What can we learn about brain function and failure by trying to understand brains as biological computers? This overarching question is what drives the work of Nathan Urban, Ph.D., associate director of the Brain Institute, Professor in the Department of Neurobiology, co-director of the Center for the Neural Basis of Cognition, and Vice Provost for Special Projects at the University of Pittsburgh. Urban’s laboratory uses physiological, imaging, behavioral, and computational approaches to address questions about the functional circuitry of the mouse sensory systems, especially the olfactory bulb and the sense of smell. Understanding computational properties of brain networks often requires new approaches for measuring and manipulating brain activity and behavior. Thus, Urban also is interested in the application and development of techniques that facilitate this sort of parallel data acquisition in vitro and in vivo.
One of Urban’s long-term research interests centers around understanding the physiological mechanisms underlying the functional and computational properties of brain neuronal networks. That is, he wants to understand the brain as an organ of biological computation and by extension to understand brain dysfunction as examples of failed computation. This interest has led Urban to work closely with mathematicians and statisticians to develop new ways of analyzing neurons and their functions, focusing on how brains and computers differ in the functioning of their elementary parts. For example, Urban and colleagues noted that unlike electronic computers, brains are made up of noisy, unreliable, highly individual neurons – a stark contrast from electronic computers which are made up of billions of identical transistors and other circuit elements. Urban and colleagues has shown how the brain improves its function by a kind of “crowd sourcing” using the diversity of neuronal properties to improve its computational power. This cell-to-cell variability in intrinsic neuronal properties can therefore be considered to be a beneficial "feature" of brain computation rather than a "bug" of biological imprecision. Multiple neurons performing similar functions can be thought of as making up “dream teams” of neurons that allow the brain to better extract information from incoming stimuli. (2013 PNAS). In 2014, Urban, led a team that received National Science Foundation funding to extend these new approaches to understanding this collective computation of neuronal groups.
One key aspect of this approach in the last few years has been gaining an understanding of the role of different sources of variability in neuronal computation. Beginning early in his career, Urban worked with Pitt mathematician Bard Ermentrout to understand how groups of neurons come to synchronize their firing (2005, Physical Review Letters). Ultimately, the work has helped scientists understand mechanisms of neuronal synchrony that play a role in the oscillatory patterns of brain activity that are disrupted in diseases like schizophrenia. This mechanism, which he named stochastic synchrony, seems to provide a basis by which additional noisy input from synapses or even from sensory stimuli can generate patterns of brain activity that may enhance sensitivity and selectivity in sensory systems. Urban’s work in this area and its application to deciphering the role of neuronal synchronization in schizophrenia also has received funding from the National Institute of Mental Health and the National Science Foundation.
Finally, Urban has applied these approaches to study mouse models of autism. In this case, his laboratory is looking at trial to trial variability as a kind of “noise.” Human studies demonstrate that reliability of sensory evoked responses is impaired in autistic subjects. Urban’s goal is to identify sources of variability of neuronal responses in our mouse models and determine whether this kind of noise differs between control mice and those with autism-related mutations. His work in this area has been funded by the Simons Foundation Autism Research Initiative.
Prior to joining the University of Pittsburgh in 2015, Urban was interim provost at Carnegie Mellon University and director of BrainHub, CMU’s campus-wide interdisciplinary brain research initiative. In 2010, Urban was named the Dr. Frederick A. Schwertz Distinguished Chair in the Life Sciences. Earlier in his career, he held an Eberly Family Faculty Development Chair. In the 2000’s he was named a top 50 leader in science and technology by Scientific American, secured a prestigious award from the International Human Frontier Science Program to understand neural integration, and received a Young Investigator Award from the Association for Chemoreception. Urban’s training was supported by a Rhodes Scholarship, as well as a Chancellor’s Scholarship from the University of Pittsburgh, a pre-doctoral fellowship from the Howard Hughes Medical Institute and postdoctoral funding from the Alexander von Humboldt Foundation. He was a post-doctoral fellow at the Max-Planck Institute for Medical Research in Heidelberg, Germany.
Atlas of Behavior
How do behaviors result from neural computations, and how can we use large-scale behavioral datasets to advance science? To tackle these questions, Urban and a group of investigators are using high definition behavioral analysis to create what he hopes will emerge as a series of atlases of behavior. Urban’s team is developing technologies to capture animal behavior in an effort to correlate them with biological datasets. The long-term goal is to collect and analyze large sets of behavioral data to build probabilistic models, or atlases, to understand and predict the behavior of individuals and groups. These atlases should have profound impacts, according to Urban. For example, he anticipates that they will provide important insight into normal and abnormal brain development and the role of genes in disease. This information should facilitate both the measurement and improvement of human performance, and our understanding of how behaviors change in response to disease or treatment for a given disorder. (see Atlas of Behavior for more details).
