The USC Professor Who Pioneered Socially Assistive Robotics
When Maja Matarić first envisioned a career in robotics engineering, the field she wanted to enter did not yet exist. Rather than wait for it to emerge, she helped create it. In 2005, Matarić and her doctoral student David Feil-Seifer presented a defining paper at the International Conference on Rehabilitation Robotics that introduced socially assistive robotics—a new area focused on using social interaction, rather than physical manipulation, to support human health and learning.
Today, Matarić is a professor of computer science, neuroscience, and pediatrics at the University of Southern California’s Viterbi School of Engineering in Los Angeles. An IEEE Fellow, she leads the Interaction Lab at USC, where her team develops robots designed to provide personalized therapy and care through conversation, gameplay, and emotional responsiveness. Her work spans applications for children with autism spectrum disorder, elderly individuals, stroke patients, and, more recently, university students experiencing anxiety and depression.
Matarić’s contributions have been recognized with numerous honors, including the 2025 Robotics Medal from MassRobotics, a Boston-based nonprofit that supports female researchers advancing robotics. At the award ceremony, she described feeling overwhelmed by the sense of community and appreciation in the room, noting that many attendees were colleagues she had known for years.
From Belgrade to Boston: A Journey Shaped by Curiosity and Care
Born in Belgrade, Serbia, Matarić grew up in a household where engineering and storytelling coexisted—her father was an engineer, her mother a writer. After her father’s passing when she was 16, she and her mother moved to the United States. She credits her father for sparking her interest in engineering and her uncle, an aerospace engineer, for introducing her to computer science.
Matarić earned her bachelor’s degree in computer science from the University of Kansas in 1987. She then joined MIT’s Artificial Intelligence Laboratory, where she worked under Rodney Brooks, an IEEE Life Fellow known for pioneering behavior-based robotics. Inspired by how animals utilize environmental landmarks to navigate, she developed Toto, one of the first navigating behavior-based robots, which used sonar and distributed mapping to move through MIT’s AI Lab building.
She completed her master’s in AI and robotics at MIT in 1990 and continued under Brooks for her doctoral studies, focusing on distributed algorithms that enabled teams of up to 20 robots to coordinate complex tasks like object search and environmental exploration. Matarić received her Ph.D. From MIT in 1994.
After a brief tenure as an assistant professor at Brandeis University, where she founded the Interaction Lab, she joined USC in 1997. Three years later, in 2002, she helped establish the Center for Robotics and Embedded Systems, now known as the Robotics and Autonomous Systems Center (RASC), which emphasizes human-centered, scalable robotic systems and interdisciplinary collaboration across USC.
Defining a New Field: Socially Assistive Robotics
Matarić’s shift toward socially assistive robotics was deeply personal. After giving birth to her first child in 1998, she began reflecting on how to answer her daughter’s inevitable question: “Why do you work with robots?” She wanted to be able to say more than “I publish papers” or “it’s well-recognized.” Her goal became clear: “Mommy’s robots aid people.”
In 2005, Matarić and Feil-Seifer formally defined socially assistive robotics as a discipline where robots support users through social interaction—such as encouragement, companionship, and motivation—rather than physical labor. Their paper at the International Conference on Rehabilitation Robotics was the first to articulate this distinction, laying the foundation for a new research direction.
That same year, she founded the Interaction Lab at USC with a clear mission: to create robots that provide social, not physical, support. Over time, the lab developed a series of platforms tailored to specific therapeutic and educational needs.
Robots That Teach, Motivate, and Connect
One of Matarić’s longest-running projects involves using robots to improve communication skills in children with autism spectrum disorder (ASD). Her team developed Bandit, a 56-centimeter-tall robot with a humanlike head, torso, and arms, equipped with FireWire cameras for eyes and movable mouth and eyebrows to express a range of facial expressions. Bandit can pan and tilt its head, play games, and offer verbal affirmation.
Studies showed that when interacting with Bandit, children with ASD exhibited behaviors rarely seen in other contexts—such as initiating play and imitating the robot’s actions. The lab also adapted Bandit to assist elderly users and stroke patients, programming it to guide and motivate them through seated aerobics and other daily movement exercises.
Over the years, the Interaction Lab introduced additional robots: Kiwi, an owl-shaped design that helped children with ASD learn social and cognitive skills while encouraging physical activity in isolated seniors; and Blossom, originally developed at Cornell University and later adapted by USC to be more affordable and customizable. Blossom has been used in studies exploring how robots can support students with anxiety or depression through cognitive behavioral therapy (CBT) practice.
Robots vs. Chatbots: Testing Mental Health Support in Real-World Settings
Matarić’s interest in mental health applications grew as large language model (LLM) chatbots began being promoted as accessible tools for anxiety and depression support. While acknowledging the appeal of chatbots—especially given barriers to therapy access and insurance coverage—she questioned whether a physical robot might offer advantages over a screen-based interface.
To test this, her team conducted a two-week study in USC dorms, assigning students randomly to complete daily CBT exercises either with a chatbot or with Blossom, both powered by the same LLM. Participants completed clinical assessments before and after each session to measure psychiatric distress.
The results showed a significant reduction in distress among those who interacted with the robot, while no such improvement was observed in the chatbot group. Transcript analysis revealed that the robot was more effective at engaging users, even though both systems used identical language models. Matarić attributed this to the robot’s physical presence, facial expressions, and ability to create a sense of shared space and responsiveness.
Based on these findings, in 2024 Matarić received a grant from the U.S. National Institute of Mental Health (NIMH) to conduct a six-week clinical trial involving 120 USC students. The study, currently underway, examines Blossom’s effectiveness in delivering CBT practice, with participants wearing Fitbits to track physiological responses such as heart rate and sleep patterns. Researchers are also assessing subjective factors like feelings of connection to the robot, intrinsic motivation, engagement, and adherence to the therapy protocol.
Matarić emphasizes that working with human participants requires engineers to develop skills beyond coding and hardware—such as understanding psychology, ethics, and user-centered design. She praises her graduate students for embracing this interdisciplinary challenge, noting that their empathy and care are as vital as their technical abilities.
Building Community Through IEEE
Matarić joined the Institute of Electrical and Electronics Engineers (IEEE) as a graduate student in 1992, the same year she published her first paper in IEEE Transactions on Robotics and Automation. The paper, titled “Integration of Representation Into Goal-Driven Behavior-Based Robots,” described her work on Toto and helped establish her early reputation in the field.
As an active member of the IEEE Robotics and Automation Society, she regularly attends major conferences, including the IEEE International Conference on Robotics and Automation (ICRA), the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), and the ACM/IEEE International Conference on Human-Robot Interaction (HRI), which aligns closely with her research focus.
She credits IEEE Life Fellow George Bekey, founding editor-in-chief of IEEE Transactions on Robotics, with helping recruit her to USC after recognizing her work through her graduate advisor, Rodney Brooks. Bekey’s editorial guidance was instrumental in shaping early robotics literature, and Matarić notes that her own contributions built on the foundation he helped establish.
Reflecting on her long-standing involvement with IEEE, Matarić encourages students to join professional societies not only for technical growth but for personal connection. “Joining a society has an impact, and it can be personal,” she says. “That’s why I recommend my students join the organization—because it’s important to get out there and get connected.”
As the NIMH-funded trial progresses, the Interaction Lab continues to refine Blossom’s ability to adapt to individual users—adjusting its movements, exercise recommendations, and feedback based on real-time data and user preferences. The goal is not to replace human therapists, but to create accessible, engaging tools that can extend support to those who might otherwise travel without.
For Matarić, the ultimate measure of success remains simple: whether her robots help people live better lives. That question, first inspired by her daughter’s curiosity, continues to drive a research agenda that blends engineering rigor with deep human purpose.
Related reading