Age-Related Differences in Children’s Spontaneous Gesturing with a Robot versus Human Instructor

(Wilson et al., 2025)

This study examined how children of different ages naturally use gestures when comuicating with social robots versus human instructors. We analyzed pointing, showing, and conventional gestures by 5 to 8 year old children. Our findings reveal that while older children showed decreased rates of show and conventional gestures with humans, their gesture rates remained remarkably consistent across all ages.

ToMCAT: Benchmark for Socially Assistive Robots with Theory of Mind

(Wilson et al., 2025)

This paper introduces a new benchmark for evaluating Theory of Mind capabilities in socially assistive robots. Unlike existing evaluations, ToMCAT features complex goals, multiple solution paths, and real mistakes using tangram puzzle assembly data from real children. We tested baseline performance using an analogical reasoing algorithm and a large language model. The analogical reasoning approach achieved perfect accuracy once at least half of the piece relationships were correctly established, while the LLM correctly identified complete puzzles only 79% of the time and performed even worse on incomplete or error-containing puzzles.

The dataset for this benchmark is now publicly available and can be found at https://github.com/FandM-CARES/ToMCAT

What does a Draw a Robot Task (DART) tell us about how children will interact with a robot?

(Howard et al., 2024)

In August, we presented a short paper exploreing how age and childhood exposure to technology influence DART responses. We also examine how DART results influence subsequent interactions with and attention to a real robot. We find a surprising lack of significant correlations between the DART and other measures, except in the oldest age group of children (7-year-olds). As such, we recommend using this task with older children or supplementing it with other implicit tasks to fully understand early robot perceptions.

How do we tell a robot what to do?

(Elbeleidy & Wilson, 2024)

In September, we presented our work in collaboration with Peerbots on a multi-perspective approach to designing social robot applications. Designing autonomous robots, teleoperated robots, or robots programmed by an end-usere each present a tradeoff between some advantages and limitations, and there is an opportunity to integrate these approaches where people benefit from the best-fit approach for their use case. In this work, we propose integrating these seemingly distinct robot control approaches to uncover a common data representation of social actions defining social expression by a robot. We demonstrate the value of integrating an authoring system, teleoperation interface, and robot planning system by integrating instances of these systems for robot storytelling. By relying on an integrated system, domain experts can define behaviors through end-user interfaces that teleoperators and autonomous robot programmers can use directly thus providing a cohesive expert-driven robot system.

How do we design an inclusive approach to developing machine ethics curriculum?

(Muchuwa et al., 2025)

Recently accepted to the Student Research Competition at SIGCSE TS 2025, F\&M student El Muchuwa descibes our pioneering approaching for an inclusive curriculum design process to broaden accessibility to machine ethics education. In collaboration with Lee Franklin at the F\&M Faculty Center, we are developing an approach that uses a “source” course to develop materials for seven “target” courses. The source course is a machine ethics curriculum development course in which students and faculty collaboratively build curricular materials for integration into non-computer science courses. Here we describe the development of the “source” course using a curriculum co-creation process that leverages student and faculty expertise. The process emphasizes an inclusive design approach, rooted in continuous stakeholder feedback and consistent, transparent communication. The products of this process include course materials that incorporate underrepresented ethical frameworks. Additionally, it features peer-reviewed journal assignments that promote reflective learning and sharing of diverse perspectives, as well as a final module project in which students collaborate with faculty to co-create curricular materials. Our approach aims to broaden a culturally relevant understanding of ethical challenges in technology while ensuring that the curriculum resonates with diverse student backgrounds.

What would be a more comprehensive evalaution of machine theory of mind?

(Wilson et al., 2025)

Recently accepted to be presented at the ToM4AI workshop in March, we propose a new dataset that introduces new challenges by requiring explicit reasoning over beliefs and knowledge. In collaboration with Barnstorm Research and the Navy Research Lab, we argue that while the AI community has made incredible gains in modeling and simulating theory of mind (ToM), there is a new for ToM evaluations to go beyond predicting an agent’s intentions based on observations of their actions. Our new Tangram dataset represents observations of a child assembling a tangram puzzle with the assistance of a human or robot instructor and interventions that the instructor has made to assist the child. Due in part to the intricacies of this task, the dataset supports evaluating ToM capabilities related to goal recognition, plan recognition, and false belief reasoning. Additionally, the dataset supports investigation into how ToM reasoning connects to the subsequent actions of the ToM reasoner. Altogether, the Tangram dataset has qualities that will help push forward AI for ToM.

The starting date is ideally in early summer, but late summer or early fall dates may be considered. The intern must be currently enrolled in a graduate program and will need to be onsite in Lancaster, PA. This is the first year of a 5 year program, and applicants may be invited to reapply for subsequent years.

The CARES Lab investigates cutting edge research at the intersection of human-robot interaction, cognitive science, psychology, artificial intelligence, and ethics. Our goal is to ensure the ethical application of robots and other technologies by better understanding and communicating with the user. We develop explainable algorithms that allow end-to-end inspection of a robot’s decision making, and we design and conduct detailed experiments to understand the impact a robot has on its users. Please check out our lab website for more information: https://fandm-cares.github.io/

For this position, we are looking for highly motivated candidates with strong background in either design and execution of child-robot interaction experiments or symbolic AI algorithm development. The candidate will also be expected to spend a small portion of their time mentoring undergraduate researchers. The ideal candidate is currently enrolled in a PhD program related to human-robot interaction or artificial intelligence and has a track record of quality publications in related venues. Excellent communication (both written and verbal) are essential.

Review of submissions will begin immediately. Interested candidates are encouraged to apply as soon as possible, as the hiring of a successful candidate will happen on a first-come-first-served basis. To apply go to https://fandm.interviewexchange.com/jobofferdetails.jsp;jsessionid=3ADD1E30896651B792E049B2761AB69E?JOBID=175255 Please include your ideal start date in the cover letter.

Through this award, I will be able to support the development of a number of undergraduate researchers over the next five years. Additionally, I will be looking to hire a graduate student intern each summer. This will be a unique opportunity to bring graduate level research to a small liberal arts college.

I’m particularly proud that both the research and education plan for this work emphasizes the ethical application of technology. The research includes developing transparent algorithms that allow for detailed inspection of the reasoning and decision-making processes of a social robot. On the education side, I will be broadening access to critical knowledge on the ethical and social implications of technology by developing curricular materials for courses outside of computer science.

Let me now briefly tell you a little more about the proposed work.

Social robots that assist young children provide new learning opportunities in the classroom. If a robot can reason about a child’s goals, intentions, and beliefs, then it can provide better help. Current approaches in social robotics have not used theory of mind reasoning (ToM). The goal of this research is to develop ToM algorithms. This will allow social robots to infer a child’s goals, intentions, and beliefs. The robot can then respond with more effective help. The algorithms use analogical reasoning to make each inference step. This provides transparency and explainability in the robot’s reasoning. The aim is that children and parents will see the robot as effective, helpful, and trustworthy. The outcome will contribute to more ethical applications of robots. The educational aspect of this project aims to create a more informed citizenry. The project will develop and integrate machine ethics materials across curricula. These materials will address the social and ethical implications of technology. The research and education goals both address the broader need for ethical technology.

From a technical perspective, there are three key goals for this research. The first is to develop algorithms for ToMreasoning.  The second goal is to show that the algorithms are transparent and explainable. The third goal is to show that social robots using ToM reasoning will be more effective. The expected outcome is that people will have a more positive view of these robots. The robot needs to use ToM reasoning in two challenging situations. First, the robot may know something that the child does not know. Also, the child may know something that the robot does not know. To enable ToM reasoning, the project explores novel applications of analogical chaining. This approach uses analogies to make a series of inferences. The project lead will partner with development psychologists. Together, they will determine the effectiveness of the ToM reasoning. They will examine how parents and children perceive the robot. The development and evaluation of these algorithms will advance five critical research areas. First, the project will advance analogical reasoning. It will apply analogical reasoning to problems related to human-robot interaction (HRI). Second, the research will produce a novel algorithm for theory of mind reasoning for use in HRI. Third, the research will produce unique datasets featuring complex theory of mind scenarios. Fourth, researchers will provide new knowledge on how children and parents see robots. Last, the researchers will collect knowledge from experts. This knowledge will suggest how a robot should help and support children. One broader impact of this project is the effects it will have on undergraduate students. The project aims to support the growing diversity of computer science. Part of this plan is to recruit students from all backgrounds. Students will support all aspects of the research. The intent is to encourage more students to consider further research opportunities.

To facilitate the design of socially assistive robots (SARs), we present an architecture to generate assistive behavior for social robots given a high-level description of the intent of the assistance. Our approach features an ontology of assistive intents, a hierarchical task network planner, and robot middleware. We demonstrate the behaviors on two robot platforms and compare the behaviors. While many of the behaviors are similar, challenges remain in generating behaviors that will be presented consistently across multiple platforms.