Robots that engage with people in a human-like manner are more likely to be perceived as having their own thoughts and beliefs, a study carried out at the Italian Institute of Technology has found.


The finding may help roboticists to create therapeutical AIs that vulnerable patients can more readily bond with, the researchers say.

The team carried out a series of experiments involving 119 participants interacting with a human-like test robot named iCub.

First, the researchers remotely controlled iCub and had it act welcoming and gregarious. While in others they made it act more mechanically and deactivated its eyes so it no longer made eye contact.

In all of the experiments, the participants were asked to watch documentary videos alongside the iCub. In the first batch of experiments, the bot was programmed to respond to the videos with sounds and facial expressions of sadness, awe or happiness, while in the second it simply responded with beeps and mechanical motions.

Both before and after the participants’ interactions with the bot, the team asked them to fill in a questionnaire that asked them to assess pictures of the robot as it carried out a number of different actions such as interacting with tools. They were then asked to judge the robot’s intentions and motivations.

The team found that the participants who watched videos with the more human-acting robot were more likely to rate the its actions as having thought, intent and emotion compared to those who watched with the machine-like robot.

This shows that humans can relate more readily to robots that not only look humanoid but act in a more human manner. The finding could inform the future development of social robots, the researchers say

“As artificial intelligence increasingly becomes a part of our lives, it is important to understand how interacting with a robot that displays human-like behaviours might induce higher likelihood of attribution of intentional agency to the robot,” said co-author Prof Agnieszka Wykowska, a principal investigator at the Italian Institute of Technology.

“Social bonding with robots might be beneficial in some contexts, like with socially assistive robots. For example, in elderly care, social bonding with robots might induce a higher degree of compliance with respect to following recommendations regarding taking medication.

“Determining contexts in which social bonding and attribution of intentionality is beneficial for the well-being of humans is the next step of research in this area.”

Read more about AI:



Jason Goodyer
Jason GoodyerCommissioning editor, BBC Science Focus

Jason is the commissioning editor for BBC Science Focus. He holds an MSc in physics and was named Section Editor of the Year by the British Society of Magazine Editors in 2019. He has been reporting on science and technology for more than a decade. During this time, he's walked the tunnels of the Large Hadron Collider, watched Stephen Hawking deliver his Reith Lecture on Black Holes and reported on everything from simulation universes to dancing cockatoos. He looks after the magazine’s and website’s news sections and makes regular appearances on the Instant Genius Podcast.