The gaze is a very powerful and important signal during human-human communication and interaction, it conveys intentions and informs the decisions of others. What happens when a human being and a human looking at each other interact?

Researchers at IIT-Istituto Italiano di Tecnologia (Italian Institute of Technology) have investigated whether a robot’s gaze affects the way people think in the context of social decision-making. What they found is that the gaze exchanged with the robot affects the neural activity of the human being, influencing, and in particular delaying, decision-making processes.

Thus, the robot’s gaze causes humans to perceive it as a social cue. These findings have powerful implications for contexts where humans may find applications such as co-workers, clinical support, or home assistants.

The study published in robotics science Today, it is designed under a larger comprehensive project led by Agnieszka Wykowska, IIT coordinator of the “Social Perception in Human-Robot Interaction” lab, and funded by the European Research Council (ERC). The project, called “InStance,” addresses the question of when and under what conditions people treat robots as intended beings. That is, in order to explain and explain the behavior of a robot, people refer to mental states such as beliefs or desires.

The authors of the paper are Marwen Belkaid, Kyveli Kompatsiari, Davide de Tommaso, Ingrid Zablith and Agnieszka Wykowska.

In most everyday situations, the human mind needs not only to make decisions but also to anticipate and predict the behavior of others. In such contexts, outlook can be informative about other people’s intentions, goals, and upcoming decisions. Humans pay attention to the eyes of others, and the brain reacts intensely when someone looks at them or directs their gaze to a specific event or place in the environment. Researchers have investigated this type of interaction with the robot.

Robots will be more and more present in our daily lives. This is why it is important to understand not only the technological aspects of robot design, but also the human aspect of human-robot interaction. Specifically, it is important to understand how the human brain processes behavioral signals transmitted by robots.”

Agnieszka Wykowska, first author of the study and principal investigator, Italian Institute of Technology

Wikoska and her research group asked a group of 40 participants to play a strategy game – the game of chicken – with the iCub robot while they measured participants’ behavior and neural activity, the latter via electroencephalography (EEG). The game is a strategy game, depicting a situation in which two drivers of simulated cars move towards each other on a collision course and the outcome depends on whether the players satisfy or keep going straight.

The researchers found that participants were slower to respond when the iCub generated a gaze reciprocal during decision-making, compared to a gaze avoided. Delayed responses may indicate that reciprocal gaze entails higher cognitive effort, for example by eliciting more thought about iCub choices or a higher degree of suppression of a potentially distracting gaze stimulus, which was irrelevant to the task.

“Think of playing poker with a robot. If a robot looks at you during the moment when you need to decide on the next move, you will have a much more difficult time deciding, for a situation where the robot is staring away. Your brain will also need to use laborious and expensive processes to try Ignore ‘that robot look’ Wikoska explains as well.

These findings suggest that the robot’s gaze “hijacks” the “social-cognitive” mechanisms of the human brain – causing the brain to respond to the robot as if it were a social agent. In this sense, for a robot “being social” may not always be beneficial to humans, interfering with their performance and speed of decision-making, even if their mutual interaction is fun and attractive.

Wikoska and her research group hope that these findings will help robotics scientists design bots that display the behavior most appropriate for a particular context of application. Human beings with social behaviors may be useful in helping to care for the elderly or care for children, as in the case of the iCub robot, being part of an experimental treatment in the treatment of autism. On the other hand, when task focus is needed, such as in factory settings or in air traffic control, having a robot with social cues can be a distraction.