The neocortex, a structure essential for human intelligence, is less than five millimeters thick. There, in the outermost layer of the brain, 20 billion neurons process countless sensory perceptions, plan actions, and form the basis of our consciousness. How do these neurons process all this complex information? That largely depends on how they are connected to each other.
What was known until now about the neuronal architecture of the human neocortex was based mainly on findings made from studying animal models such as mice.
The human neocortex is much thicker and more complex than that of the mouse. However, the scientific community had until now assumed (partly due to lack of data) that it followed the same basic principles of connectivity.
Now, in a new study, for which cutting-edge technology and exceptionally difficult to obtain tissue samples have been used, it has been proven that, contrary to what was assumed until now, the nerve cells of the human neocortex are very connected. different from those of mice.
The study was carried out by a team including, among others, Yangfan Peng and Jörg Geiger, from the Neuroscience Research Center, attached to the Charité University Hospital in Berlin, Germany.
The researchers specifically verified that, in the case of mice, neighboring neurons frequently communicate with each other as if they were in dialogue. One neuron sends a signal to another and the other neuron sends the signal back. This means that information often flows in recurring loops.
The study’s authors found that only a small fraction of the neurons observed in human brain tissue interacted with each other. In humans, information tends to flow in only one direction. It rarely returns to the starting point, either directly or through cycles.
This increases the efficiency and ability of the human brain to process information.
Reconstruction of a local set of neurons in the neocortex of a human being. (Image: © Charité / Sabine Grosser)
This finding is not only useful for neurology, but will also be useful for making progress in the field of artificial neural networks (computerized structures that mimic human brain architecture).
In an experiment with artificial neural networks, Peng and his colleagues put them to work on a task that often requires artificial intelligence: recognizing correct numbers from audio recordings of spoken digits. The artificial neural network that imitated human structures achieved more correct responses in this speech recognition task than the one modeled in mice. It was also more efficient, since the same performance required the equivalent of 380 neurons in the mouse model, but only 150 in the human.
The study is titled “Directed and acyclic synaptic connectivity in the human layer 2-3 cortical microcircuit.” And it has been published in the academic journal Science. (Fountain: NCYT de Amazings)