Humans sense a collaborating robot as part of their “extended” body

Genoa (Italy), 11 September 2025 - Researchers from the Istituto Italiano di Tecnologia (IIT) in Genoa (Italy) and Brown University in Providence (USA) have discovered that people sense the hand of a humanoid robot as part of their body schema, particularly when it comes to carrying out a task together, like slicing a bar of soap. The study has been published in the journal iScience and can pave the way for a better design of robots that have to function in close contact with humans, such as those used in rehabilitation.

The project, led by Alessandra Sciutti, IIT Principal Investigator of the CONTACT unit at IIT, in collaboration with Brown University professor Joo-Hyun Song, explored whether unconscious mechanisms that shape interactions between humans also emerge in interactions between a person and a humanoid robot.

Researchers focused on a phenomenon known as the “near-hand effect”, in which the presence of a hand near an object alters visual attention of a person, because the brain is preparing to use the object. Moreover, the study considers the human brain's ability to create its “body schema” to move more efficiently in the surrounding space, by integrating objects into it as well.

Through an unconscious process shaped by external stimuli, the brain builds a “body schema” that helps us avoid obstacles or grab objects without looking at them. Any tools can become part of this internal map as long as they are useful for a task, like a tennis racket that feels like an arm extension to the player who uses it daily. Since body schema is constantly evolving, the research team led by Sciutti explored whether a robot could also become part of it.

Giulia Scorza Azzarà, PhD student at IIT and first author of the study, designed and analyzed the results of experiments where people carried out a joint task with iCub, the IIT’s child-sized humanoid robot. They sliced a bar of soap together by using a steel wire, alternately pulled by the person and the robotic partner.

After the activity, researchers verified the integration of the robotic hand into the body schema, quantifying the near hand effect with the Posner cueing task. This test challenges participants to press a key as quickly as possible to indicate on which side of the screen an image appears, while an object placed right next to the screen influences their attention. Data from 30 volunteers showed a specific pattern: participants reacted faster when images appeared next to the robot’s hand, showing that their brains had treated it much like a near hand. Thanks to control experiments, researchers proved that this effect appeared only in those who had sliced the soap with the robot.

The strength of the near hand effect also depended on how the humanoid robot moved. When the robot’s gestures were broad, fluid, and well synchronized with the human ones, the effect was stronger, resulting in a better integration of iCub’s hand into the participant’s body schema. Physical closeness between the robotic hand and the person also played a role: the nearer the robot’s hand was to the participant during the slicing task, the greater the effect.

To assess how participants perceived the robot after working together on the task, researchers gathered information through questionnaires. The results show that the more participants saw iCub as competent and pleasant, the more intense the cognitive effect was. Attributing human-like traits or emotions to iCub further boosted the hand’s integration in the body schema; in other words, partnership and empathy enhanced the cognitive bond with the robot.

The team carried out experiments with a humanoid robot under controlled conditions, paving the way for a deeper understanding of human-machine interactions. Psychological factors will be essential to designing robots able to adapt to human stimuli and able to provide a more intuitive and effective robotic experience. These are crucial features for application of robotics in motor rehabilitation, virtual reality, and assistive technologies.

The research is part of the ERC-funded wHiSPER project, coordinated by IIT’s CONTACT (COgNiTive Architecture for Collaborative Technologies) unit.