The human brain creates abstract concepts: table, tree, or even a musical note. Up to now, there has been a broad consensus among scientists that the creation of these concepts requires a complex interaction and perfectly orchestrated between the neurons. But a new mathematical research, published in the journal Scientific Report, points to a different idea. Instead of using a lot neurons, our brain creates concepts with less.
“The individual neurons can create abstract concepts. In a sense, our neurons are more of these lists what we thought before,” he says Valeri Makarovresearcher at the Instituto de Matematica Interdisciplinary Universidad Complutense de Madrid (UCM).
To achieve these results, the mathematicians created a neural network, 3.200 cells in the selective coating and 1,600 in the level of concept, which simulates the human brain, and tested with… the music. “To this computer, we have given different pieces of music, among them, the Beethoven’s ninth Symphony and then, your computer, or the artificial brain, has learned to listen and distinguish the notes,” explains Makarov.
“Our brain creates this type of cells conceptual are able to perceive the different notes, and then, from the notes, we can compose more complex concepts, such as melodies,” he added. That is to say, when you play the note “fa” is the cell associated with conceptual is enabled, as that represents the concept of “tree”.
The neuron “Jennifer Aniston”
A laboratory of the University of Leicester have collaborated on the research found that one of these cells, the concept and called it the “Jennifer Aniston”, as the american actress. Scientists have opened up the skull of a patient, they put an electrode and recorded their neural activity.
We were started to teach the different images of famous personalities, well-known places. When they have shown different pictures of the actress discovered that a neuron increased its activity, while not what it was when he was about to images of other actresses, or Brad Pitt, that is to say, the neuron associated their image to a concept. And this is the reason why they called it that.
Makarov assures that after this job you want to apply this theory to the field of artificial intelligence to create a robot that can understand its environment and able to communicate.
In addition, according to the researcher, also, they are working with images of the lungs a hospital in England, the development of a network of artificial that is able to identify the different diseases, such as asthma. “We still do not have conclusive results, but it is the next step,” says Makarov.