Max Planck Institute for Dynamics and Self-Organization -- Department for Nonlinear Dynamics and Network Dynamics Group
Personal tools
Log in

Self-Organization instead of Environment and Genes

Date: 04.11.2010

Apart from the environment and genetic factors, self-organized processes play a crucial role in brain development. This is the result of a study conducted by an international team of scientists, among them researchers from the Max Planck Institute for Dynamics and Self-Organization, the Bernstein Center for Computational Neuroscience and the University of Göttingen (all in Germany). In the brains of ferrets, tree shrews, and galagos the researchers discovered a surprising similarity: The nerve cells within the visual cortex are arranged in the same way. Neither early influences of the environment, nor inheritance can explain this result. However, the scientists were able to predict this brain architecture using a mathematical model that describes how neuronal circuits develop in a self-organized way.
(Science Magazine, online-version, November 4th, 2010)

Nerve cells in the visual cortex react to defined picture elements such as edges and contours. Each cell has a so-called orientation preference: It is specialized for processing edges oriented in a certain direction, for example horizontal, vertical or inclined edges. If you imagine all cells of the same orientation preference colored with the same color, you obtain the map of orientation preference. The fundamental structural element of this map, that is repeated many thousand times throughout the visual cortex, is a pinwheel: areas of the same orientation preference meet in one point – like the wings of a toy pinwheel (see figure 2).

While earlier studies had led researchers to expect that the distribution of the pinwheels in the visual cortices of different species would differ greatly, the researchers led by Prof. Dr. Fred Wolf from the Max Planck Institute for Dynamics and Self-Organization in Göttingen found an astonishing similarity for ferrets, tree shrews and galagos. One trait of this common design is the density of the pinwheels. For the named species, this density and a large number of further properties are exactly the same. This finding cannot be explained by an inherited genetic blueprint, because the last mutual ancestor of ferret, tree shrew and galago lived more than 65 million years ago in the age of the dinosaurs. Since then, there would have been plenty of time for the brains to develop in a different manner. Apart from that, there are mammals that are much more closely related than the studied species and whose visual cortices are not similar. But also the influences of experience on early brain development cannot account for the new results. The studied species deal with completely different environmental conditions in their natural habitats.

Empirically and theoretically the researchers showed, that the same density of pinwheels can best be explained by self-organized processes in brain development. As soon as the animals begin to see after their birth, the orientation preference maps form as on their own. The mathematical analysis of neuronal self-organization showed, that very few requirements suffice to create the nerve cell architecture found in the study. One of these requirements is that neurons are able to send direct signals to each other spanning large distances. If this requirement and a few others are fulfilled, the neurons in the model work together to form a quasi-periodic pattern of their preferred orientation. Quasi-periodic means, that no pattern is ever repeated exactly.

“Well-known examples for self-organization are Mexican waves propagating through a stadium during sports events or stop-and-go-waves in traffic that can occur even without an external impairment of the traffic flow”, says Matthias Kaschube, Lewis-Sigler Fellow at Princeton-University and the paper’s first author. In these examples - as with all other processes of self-organization - there is not a hidden “script” forcing the elements (fans or vehicles in the examples above) to act as they do. The elements’ motion results only from the way they influence each other.

In the past decades, scientists have worked out how mathematical models can help to understand such self-organized processes for many non-biological systems. As the theoretical physicist Fred Wolf stresses, the new results provide custom-made mathematical concepts for the understanding of how neuronal elements interact within the visual cortex.

Original publication: Matthias Kaschube, Michael Schnabel, Siegrid Löwel, David M. Coppola, Leonard E. White, Fred Wolf: Universality in the Evolution of Orientation Columns in the Visual Cortex, Science 330, 1113 (2010);, Nov 4th, 2010, DOI: 10.1126/science.1194869

For additional information, please contact:
Dr. Birgit Krummheuer
Press office
Max Planck Institute for Dynamics and Self-Organization
Tel.: +49 551 5176 668
Mobile: +49 173 3958625

Prof. Dr. Fred Wolf
Max Planck Institute for Dynamics and Self-Organization, Bernstein Center for Computational Neuroscience, and University Göttingen
E-mail: follow this link