Jean-Marc Fellous
Institute for Neural Computation at the University of California, San Diego
Director of the Computational and Experimental Neuroscience Laboratory
Seminar Information
In the reductionist tradition, spatial navigation in animal neuroscience has mostly been studied in relatively small and simple environments, using simple tasks. It is clear however that complex spatial navigation in the ‘real world’ involves sophisticated reasoning at multiple spatial and temporal scales. I will present some recent behavioral and electrophysiological results using a task conducted in a large ‘megaspace’. In this unusual environment, we found that place cells become multifield and multiscale, challenging therefore most of the common computational models and understanding of hippocampal codes. Time permitting, I will also show some results and ongoing work involving different kinds of complexity including a spatial navigation optimization task, a task involving rat-robot interactions, and a task involving obstacle-rich environments. Understanding how the brain achieves efficient and complex spatial navigation is likely to provide new insights that may benefit other areas of inquiries including Artificial Intelligence, navigation algorithms in autonomous robots and the understanding of the specific neural computations involved during complex behaviors in humans and non-human primates.
Dr. Fellous did his Engineering undergraduate work at the Ecole Centrale Mediterranee (Marseille, France) in Computer Languages and Electronics with an internship at Imperial College (London, UK). He did his Masters in Artificial Intelligence and Pattern Recognition at Paris VI University (Paris, France) in Parallel Logic Programing and a Ph.D. In Computer Science and Artificial Intelligence at the University of Southern California (Los Angeles, CA), working on biologically inspired computational models of face recognition under the advising of Drs. M.A. Arbib and C. von der Malsburg. He then held a dual postdoctoral position in the laboratory of Dr. J. Lisman at Brandeis University (Boston, MA) working on carbachol-driven oscillations in the hippocampal slice, and with Dr. L. Zebrowitz on connectionists models of face perception. He then moved to the Salk Institute for Biological Studies (San Diego, CA) in the laboratory of Dr. T.J. Sejnowski to work on experimental and computational aspects of spike timing reliability and precision. Dr. Fellous became an Assistant Professor in Biomedical Engineering at Duke University (Durham, NC) in 2004, and joined the University of Arizona (Tucson, AZ) in 2006. where he was a full Professor in the Departments of Psychology and Biomedical Engineering, with affiliations in Neuroscience, Physiology, Applied Mathematics and Cognitive Science. He recently joined the Institute for Neural Computation at the University of California San Diego. He is the director of the Computational and Experimental Neuroscience Laboratory. The current interests of the laboratory include 1) Complex spatial navigation in large environments, 2) Decision making and memory consolidation during sleep, 3) The role of neuromodulatory substances in neural computations and 4) the neural bases of emotions.
A list of publications from the laboratory can be found at: https://cenl.ucsd.edu/