Research News

Appropriately wired: Neurons determine the architecture of their networks

Neurons interconnect to form networks. The architecture of these neural networks provides the basic framework for all information processing in the brain. In the cerebral cortex of mammals, neural networks have developed a complexity that exceeds that in other brain regions or species. It is believed that this enormous complexity is one of the prerequisites for the emergence of higher brain functions in cortex, such as the transformation of sensory stimuli into perception or consciousness.

Scientists at the MPI in Bonn, in collaboration with researchers from Berlin, Tübingen, Amsterdam and Munich, have identified basic principles that can predict the complex wiring patterns in cortical networks. To this end, the researchers spent more than a decade reconstructing neurons in the cortex of rats (see figure), and then used these data to generate digital models of neural networks. This combination of experiment and computational modeling allowed the researchers to investigate systematically under which assumptions network models emerge whose wiring patterns correspond to those observed in the animal. Surprisingly, they found that whether and how strongly neurons are connected has neither to be genetically predetermined, nor does it necessarily emerge as a result of learning. In turn, knowledge of the morphologies of the neurons in a brain region was sufficient to predict realistic wiring patterns from subcellular to network scales.

"Our findings indicate that the architecture of a neural network is strongly impacted by the morphological properties of its neuronal constituents," says Dr. Oberlaender from the MPI in Bonn, and principal investigator of the study. The identified principles are applicable to reconstructions of any neural network, as demonstrated in the study for the human cortex. In the future, it will therefore be possible to disentangle in different brain regions and in different species, whether connections between neurons emerged due to morphological properties, whether they were genetically predetermined, and to what degree they might have been influenced by learning.

The study was published in 'Cell Reports' on 12/04/2022. Read the publication.

Microscopic image of two interconnected neurons in rat cortex.

For further information please contact:

Prof. Dr. Marcel Oberlaender
Max Planck Group Leader