agents that learn efficiently and effectively from their instruction, it is important to understand how people, that are not experts in Machine Learning or robotics, will try to teach social robots. In prior work we have shown that human trainers use positive and negative feedback differentially when interacting with a reinforcement learning agent. In this paper we present experiments and implementations on two platforms, a robotic and a computer game platform, that explore the asymmetric communicative intents of positive and negative feedback from a human partner, in particular that negative feedback is both about the past and about intentions for future action.

A crucial skill for fluent action meshing in human team activity is a learned and calculated selection of anticipatory actions. We believe that the same holds for robotic teammates, if they are to perform in a similarly fluent manner with their human counterparts. In this work, we describe a model for human-robot joint action, and propose an adaptive action selection mechanism for a robotic teammate, which makes anticipatory decisions based on the confidence of their validity and their relative risk. We conduct an analysis of our method, predicting an improvement in task efficiency compared to a purely reactive process. We then present results from a study involving untrained human subjects working with a simulated version of a robot using our system. We show a significant improvement in best-case task efficiency when compared to a group of users working with a reactive agent, as well as a significant difference in the perceived commitment of the robot to the team and its contribution to the team's fluency and success. By way of explanation, we raise a number of fluency metric hypotheses, and evaluate their significance between the two study conditions.

The sixteen papers in this special section are devoted to human-robot interaction models and tools.

When humans learn a new motor skill from a teacher, they learn using multiple channels. They receive high level information aurally about the skill, visual information about how another performs the skill, and at times, tactile information from the teacher's physical guidance. This research proposes a novel approach where the student receives real-time tactile feedback, simultaneously over all joints, delivered through a wearable robotic system. This tactile feedback can supplement the visual or auditory feedback from the teacher. Our results using a 5-DOF robotic suit show a 27% improvement in accuracy while performing the target motion, and an accelerated learning rate of up to 23%. We report both of these results with high statistical significance (p les 0.01). This research is intended for use in a diverse set of applications including sports training, motor rehabilitation after neurological damage, dance, postural retraining for health, and many others. We call this system tactile interaction for kinesthetic learning (TIKL).

Human-robot interaction research is maturing to the point where we can build systems that interact with people in their daily lives and provide support for particular needs. We propose a sociable robot system to help people who are losing weight to better track their progress. We describe related work in sociable robotics and ubiquitous computing and define the term sociable robot systems as systems that comprise a sociable robot, other technological devices, methods for interaction, and methods for relationship creation and maintenance. The system currently being implemented is described as well as the plans for testing the system in a real-world setting.

Robust joint visual attention is necessary for achieving a common frame of reference between humans and robots interacting multimodally in order to work together on real-world spatial tasks involving objects. We make a comprehensive examination of one component of this process that is often otherwise implemented in an ad hoc fashion: the ability to correctly determine the object referent from deictic reference including pointing gestures and speech. From this we describe the development of a modular spatial reasoning framework based around decomposition and resynthesis of speech and gesture into a language of pointing and object labeling. This framework supports multimodal and unimodal access in both real-world and mixed-reality workspaces, accounts for the need to discriminate and sequence identical and proximate objects, assists in overcoming inherent precision limitations in deictic gesture, and assists in the extraction of those gestures. We further discuss an implementation of the framework that has been deployed on two humanoid robot platforms to date.

In Socially Guided Machine Learning we explore the ways in which machine learning can more fully take advantage of natural human interaction. In this work we are studying the role real-time human interaction plays in training assistive robots to perform new tasks. We describe an experimental platform, Sophie's World, and present descriptive analysis of human teaching behavior found in a user study. We report three important observations of how people administer reward and punishment to teach a simulated robot a new task through Reinforcement Learning. People adjust their behavior as they develop a model of the learner, they use the reward channel for guidance as well as feedback, and they may also use it as a motivational channel.

As robots become an everyday part of the complicated environment of the human world it will be important for such systems to feature a full body sense of touch capable of detecting a wide variety of tactile inputs. Such "sensitive skins" can provide much benefit in human robot interaction, specifically in the realm of robotic companions for therapeutic or service applications. In this paper we present a set of design criteria for how such "skins" should be designed. Based on this criteria, a "skin" which features temperature, force, and electric field sensing is described. Results from early experiments with this skin show how the sensors of the multi-modal skin complement each other and allow the distinction between social and affective classes of touch to be distinguished from touch with physical objects

While reinforcement learning (RL) is not traditionally designed for interactive supervisory input from a human teacher, several works in both robot and software agents have adapted it for human input by letting a human trainer control the reward signal. In this work, we experimentally examine the assumption underlying these works, namely that the human-given reward is compatible with the traditional RL reward signal. We describe an experimental platform with a simulated RL robot and present an analysis of real-time human teaching behavior found in a study in which untrained subjects taught the robot to perform a new task. We report three main observations on how people administer feedback when teaching a robot a task through reinforcement learning: (a) they use the reward channel not only for feedback, but also for future-directed guidance; (b) they have a positive bias to their feedback -possibly using the signal as a motivational channel; and (c) they change their behavior as they develop a mental model of the robotic learner. In conclusion, we discuss future extensions to RL to accommodate these lessons

Nonverbal communication plays an important role in coordinating teammates' actions for collaborative activities. In this paper, we explore the impact of non-verbal social cues and behavior on task performance by a human-robot team. We report our results from an experiment where naive human subjects guide a robot to perform a physical task using speech and gesture. Both self-report via questionnaire and behavioral analysis of video offer evidence to support our hypothesis that implicit non-verbal communication positively impacts human-robot task performance with respect to understandability of the robot, efficiency of task performance, and robustness to errors that arise from miscommunication.

We want to build robots capable of rich social interactions with humans, including natural communication and cooperation. This work explores how imitation as a social learning and teaching process may be applied to building socially intelligent robots, and summarizes our progress toward building a robot capable of learning how to imitate facial expressions from simple imitative games played with a human, using biologically inspired mechanisms. It is possible for the robot to bootstrap from this imitative ability to infer the affective reaction of the human with whom it interacts and then use this affective assessment to guide its subsequent behavior. Our approach is heavily influenced by the ways human infants learn to communicate with their caregivers and come to understand the actions and expressive behavior of others in intentional and motivational terms. Specifically, our approach is guided by the hypothesis that imitative interactions between infant and caregiver, starting with facial mimicry, are a significant stepping-stone to developing appropriate social behavior, to predicting others' actions, and ultimately to understanding people as social beings.

The sense of touch is one of the most important sensory system in humans. This paper describes an initial step toward the realization of a fully "sensitive skin" for robots in which somatic sensors of varying modalities such as touch, temperature, pain, and proprioception combine, as if letters in an alphabet, to create a more vivid depiction of the world and foster richer human robot interactions. We have developed a new "sensitive" hand, covered in a lifelike silicone "skin" to explore the importance of touch and the formation of the somatic alphabet in the context of our humanoid robot, Leonardo. From initial tests the populations of these sensors show the potential for similar performance to both the mechanoreceptors in human skin and the cortical neurons in the somatosensory cortex.

Social robots are robots that help people as capable partners rather than as tools, are believed to be of greatest use for applications in entertainment, education, and healthcare because of their potential to be perceived as trusting, helpful, reliable, and engaging. This paper explores how the robot's physical presence influences a person's perception of these characteristics. The first study reported here demonstrates the differences between a robot and an animated character in terms a person's engagement and perceptions of the robot and character. The second study shows that this difference is a result of the physical presence of the robot and that a person's reactions would be similar even if the robot is not physically collocated. Implications to the design of socially communicative and interactive robots are discussed.

We view the problem of machine learning as a collaboration between the human and the machine. Inspired by human-style tutelage, we situate the learning problem within a dialog in which social interaction structures the learning experience, providing instruction, directing attention, and controlling the complexity of the task. We present a learning mechanism, implemented on a humanoid robot, to demonstrate that a collaborative dialog framework allows a robot to efficiently learn a task from a human, generalize this ability to a new task configuration, and show commitment to the overall goal of the learned task. We also compare this approach to traditional machine learning approaches.

The growing field of human-robot interaction (HRI) demands robots that move fluidly, gracefully, compliantly and safely. This paper describes our recent work in the design and evaluation of long-travel voice coil actuators (VCAs) for use in robots intended for interacting with people. The basic advantages and shortcomings of electromagnetic actuators are discussed and evaluated in the context of human-robot interaction, and are compared to alternative actuation technologies. Voice coil actuators have been chosen for their controllability, ease of implementation, geometry, compliance, biomimetic actuation characteristics, safety, quietness, and high power density.

A fully "sensitive skin" can be thought of as the ultimate goal for the application of somatic sensors. This paper describes further work in the creation of a "somatic alphabet" for humanoid robotics. Populations of individual force sensing resistors are combined into receptive fields. This paper details the algorithms used to infer direction of motion of the centroid of a stimulus as well as orientation.

This paper explores the topic of human-robot interaction (HRI) from the perspective of designing sociable autonomous robots-robots designed to interact with people in a human-like way. There are a growing number of applications for robots that people can engage as capable creatures or as partners rather than tools, yet little is understood about how to best design robots that interact with people in this way. The related field of human-computer interaction (HCI) offers important insights, however autonomous robots are a very different technology from desktop computers. In this paper, we look at the field of HRI from an HCI perspective, pointing out important similarities yet significant differences that may ultimately make HRI a distinct area of inquiry. One outcome of this discussion is that it is important to view the design and evaluation problem from the robot's perspective as well as that of the human. Taken as a whole, this paper provides a framework with which to design and evaluate sociable robots from a HRI perspective.

As robot designers, we tend to emphasize the cognitive aspect of intelligence when designing robot architectures while viewing the affective aspect with skepticism. However, scientific studies continue to reveal the deeply intertwined and complementary roles that cognition and emotion play in intelligent decision-making, planning, learning, attention, communication, social interaction, memory, and more. Such findings provide valuable insights and lessons for the design of autonomous robots that must operate in complex and uncertain environments and perform in cooperation with people. This paper presents a concrete implementation of how these insights have guided our work, focusing on the design of sociable autonomous robots that interact with people as capable partners.

This work motivates interactive robot theatre as an interesting test bed to explore research issues in the development of sociable robots and to investigate the relationship between autonomous robots and intelligent environments. We present the implementation of our initial exploration in this area highlighting three core technologies. First, an integrated show control software development platform for the design and control of an intelligent stage. Second, a stereo vision system that tracks multiple features on multiple audience participants in real-time. Third, an interactive, autonomous robot performer with natural and expressive movement that combines techniques from character animation and robot control.

This paper explores the expression of emotion in synthesized speech for an anthropomorphic robot. We adapted several key emotional correlates of human speech to the robot speech synthesizer to allow the robot to speak in either an angry, calm, disgusted, fearful, happy, sad, or surprised manner. We evaluated our approach thorough an acoustic analysis of the speech patterns for each vocal affect and studied how well human subjects perceive the intended affect.

I outline my approach for the design of a robot that can engage humans in a natural, intuitive, social manner. We have found that humans perceive and interpret the robot's actions as socially significant and possessing communicative value. This has led us to define a very different set of constraints and challenges for autonomous robot control that lie along a social dimension. The goal of this research program is to pioneer a path toward the creation of sociable robots. Our journey should be a responsible one, well-conceived and well intentioned. For this reason, it is important that this work raises some of the philosophical and ethical questions regarding how building such technologies shapes our self-understanding, and how these technologies might impact society.

Ballard (1991) described the implications of having a visual system that could actively position the camera coordinates in response to physical stimuli. In humanoid robotic systems, or in any animate vision system that interacts with people, social dynamics provide additional levels of constraint and additional opportunities for processing economy. In this paper, we describe an integrated visual-motor system that was implemented on a humanoid robot to negotiate the robot's physical constraints, the perceptual needs of the robot's behavioral and motivational systems, and the social implications of the motor acts.

In order to interact socially with a human, a robot must convey intentionality, that is, the human must believe that the robot has beliefs, desires, and intentions. We have constructed a robot which exploits natural human social tendencies to convey intentionality through motor actions and facial expressions. We present results on the integration of perception, attention, motivation, behavior, and motor systems which allow the robot to engage in infant-like interactions with a human caregiver.