Mathematical Tools Help Map Change in Neuroscience

February 22, 2010

Mathematical Tools Help Map Change in Science Researchers Martin Rosvall and Carl T. Bergstrom (University of Washington) have tracked how the field of neuroscience evolved from an interdisciplinary specialty to a fully fledged scholarly endeavor.

The researchers explained their study in an article in PLoS One. They looked at the changing patterns of 35 million citations in more than 7000 scientific journals. This network represents the flow of information among researchers the world over, and the results showed that significant changes took place in the life sciences in the past decade.

"People have done a great deal of work on how to find the important features of a network at one specific point in time," Rosvall said. "But we have not had ways of looking at how these networks change over time."

According to Bergstrom, "detecting structural changes in large networks is a problem that consists of two parts. First, we identify statistically significant changes in the structure of a network, and second, we provide an intuitive way to visualize these changes."

The researchers showed that bootstrap resampling accompanied by significance clustering provides a solution. To connect changing structures with the changing function of networks, they highlighted the significant structural changes with alluvial diagrams, realizing de Solla Price's vision of mapping change in science.

The researchers believe that these mathematical methods can be applied to mapping changes in technology, biology, medicine, and finance.

Source: University of Washington Press Release (January 26, 2010)