Transparent and consistent reporting of user studies is essential for advancing scientific knowledge. In human-robot interaction (HRI), studies are often reported incompletely, even in top-tier venues, limiting proper evaluation, replication, and practical application of findings in practice. This study aimed to generate expert consensus on a reporting checklist for HRI user studies and provide a validated tool to improve transparency, reproducibility, and methodological rigor in the field, leading to easier translation of research into practice. A two-round Delphi study was conducted with 34 HRI experts from academia and industry from over 12 countries. An international panel of nine interdisciplinary experts first synthesized a preliminary list of 116 reporting items from the literature. Experts rated the importance of each item and provided qualitative feed- back. Consensus was defined as 70% agreement, and items were iteratively refined through anonymous online surveys. Overall, consensus was achieved on 106 items, encompassing both essential and context-dependent elements in nine domains. The resulting RUSH checklist (Reporting User Studies in Human-Robot Inter- action) provides the first community-endorsed, consensus-based reporting guideline for HRI user studies.

The study of classroom discourse is essential for enhancing child development and educational outcomes in academic settings. Prior research has focused on the annotation of conversational talk-turns within the classroom, offering a statistical analysis of the various types of discourse prevalent in these environments. In this work, we explore the generalizability and transferability of text classifiers trained to predict these discourse codes across educational domains. We examine two distinct English-language classroom datasets from the domains: literacy and math. Our results show that models exhibit high accuracy and generalizability when the training and test datasets originate from the same or similar domains. In situations where limited training data is available in new domains, few shot and zero shot exhibit more resiliency and aren’t as effected as their supervised counterparts. We also observe that accompanying each talk turn with dialog-level context improves the accuracy of the generative models. We conclude by offering suggestions on how to enhance the generalization of these methods to novel domains, proposing directions for future studies to investigate new methods for boosting the model adaptability across domains.

After-action reviews (AARs) are professional discussions that help operators and teams enhance their task performance by analyzing completed missions with peers and professionals. Previous studies comparing different formats of AARs have focused mainly on human teams. However, the inclusion of robotic teammates brings along new challenges in understanding teammate intent and communication. Traditional AAR between human teammates may not be satisfactory for human-robot teams. To address this limitation, we propose a new training review (TR) tool, called the Virtual Spectator Interface (VSI), to enhance human-robot team performance and situational awareness (SA) in a simulated search mission. The proposed VSI primarily utilizes visual feedback to review subjects' behavior. To examine the effectiveness of VSI, we took elements from AAR to conduct our own TR, and designed a 1$\times 3$between-subjects experiment with experimental conditions: TR with (1) VSI, (2) screen recording, and (3) non-technology (only verbal descriptions). The results of our experiments demonstrated that the VSI did not result in significantly better team performance than other conditions. However, the TR with VSI led to more improvement in the subjects' SA over the other conditions.

In an interdisciplinary and evolving research field like human-robot interaction, clear and precise results reporting is essential for study comparability and replicability. To address the lack of a standard for such reporting and, at the same time, provide guidance for novices in the field, we have developed a web-based reporting form to capture human-robot interaction studies, serving as a model for how conferences could adopt it into the submission pipeline. In this work, we present a formative evaluation of this form regarding its level of detail, format and clarity, and the perceived benefits for authors, reviewers, and the community as a whole. We report the expert review of nine researchers who highlight the substantial value of this tool. In addition, these experts also provide suggestions for improvements to its form and the addition of details surrounding qualitative reporting.

Learning from demonstration (LfD) is a common method for teaching novel tasks to social robots, but non-experts can struggle to teach optimally without understanding the robot’s failures. Interpretability techniques present a possible solution to this problem, but research on interpretability in social robotics has been limited and focuses on improving users’ perceptions of robots rather than helping them understand the robot’s internal model and the true causes of its failures. We address this gap with an online study evaluating how well non-experts can diagnose a social robot’s failures during learning using causal explanations and a novel visual transparency interface. While neither method improved performance for all users, participants who displayed high interaction with the visual interface showed a significantly improved ability to diagnose errors. Our findings suggest visual interfaces may be a promising alternative to causal explanations for teaching social robots and highlight the challenges that remain in helping non-experts understand social robot failures during LfD.

The inexorable progress of technology brought forth an era where robots increasingly integrate into human life which necessitates the understanding of human-robot interactions (HRI). This study unravels the details of HRI within interactive storytelling contexts. Through a between-subject experiment with 28 participants, we assessed response latency and utterance lengths to interactive story narrations delivered by either a human or a robot. Findings indicated that participants displayed longer response latency interacting with the robot narrator while articulating shorter utterances compared to the human condition where participants displayed longer utterances and shorter response latency. These observations suggest significant differences in cognitive and communicative strategies in human-human versus human-robot interactions. The results underscore the challenges and potential of designing social robots that are time-sensitive in interacting with humans. Future explorations should focus on the cognitive and emotional drivers behind these interactions.

Social robots are being introduced in a variety of educational domains with great success. These social robots are often designed by drawing inspiration from practices held by human teachers. Contingent mediations are a prime example of a high-quality teaching practice that supports better outcomes in human-human teaching. This can inform the design of social robots. Current research on robot use in education has focused on curriculum-level contingent mediations where the difficulty level of subsequent tasks are adjusted to the current capabilities of a learner. However, task-level contingent mediations that provide support for students’ learning during a specific task remain unexplored. This research investigates whether patterns of task-level mediations differ between a robot, virtual, and human agent, as well as their effects on learning outcomes. To investigate these research questions, we designed instruction that utilizes contingent mediational flows, based on formative assessment data, to be delivered by a human, robot, and virtual agent to teach grade 3-5 children science words. We identified 23 unique instructional patterns. Then, we compared these patterns across teaching agents and their effects on learning outcomes. Overall, our study demonstrated that high-quality contingent mediations support children’s science vocabulary in learning regardless of the teaching agent.

This qualitative study explored 20 grade 3-5 children’s perceptions of learning STEM vocabulary with an educational social robot. A semi-structured interview protocol was used to elicit children’s perceptions. Interviews were recorded and transcribed into a database reflecting one talk-turn per row (911 talk-turns total). Two coders used emergent coding and constant comparative method to identify talk-turns that reflected children’s perceptions of the assets and drawbacks of learning STEM vocabulary with the robot. Findings identified new assets and drawbacks about the robot’s instruction, and new drawbacks about the robot’s speech, which were not explored in previous research. Findings suggest design implications, including designing robots with the capacity for more individualization of instruction and adjustable movement and speech features based on learners’ preferences.

Backchanneling models, designed to enhance the interactive capabilities of robots, have primarily been trained on human-human interaction data. However, applying such data directly to social robots raises concerns due to dissimilarities in the way humans and robots exhibit verbal and nonverbal behaviors, particularly in the domain of emotional backchannels. This research aims to address this gap by conducting an exploratory study on the differences in human backchanneling behaviors during interactions with humans and social robots in various emotional contexts (e.g., happy and sad). Our findings reveal significant variations in emotionally specific backchannels between human-human and human-robot interactions under different emotional contexts. These results highlight the importance of designing backchanneling models that are tailored for human-robot interactions.

For socially assistive robots to achieve widespread adoption, the ability to learn new tasks in the wild is critical. Learning from Demonstration (LfD) approaches are a popular method for learning in the wild, but current methods require significant amounts of data and can be difficult to interpret. Interactive Task Learning (ITL) is an emerging learning paradigm that aims to teach tasks in a structured manner, minimizing the need for data and increasing transparency. However, to date ITL has only been explored for physical robotics applications. Additionally, minimal research has explored how usable existing ITL systems are for non-expert users. In this work, we propose a novel approach to learn social tasks via ITL. This system utilizes recent advances in Natural Language Understanding (NLU) to learn from natural dialogue. We conducted a pilot study to compare the ITL system against an LfD approach to investigate differences in teaching performance as well as teachers' perceptions of trust and workload towards these systems. Additionally, we analyzed the teaching behavior of participants to identify successful and unsuccessful teaching strategies. Our findings suggest ITL could provide more transparency to users and improve performance by correcting speech recognition errors. However, participants generally preferred LfD and found it an easier teaching method. From the observed teaching behavior, we identify existing challenges in ITL for non-experts to teach social tasks. Using this, we propose areas of improvement toward future ITL learning paradigms that are intuitive, transparent, and performant.

Ethics have become a core consideration in human-robot interaction (HRI) due to ample opportunity for both positive and negative impact on humans. HRI literature has expounded on ways to produce ethical social robots, especially participatory design (PD) that integrates anticipated users and other stakeholders as designers themselves to ensure their values are integrated into robot design. We draw attention to the ethics of participation in robot design, distinct from the ethics of the robot ultimately designed. We propose an approach to foregrounding ethics in PD processes through co-construction of robot PD protocols with stakeholders. We call this ”pre-PD” because it entails expanding the boundaries of PD beyond the product of design (the robot) to also include the participatory activities that enable design. Contributions of the paper include: (1) a case study of pre-PD for sexual violence mitigation robots to demonstrate feasibility of stakeholders co-constructing robot PD protocols, and (2) an actionable framework for HRI researchers to use when constructing their own PD protocols with stakeholders, informed by reflection on the case study.

This study aimed to evaluate the effectiveness of robot-mediated training for job interviews for young adults with autism spectrum disorder (ASD). The six-week intervention involved mock job interviews with a Furhat social robot to target nonverbal behaviors and communication skills. To measure the efficacy of the intervention, four common nonverbal behavioral challenges among individuals with ASD were investigated: eye gaze, excessive body movement, atypical vocalization, and orientation toward the interviewer. Results indicated varying levels of improvement among participants, with some showing consistent improvement and others exhibiting unexpected results from session to session. This underscores the need for personalized, objective, and quantitative analysis. The study highlights the importance of addressing nonverbal communication challenges for individuals with ASD and equipping them with the necessary job market skills. While the pilot results from robot-mediated training appear promising, further research with a larger group including a wider range of participants with ASD is required to generalize the outcomes.

Turn-taking is a fundamental behavior during human interactions and robots must be capable of turn-taking to interact with humans. Current state-of-the-art approaches in turn-taking focus on developing general models to predict the end of turn (EoT) across all contexts. This demands an all-inclusive verbal and non-verbal behavioral dataset from all possible contexts of interaction. Before robot deployment, gathering such a dataset may be infeasible and/or impractical. More importantly, a robot needs to predict the EoT and decide on the best time to take a turn (i.e, start speaking). In this research, we present a learning from demonstration (LfD) system for a robot to learn from demonstrations, after it has been deployed, to make decisions on the appropriate time for taking a turn within specific social interaction contexts. The system captures demonstrations of turn-taking during social interactions and uses these demonstrations to train a LSTM RNN based model to replicate the turn-taking behavior of the demonstrator. We evaluate the system for teaching the turn-taking behavior of an interviewer during a job interview context. Furthermore, we investigate the efficacy of verbal, prosodic, and gestural cues for deciding when to begin a turn.

Learning from demonstration (LfD) approaches have garnered significant interest for teaching social robots a variety of tasks in healthcare, educational, and service domains after they have been deployed. These LfD approaches often require a significant number of demonstrations for a robot to learn a performant model from task demonstrations. However, requiring non-experts to provide numerous demonstrations for a social robot to learn a task is impractical in real-world applications. In this paper, we propose a method to improve the sample efficiency of existing learning from demonstration approaches via data augmentation, dynamic experience replay sizes, and hierarchical Deep Q-Networks (DQN). After validating our methods on two different datasets, results suggest that our proposed hierarchical DQN is effective for improving sample efficiency when learning tasks from demonstration. In the future, such a sample-efficient approach has the potential to improve our ability to apply LfD approaches for social robots to learn tasks in domains where demonstration data is limited, sparse, and imbalanced.

Parents of children with developmental disabilities face significantly higher workloads than parents of neurotypical children due to their higher care giving demands. Consequently, parents of children with developmental disabilities often face emotional, physical, mental, and social health declines. Currently there has been significant research and development of robots for providing care to children with developmental disabilities to address a variety of care giving scenarios. However, it is presently unclear whether parents would be comfortable with robots interacting with their children in these different child-robot interaction scenarios. In this paper, we investigate parental comfort toward robots caring for children with developmental disabilities in a variety of interaction scenarios and the influence of parental negative attitudes toward robots as well as trust on their comfort toward robots in these scenarios. Overall, our findings suggest that US parental attitudes, trust, and comfort toward robots caring for children with developmental disabilities are neutral. Parents were most comfortable with a robot serving as a teaching assistant to children with a developmental disability and least comfortable as a bus driver. Furthermore, trust for robots had a medium positive association with comfort with child-robot interactions and negative attitudes toward robots had a medium negative association with comfort with child-robot interactions.

Children diagnosed with autism spectrum disorder (ASD) typically work towards acquiring skills to participate in a regular classroom setting such as attending and appropriately responding to an instructor’s requests. Social robots have the potential to support children with ASD in learning group-interaction skills. However, the majority of studies that target children with ASD’s interactions with social robots have been limited to one-on-one interactions. Group interaction sessions present unique challenges such as the unpredictable behaviors of the other children participating in the group intervention session and shared attention from the instructor. We present the design of a robot-mediated group interaction intervention for children with ASD to enable them to practice the skills required to participate in a classroom. We also present a study investigating differences in children’s learning behaviors during robot-led and human-led group interventions over multiple intervention sessions. Results of this study suggests that children with ASD’s learning behaviors are similar during human and robot instruction. Furthermore, preliminary results of this study suggest that a novelty effect was not observed when children interacted with the robot over multiple sessions.

Robot-mediated therapy is an emerging field of research seeking to improve therapy for children with Autism Spectrum Disorder (ASD). Current approaches to autonomous robot-mediated therapy often focus on having a robot teach a single skill to children with ASD and lack a personalized approach to each individual. More recently, Learning from Demonstration (LfD) approaches are being explored to teach socially assistive robots to deliver personalized interventions after they have been deployed but these approaches require large amounts of demonstrations and utilize learning models that cannot be easily interpreted. In this work, we present a LfD system capable of learning the delivery of autism therapies in a data-efficient manner utilizing learning models that are inherently interpretable. The LfD system learns a behavioral model of the task with minimal supervision via hierarchical clustering and then learns an interpretable policy to determine when to execute the learned behaviors. The system is able to learn from less than an hour of demonstrations and for each of its predictions can identify demonstrated instances that contributed to its decision. The system performs well under unsupervised conditions and achieves even better performance with a low-effort human correction process that is enabled by the interpretable model.

Socially Assistive Robots (SARs) have demonstrated success in the delivery of interventions to individuals with Autism Spectrum Disorder (ASD). To date, these robot-mediated interventions have primarily been designed and implemented by robotics researchers. It remains unclear whether therapists could independently utilize robots to deliver therapies in clinical settings. In this paper, we conducted a study to investigate whether therapists could design and implement robot-mediated interventions for children with ASD. Furthermore, we compared therapists’ performance, efficiency, and perceptions towards using a Virtual Reality (VR) and kinesthetic-based interface for delivering robot-mediated interventions. Overall, our results demonstrated therapists could independently design and implement interventions with a SAR. They were faster at designing a new intervention using VR than a kinesthetic interface. Therapists also had similar performance to delivering inperson interventions when utilizing VR to deliver interventions with the robot. Therapists reported moderate workload using the VR interface and perceived VR to be usable.

Current studies have demonstrated that Socially Assistive Robots (SARs) delivering Applied Behavior Analysis (ABA) based interventions can teach individuals with Autism Spectrum Disorder (ASD) valuable social, emotional, communication and academic skills. These robot-mediated interventions (RMIs) are typically delivered via teleoperation, which places additional or similar workloads on therapists as administering interventions directly. The autonomous delivery of ABA therapies to individuals with ASD by a robot could significantly reduce workload and improve the usability as well as acceptance of this technology. However, pre-programming the autonomy of a SAR with a limited set of interventions is not sufficient for clinical practice due to the rapidly changing and different learning needs of individuals with ASD. In order to be applicable in clinical settings, therapists must be capable of customizing and personalizing interventions to the needs of each individual. Towards this goal, in this paper we present the initial development and deployment of a proof-of-concept Learning from Demonstration (LfD) system in-the-wild to learn the verbal behavior of therapists during the delivery of an ABA-based intervention to children with ASD. We also present preliminary data on the results of a policy trained on data collected from demonstrations provided during this in-the-wild deployment of our LfD system.

Socially assistive robots (SARs) are currently being developed to assist in the delivery of Applied Behavior Analysis (ABA) therapies to individuals diagnosed with Autism Spectrum Disorder (ASD). Although SARs have demonstrated positive outcomes, minimal research has focused on investigating needs of the therapists that deliver treatments. Therapist perspectives are important as they will likely be the primary end-users of SARs. In this study, we investigated the perceptions and design requirements of ABA therapists towards SARs and the interfaces used to operate them. Therapists were interviewed after they independently designed, developed, and implemented their own robot-mediated interventions. Overall, therapists’ general perceptions towards integrating a SAR within their existing workflow was positive and they expected that children would benefit from ABA therapies delivered by a SAR. The therapists also provided insights on design requirements for utilizing SARs and their interfaces as well as potential clinical and future use cases for this technology.

Autonomous vehicle technology is rapidly developing but the current state-of-the-art still has limitations and requires frequent human intervention. However, handovers from an autonomous vehicle to a human driver are challenging because a human operator may be unaware of the vehicle surroundings during a handover which can lead to dangerous driving outcomes. There is presently an urgent need to develop advanced driver-assistance systems capable of monitoring driver situational awareness within an autonomous vehicle and intelligently handing-over control to a human driver in emergency situations. Towards this goal, in this paper we present the development and evaluation of a vision-based system that identifies visual cues of a driver's situational awareness including their: head pose, eye pupil position, average head movement rate and visual focus of attention.

Group-based recreational activities have shown to have a number of health benefits for people of all ages. The handful of social robots designed to facilitate such activities are currently only able to implement a priori known recreational activities that have been pre-programmed by human experts. Once deployed in their intended facility, these robots are not able to learn new activities from non-expert humans. In this paper, we present the development of a novel learning from demonstration (LfD) system architecture for a social robot in order for it to learn from non-expert teachers the structure of an activity and monitor the execution of the new activity. In order to obtain user compliance, personalized persuasive strategies are also learned by the robot to use while implementing the activity during human-robot interactions (HRI) with the intended users. The architecture has been integrated into our socially assistive robot Tangy to learn the group-based activity Bingo. System performance experiments were conducted with Tangy to first learn to facilitate Bingo from non-expert teachers and then use the learned activity to physically facilitate Bingo with multiple users. The results showed Tangy was able to effectively and efficiently learn the new Bingo activity structure as well as personalize its persuasive strategies to individual users in order to obtain activity compliance.

In this paper we present the development of a novel multi-user human-robot interaction (HRI) system architecture to allow the social robot Tangy to autonomously plan, schedule and facilitate multi-user activities while considering the users' schedules. During scheduled activities, the robot is able to interact with a group of users by providing both group-based and individualized assistance based on the current state of the activity and the needs of the individual users engaged in the social interactions. Such planning and scheduling of daily activities of a social robot while reasoning about multiple user schedules has not yet been addressed in the literature. Herein, the HRI multi-user activities we consider are a series of Bingo games. System performance experiments presented in the paper validate the use of the proposed multiuser system architecture in: 1) planning and scheduling daily Bingo games for Tangy to facilitate while considering the individual schedules of the users, and 2) determining the appropriate behaviors of the robot with respect to individuals and groups of people while providing game reminders prior to a Bingo game starting and also while facilitating the game itself.

As older adults age, they are more likely to reside in long-term care facilities due to the decline in cognitive and/or physical abilities that prevent them from living independently. With a rapidly aging population there is an increasing demand on long-term care facilities to care for older adults. Such facilities need to provide medical services, assistance in activities of daily living, and scheduled leisure activities to improve health and quality of life. However, as the need for long-term care is increasing, the care workforce is faced with decreasing numbers of healthcare staff and high turnover rates. Our research focuses on the design of socially assistive robots to plan, schedule, and facilitate social and cognitive interventions for residents in long-term care facilities. In this paper, we investigate the specific design considerations and the impressions of long-term care residents, healthcare professionals, and family members on a socially assistive robot designed to autonomously facilitate cognitively and socially stimulating leisure activities. Thematic analysis of focus group sessions conducted at a long-term care facility with the aforementioned individuals revealed important design considerations for the development and integration of a socially assistive robot in long-term care facilities.

Studies have shown that cognitive and social stimulation is crucial to the overall health of older adults including psychological, cognitive and physical well-being. However, activities to promote such stimulation are often lacking in long-term care facilities. Our work focuses on the use of social robotic technologies to provide person-centered cognitive interventions. Namely, this paper presents an HRI study with the unique human-like socially assistive robot Brian 2.1, in order to investigate the use and acceptability of the expressive human-like robot by older adults living in a longterm care center. Current studies with social robots for the elderly have been mainly directed towards collecting data on the acceptance and use of animal-like robots. Herein, we aim to determine if the robot's human-like assistive and social characteristics result in the elderly having positive attitudes towards the robot as well as accepting it as an interactive cognitive training tool.