Researchers at the Norwegian University of Science and Technology (NTNU) conducted a pilot trial with Pepper, a humanoid social robot developed by SoftBank Robotics, to test whether robot feedback could support learning and reduce teacher workload.
Fifteen participants aged 16 to 40 wore eye-tracking glasses while searching for specific objects in a series of images displayed for 10 seconds each.
During the task, participants received continuous feedback — positive, neutral, or negative — either from Pepper or from a human assistant.
Eye-tracking data showed that responses from the robot had a greater impact on visual attention than responses from the human assistant, and average performance was somewhat better with robot feedback.
The advantage disappeared when negative feedback was delivered at the end of the task, indicating that feedback order matters.
Professor Deepti Mishra and Associate Professor Akshara Pande, who led the study at NTNU's Educational Technology Laboratory in Gjøvik, said participants may feel more comfortable receiving feedback from a robot than from a person.
The researchers emphasize that social robots are intended as tools for specific contexts such as one-to-one learning, not as replacements for teachers, and that human oversight remains essential.
The pilot's small sample size limits generalization; the team recommends larger, longer-term studies with more diverse participants.
Social robots may support learning and reduce teachers' workload, pilot trial finds
This is an independent summary. The complete reporting, supporting context and any primary documents remain with Phys.org.