Colloquia Archive
Neurons That Keep the Rhythm
March 13, 2009
AbstractNeurons have a natural tendency to synchronize their activity, with important implications for our thoughts and actions. For instance, a neural code based on synchrony is responsible for conscious attention to stimuli and movement preparation, as well as the maintenance of task-relevant representations in active memory. A challenging aspect of neural synchronization is its lack of periodicity, which cannot be explained by simple oscillatory mechanisms. A naive suggestion might be that nonperiodicity is the product of neural noise. This idea, however, is hard to validate empirically.
As an alternative, I will describe a computational theory that is completely noise-free, yet captures a large body of experimental findings on neural synchrony. Despite the apparently unpredictable behavior of the model, it nonetheless responds in a precise and highly reproducible manner to a rhythmic train of pulses. This is consistent with experimental evidence: following a rhythmic visual stimulation, neurons in several locations – including retina and cortex – produce an "omitted stimulus response" that is closely timed to the frequency of stimulation. This response is interpreted as a detection of signal novelty, and underlies short-term perceptual memory for temporal sequences.
By its ability to combine unpredictable neural activity with highly stereotyped responses to stimuli, the new theory suggests that the brain operates in a state somewhere between completely disorganized and completely structured activity. This state is characterized by transient dynamics that promote rapid responses to stimuli and prevent the brain from getting "stuck" in permanent attractors. The new theory suggests that complex patterns of brain activity can be fully explained by the deterministic laws of chaos, and that neural noise is truly superfluous. This has implications for understanding the relationship between neural activity and cognition, as well as neuropathological diseases.
BiographyJ.-P. Thivierge is a postdoctoral fellow in the Department of Psychological and Brain Sciences, Indiana University, Bloomington. He received his Ph.D. in Psychology from McGill University in 2006, and his B.A. in Psychology with honors from the University of Ottawa in 2000. His research focuses on computational and quantitative investigations of neural connectivity and plasticity. The particular focus of my work is on investigating the mechanisms by which synaptic plasticity alters neuronal interactions as a result of learning, development, and reorganization following damage. My work explores two main themes: 1) the local changes in synaptic efficacy occurring at individual synaptic junctions; and 2) the global re-organization of complex cellular networks.
Recent Publications
Thivierge, J.P. (in press). How does non-random spontaneous activity contribute to brain development? Neural Networks.[pdf]. Covered in New Scientist.
Thivierge, J.P. (2008). Neural diversity creates a rich repertoire of brain activity. Communicative & Integrative Biology, 1. [pdf]
Shultz, T.R., Thivierge, J.P., & Laurin, K. (2008). Modeling the Characteristic-to-defining Features Shift in Concept Acquisition. Proceedings of the Annual Meeting of the Cognitive Science Society. 531-536. [pdf]
Tauskela, J.S., Fang, H., Hewitt, M., Brunette, E., Ahuja, T., Thivierge, J.P., Comas, T., & Mealing, G.A. (2008). Elevated synaptic activity preconditions neurons against an in vitro model of ischemia. Journal of Biological Chemistry, 283, 34667-34676. [pdf]
Thivierge, J.P., & Cisek, P. (2008). Non-periodic synchronization in heterogeneous networks of spiking neurons. Journal of Neuroscience, 28, 7968-7978. With cover illustration. [pdf]
Thivierge, J.P. (2008). Higher derivatives of ERP responses to cross-modality processing. Neuroinformatics, 6, 35-46. [pdf]
Thivierge, J.P. (2007). Functional Data Analysis of Cognitive Events in EEG. Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, 2473-2478. [pdf]
Thivierge, J.P., & Balaban, E. (2007). Getting into shape: Optimal ligand gradients for axonal guidance. BioSystems, 90, 61-77. [pdf]
Shultz, T.R., Rivest, R., Egri, L., Thivierge, J.P., & Dandurand, F. Could knowledge-based neural learning be useful in developmental robotics? The case of KBCC. (2007). International Journal of Humanoid Robotics, 4, 245-279. [pdf]
Thivierge, J.P., & Marcus, G.F. (2007). The Topographic Brain: From Neural Connectivity to Cognition. Trends in Neurosciences, 30, 251-259. [pdf] With cover illustration.
Thivierge, J.P., Rivest, F., & Monchi, O. (2007). Spiking neurons, dopamine, and plasticity: Timing is everything, but concentration also matters. Synapse, 61, 375-390. [pdf]
