At birth, the child does not have this “consciousness”, he does not have the basis of separation between himself and others, and it is over time, through his interactions with others , that this separation develops. (Photo: The Canadian Press)
A new model developed by researchers in Montreal and France might lead to a better understanding of how the brain acquires its “consciousness”, which might then lead to better treatment of neurodevelopmental disorders and advances in the field of artificial intelligence.
This neurocomputational model was developed by an international group of researchers from the Institut Pasteur and the Sorbonne University in Paris, the CHU Sainte-Justine, Mila — the Quebec Institute of Artificial Intelligence, and the University of Montreal. .
“Until now, this model has always been made for an already formed brain, an adult brain,” said one of the authors of this work, Professor Guillaume Dumas from the Department of Psychiatry and Addiction of the Faculty of Medicine of Paris. ‘Montreal university.
“However, in our model, what we did was to start a bit like a child’s brain and to develop this model, that is to say (what are) the biological mechanisms that are necessary so that, through development and experience of the world, a brain can acquire this faculty of consciousness.
The word “consciousness”, points out Professor Dumas, is one of the most debated terms and possibly one of the least well defined in the field of cognitive science. In the context of this work, it refers to the brain’s ability to describe a sensory experience (eg, “I see a red flower” or “I hear music”).
At birth, he continues, the child does not have this “consciousness”, there is basically no separation between him and the others, and it is over time, through his interactions with others, that this separation develops.
“The model we are proposing is the steps necessary to (get to) this self-awareness,” explained Mr. Dumas. What are the most fundamental and critical mechanisms for consciousness to emerge?
The model is computational, in the sense that it is on a computer, but it is a “biophysical” model which, unlike algorithms which are programmed to arrive at a specific goal, is rather “inspired by biology”
“We simulated neurons, like those in our brain, (and they) reproduce the electrical exchanges that take place in the brain, explained Mr. Dumas. We also integrated into our model biological phenomena such as dopamine, which is a neurotransmitter associated with reward, so that the model can learn via rewards during its learning.
Understanding the emergence of consciousness might have clinical implications in the treatment of neurodevelopmental disorders, but also in the development of artificial intelligence, he concludes.
The results of this work were revealed in the scientific journal Proceedings of the National Academy of Sciences of the United States of America.
