Who am I? A seemingly simple question that has fascinated and tormented philosophers and scientists for millennia. Today, unexpectedly, new answers could be provided by robotics. Not because robots have suddenly developed self-awareness (the work of embodiment is just in its infancy and who knows if it will one day lead to results), but because they represent unique models for studying the processes that, in humans, give rise to sense of self. As an innovative study just published in explains Science Robotics (I link it here), current robots can simulate aspects such as a sense of body ownership or control over one's actions, offering a valuable window into the workings of our minds. An approach that could also shed light on the mysteries of disorders such as autism e schizophrenia, in which the sense of self appears altered.
Deconstructing the Sense of Self
The starting point is an intuition as simple as it is profound: our sense of self is not a monolith, but a mosaic of interconnected processes. “The experience of being an 'I', contained in our body and capable of acting in the world, seems natural and immediate to us,” he explains Agnieszka Wykowska, first author of the study and head of the Social Cognition in Human-Robot Interaction unit at theItalian Institute of Technology. “But in reality it is the result of multiple components, such as the sense of owning a body or controlling our actions.”
This is where robots come in. “Today we are able to build robots that can distinguish their own body from that of others and detect the consequences of their actions,” Wykowska continues. “In this sense, they can serve as embodied models of the cognitive processes that, in humans, underlie the sense of self".
But robots are not just models. They can also be used as “experimental probes” to explore the sense of self, thanks to the fact that they have a body and can interact with both humans and the environment. The authors of the study (in addition to Wykowska, Tony Prescott of the University of Sheffield and Kai Vogeley (University of Cologne) propose two complementary approaches.
Simulate to understand
The first involves programming robots to simulate the mental processes involved in experiencing the self, based on knowledge from psychology and neuroscience. “Research suggests that in humans, a sense of self develops as the brain’s ‘best explanation’ of sensory experience and its role in generating it,” Prescott explains. “A robot, as a physically embodied actor, is an ideal platform to test these theories.”
Interact to reveal
The second approach involves using robots in psychological experiments in which people interact with them while the robots exhibit social skills, such as verbal communication or joint attention. “These experiments could reveal whether people perceive robots as ‘other social animals’ and whether the mental states they attribute to them are similar to those they attribute to humans,” Vogeley adds.
Some experiments conducted by Wykowska's group at theIIT have already shown that, under certain conditions, humans develop a sense of “shared agency” with robots, when they perceive them as intentional agents and collaborate with them as a team.
Sense and development of the self, from man to robot and back
The authors also draw a parallel between the development of a sense of self in humans over the course of a lifetime and the possibility of transferring aspects of it to robots. “By age 4, children develop a sense of themselves as a continuous entity in time and recognize that others also have selves,” Wykowska notes. “We are starting to explore these aspects in robots, creating memory systems similar to autobiographical memory human”.
However, the work is still in its infancy. Current robots do not have awareness of themselves as entities that persist over time, nor do they recognize humans or other robots as having a self. But the research opens up fascinating prospects.
From Disturbances of the Self to a New Understanding
One of the most promising challenges is to use robots to understand what happens when the sense of self is compromised, as in autism e schizophrenia. “Modeling these conditions in robots could offer new insights into the mechanisms underlying the sense of self and the ways in which it can be altered,” suggests Vogeley, who is also a psychiatrist.
It's not just a scientific challenge, but of an opportunity to embrace the diversity of human experience. Understanding the many facets of the sense of self, in its typical and atypical variants, could help us build a more inclusive and empathetic society.
Towards a future of understanding
The study therefore opens up perspectives that go far beyond the robotics. “Understanding our sense of self through robots ultimately means understanding ourselves more deeply and more completely,” Wykowska concludes.
It is a fascinating journey, which unites science and philosophy, technology and humanity.
A journey that could lead us to solve one of the greatest mysteries of our existence: what it means to be an “I”. And who knows, maybe along the way we will build robots with some form of self-awareness. It would be a turning point not only technological, but existential, that would force us to redefine the boundaries between natural and artificial, between human and non-human.
But even if this scenario remains confined to science fiction for now, one thing is certain: by studying the sense of self through robots, we are actually exploring the profound nature of our humanity. And this, in itself, is already an extraordinary result.
