University of Chicago researchers are working to improve how robots interact with humans by programming them to respond empathetically and use nonverbal social cues. Sarah Sebo, assistant professor of computer science, leads the lab focusing on making robots more intuitive and engaging in their interactions.
Sebo explained that while robotics has traditionally centered on automating tasks such as delivering food or assisting with daily living, there is now a greater need to design robots that can better understand and engage with people. “People started realizing, oh, we need to really understand people better and design social behaviors in robots so that they can more effectively engage with people,” Sebo said.
Human-robot interaction blends computer science and psychology, requiring an understanding of human behavior to help robots express themselves in ways similar to humans. Sebo illustrated this by noting the difference between a robot using a laser pointer versus one using eye movements: “If the robot has a head with eyes, and the robot looks over here, you very quickly know where the robot wants you to look.”
Graduate student Lauren Wright added that the field is still new: “The fun thing about human-robot interaction is that because it’s a new field, there are so many big unanswered questions. We are measuring things between humans and robots that no one has measured before.”
Sebo also discussed how people often form attachments to robots differently than other technology. For example, when Roomba vacuum owners had their devices replaced during repairs, many wanted their original unit returned rather than accepting a new one.
In their research, Sebo’s team equips robots with verbal abilities like talking and nodding before recruiting participants for studies in which they interact with these machines. The team analyzes responses under different conditions to assess trust and rapport.
Describing how robots process communication, Sebo said: “When a robot needs to communicate with a person verbally, it first needs to understand what they said. It takes the speech and turns it into text. Then it prompts an LLM to generate a response. The robot takes that text and turns it into speech.”
One study involved participants sharing personal problems with a robot programmed either to move on or offer an empathetic response. According to Sebo: “When the robot provided this one sentence of active listening... People viewed themselves as having significantly more rapport with this robot.”
The research also explores applications in education where robots could supplement classroom teaching by providing interactive feedback without replacing teachers. Wright emphasized: “We don’t want to replace teachers, just augment what they are doing.” Their findings suggest children feel less anxious reading aloud to robots compared to humans.
However, potential risks exist as well. As Sebo noted: “Once you put a robot face and body with an LLM, it becomes even more real than this text conversation you might have with an LLM.” The team is investigating how much social connection is beneficial for children interacting with robots.
Wright referenced studies indicating students sometimes learn better from factual rather than highly personalized robotic characters: “It’s a nice philosophical push for other researchers to consider whether we need robots to be ‘characters’ or just be adaptive and responsive without a strong personality.”
Graduate student Alex Wuqi Zhang stressed safety concerns when AI software gains physical form through robotics: “We also need to make sure guardrails are in place...we need to make sure it’s safe.”
Looking ahead five to ten years, Zhang predicted wider integration of robots into daily life: “It's really important to study how specific robot behaviors affect how humans behave and how humans shape how robots behave.” Sebo concluded that successful social engagement will be essential for widespread acceptance of robotics.