An official website of the United States government
Theater-based design methods are seeing increased use in social robotics, as embodied roleplay is an ideal method for designing embodied interactions. Yet theater-based design methods are often cast as simply one possible tool; there has been little consideration of the importance of specific improvisational skills for theater-based design; and there has been little consideration of how to train students in theater-based design methods. We argue that improvisation is not just one possible tool of social robot design, but is instead central to social robotics. Leveraging recent theoretical work on Applied Improvisation, we show how improvisational skills represent (1) a set of key capabilities needed for any socially interactive robot, (2) a set of learning objectives for training engineers in social robot design, and (3) a set of methodologies for training those engineers to engage in theater-based design methods. Accordingly, we argue for a reconceptualization of Social Robotics as an Applied Improvisation project; we present, as a speculative pedagogical artifact, a sample syllabus for an envisioned Applied Improvisation driven Social Robotics course that might give students the technical and improvisational skills necessary to be effective robot designers; and we present a case study in which Applied Improvisation methods were simultaneously used (a) by instructors, to rapidly scaffold engineering students’ improvisational skills and (b) by those students, to engage in more effective human-robot interaction design.
more »
« less
This content will become publicly available on April 11, 2026
An LLM-Guided Tutoring System for Social Skills Training
Social skills training targets behaviors necessary for success in social interactions. However, traditional classroom training for such skills is often insufficient to teach effective communication — one-to-one interaction in real-world scenarios is preferred to lecture-style information delivery. This paper introduces a framework that allows instructors to collaborate with large language models to dynamically design realistic scenarios for students to communicate. Our framework uses these scenarios to enable student rehearsal, provide immediate feedback and visualize performance for both students and instructors. Unlike traditional intelligent tutoring systems, instructors can easily co-create scenarios with a large language model without technical skills. Additionally, the system generates new scenario branches in real time when existing options don't fit the student's response.
more »
« less
- Award ID(s):
- 1914635
- PAR ID:
- 10633028
- Publisher / Repository:
- AAAI
- Date Published:
- Journal Name:
- Proceedings of the AAAI Conference on Artificial Intelligence
- Volume:
- 39
- Issue:
- 28
- ISSN:
- 2159-5399
- Page Range / eLocation ID:
- 29643 to 29645
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Scherschel, H.; Rudmann, D.S. (Ed.)Passion-Driven Statistics is a project-based, introductory curriculum implemented as a course in statistics, research methods, data science, a capstone experience, and a summer research boot camp with students from a wide variety of academic settings. Funded by the National Science Foundation, the curriculum engages students in authentic projects with large, real-world data sets. Passion-Driven Statistics students were more likely to report increased confidence in working with data and increased interest in pursuing advanced statistics course work compared to students from the traditional statistics course (Dierker et al., 2018a). This presentation draws on pre/post data from 67 instructors attending a Passion-Driven Statistics faculty development workshop. Analyses evaluate instructor characteristics, attitudes, and experiences that predict its implementation. Findings show that nearly half of the instructors who reported being likely to implement passion-driven statistics and a quarter of the overall sample employed the project-based curriculum by the end of the first full academic year following the workshop. Those showing this fast uptake were more likely to be female than those who had not yet implemented (87.5% vs. 55.3%), were more likely to hold a Ph.D. (94.1% vs. 59.2%), and were more likely to be employed by a private rather than public institution (76.5% vs. 46.0%). Those showing fast uptake were also more likely to have been previously using statistical software (i.e., SAS, JMP, R, Stata, Python, or SPSS) in their target course (70.6% vs. 40.0%) and to have greater prior experience with project-based skills and fewer post workshop concerns about their likely success. Results from these analyses will guide recommendations for 1) engaging instructors ready to implement innovation in learning and teaching and 2) supporting those instructors requiring additional time, training and skills. Instructors are the most important resource for promoting innovations and all materials are open educational resourcesmore » « less
-
Role-play exercises are widely utilized for training across a variety of domains; however, they have many shortcomings, including low availability, resource intensity, and lack of diversity. Large language model-driven virtual agents offer a potential avenue to mitigate these limitations and offer lower-risk role-play. The implications, however, of shifting this human-human collaboration to human-agent collaboration are still largely unexplored. In this work we focus on the context of psychotherapy, as psychotherapists-in-training extensively engage in role-play exercises with peers and/or supervisors to practice the interpersonal and therapeutic skills required for effective treatment. We provide a case study of a realistic ''virtual patient'' system for mental health training, evaluated by trained psychotherapists in comparison to their previous experiences with both real role-play partners and real patients. Our qualitative, reflexive analysis generated three themes and thirteen subthemes regarding key interpersonal skills of psychotherapy, the utility of the system compared to traditional role-play techniques, and factors which impacted psychotherapist-perceived ''humanness'' of the virtual patient. Although psychotherapists were optimistic about the system's potential to bolster therapeutic skills, this utility was impacted by the extent to which the virtual patient was perceived as human-like. We leverage the Computers Are Social Actors framework to discuss human-virtual-patient collaboration for practicing rapport, and discuss challenges of prototyping novel human-AI systems for clinical contexts which require a high degree of unpredictability. We pull from the ''SEEK'' three-factor theory of anthropomorphism to stress the importance of adequately representing a variety of cultural communities within mental health AI systems, in alignment with decolonial computing.more » « less
-
There is increasing awareness that data science and computational thinking are critical skills for undergraduates to develop but these can be difficult to integrate into undergraduate Biology classes. In this module, we describe how we have used a system for learning the programming language R that focuses on building students? skills and confidence in data exploration, management, and visualization. This activity pairs a hands-on virtual experiment where students simulate animal movements and social interactions to provide a friendly introduction to basic data science for biologists. During the activity, students play the ?Bat Game?, an online game which students access via an internet browser. Each student controls the movement decisions of one bat within a social group. The bats must search for cows they can bite to get a meal of blood. Students take the roles of bats in a series of foraging tasks. Students must follow ?rules? and attempt to match their overall actions to those of their group members under different scenarios. The game platform collects all the locations of all bats in the game. After playing the game, students export the data they just created and analyze it to learn how to detect known patterns through basic summaries and plotting in R. All analyses and programming skills are presented in one cohesive R Markdown file, where students can read about the goals of each coding chunk, can run each chunk, and then answer questions about the biology of the social system as well as basic questions about the code used in the analyses. This approach decouples coding from statistics, assumes no prior knowledge, and uses a charismatic species to incentivize student participation. This module can be used in many courses including lab sections of large-enrollment introductory biology courses as well as smaller upper-level coursesmore » « less
-
Abstract Data science skills (e.g., analyzing, modeling, and visualizing large data sets) are increasingly needed by undergraduates in the life sciences. However, a lack of both student and instructor confidence in data science skills presents a barrier to their inclusion in undergraduate curricula. To reduce this barrier, we developed four teaching modules in the Macrosystems EDDIE (for environmental data-driven inquiry and exploration) program to introduce undergraduate students and instructors to ecological forecasting, an emerging subdiscipline that integrates multiple data science skills. Ecological forecasting aims to improve natural resource management by providing future predictions of ecosystems with uncertainty. We assessed module efficacy with 596 students and 26 instructors over 3 years and found that module completion increased students’ confidence in their understanding of ecological forecasting and instructors’ likelihood to work with long-term, high-frequency sensor network data. Our modules constitute one of the first formalized data science curricula on ecological forecasting for undergraduates.more » « less
