Comparing visual and motor cortex: representational coding versus dynamical systems
2012; Frontiers Media; Volume: 6; Linguagem: Inglês
10.3389/conf.fncom.2012.55.00026
ISSN1662-5188
Autores Tópico(s)EEG and Brain-Computer Interfaces
ResumoEvent Abstract Back to Event Comparing visual and motor cortex: representational coding versus dynamical systems Jeffrey S. Seely1*, Matthew T. Kaufman2, Adam Kohn3, Matthew A. Smith4, J. A. Movshon5, Nicholas J. Priebe6, Steven G. Lisberger7, Stephen I. Ryu2, Krishna V. Shenoy2, John P. Cunningham8 and Mark M. Churchland9 1 Columbia University, Department of Neuroscience, United States 2 Stanford University, Department of Electrical Engineering, United States 3 Albert Einstein College of Medicine, Department of Neuroscience, United States 4 University of Pittsburgh, Department of Ophthalmology, United States 5 New York University, Center for Neural Science, United States 6 University of Texas at Austin, Section of Neurobiology, School of Biological Sciences, United States 7 University of California San Francisco, Howard Hughes Medical Institute, W.M. Keck Foundation Center for Integrative Neuroscience, Department of Physiology, United States 8 Washington University, Department of Engineering, United States 9 Columbia University, Department of Neuroscience, David Mahoney Center, Kavli Institute for Brain Science, United States Systems neuroscience often employs models that explain neural responses in terms of represented stimulus features or movement parameters. These models can be powerful, but do not apply equally well to all circuits. For example, the activity of a central pattern generator is best captured by its intrinsic dynamics. Here, we examine the population response in a number of cortical areas, and ask whether responses appear stimulus driven (i.e. are better described as a function of external parameters) or internally generated (i.e. are better described by a dynamical system). We analyzed datasets (44 - 218 single and/or multi-unit isolations) from visual areas V1 and MT (recorded during the presentation of visual stimuli) and from primary motor and premotor cortex (recorded during a delayed reach task). Our analyses did not fit particular representational or dynamical models, but instead asked whether basic features of the data tended to obey or violate the expectations of a dynamical system. The principal expectation is, for a k-dimensional dynamical system, the number of temporal patterns (modes) present in the data should tend to be <= k. This can be shown analytically for linear, time-varying dynamical systems. Conversely, stimulus-driven responses can display an arbitrary variety of temporal patterns, as determined by the stimuli. We therefore compared the number of temporal patterns in the data with the overall dimensionality of the system. We considered data of the form r(n,c,t), the rate of neuron n for condition c and time t. To determine the overall dimensionality, we applied PCA considering neurons as ‘variables’ and conditions and times as ‘observations’. To determine the number of temporal patterns that the system displayed, we applied PCA again considering conditions as ‘variables’ and neurons and times as ‘observations’. For each of these two applications of PCA we found the number of dimensions required to reconstruct the data to 90% precision. We then took the ratio (overall dimensionality divided by number of temporal patterns). For simulated data from stimulus-driven models, ratios were < 1. For simulated data from dynamical systems models, ratios were > 1. For the neural data from visual areas (2 datasets) the ratio lay close to the stimulus-driven prediction (ratios = 0.8, 1.1), while the population response from the motor areas (4 datasets) appeared more dynamical (ratios = 1.6, 1.8, 1.8, 2.4). These results indicate that, in visual areas the overall structure of the data is consistent with responses being largely a function of stimulus parameters. In contrast, in motor areas the structure of the data argues that responses are more strongly dominated by internal dynamics. Keywords: dynamical systems, Motor Cortex, Visual Cortex, modeling, data analysis Conference: Bernstein Conference 2012, Munich, Germany, 12 Sep - 14 Sep, 2012. Presentation Type: Poster Topic: Abstracts Citation: Seely JS, Kaufman MT, Kohn A, Smith MA, Movshon JA, Priebe NJ, Lisberger SG, Ryu SI, Shenoy KV, Cunningham JP and Churchland MM (2012). Comparing visual and motor cortex: representational coding versus dynamical systems. Front. Comput. Neurosci. Conference Abstract: Bernstein Conference 2012. doi: 10.3389/conf.fncom.2012.55.00026 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 02 Jun 2012; Published Online: 12 Sep 2012. * Correspondence: Mr. Jeffrey S Seely, Columbia University, Department of Neuroscience, New York, NY, United States, jsseely@gmail.com Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Jeffrey S Seely Matthew T Kaufman Adam Kohn Matthew A Smith J. A Movshon Nicholas J Priebe Steven G Lisberger Stephen I Ryu Krishna V Shenoy John P Cunningham Mark M Churchland Google Jeffrey S Seely Matthew T Kaufman Adam Kohn Matthew A Smith J. A Movshon Nicholas J Priebe Steven G Lisberger Stephen I Ryu Krishna V Shenoy John P Cunningham Mark M Churchland Google Scholar Jeffrey S Seely Matthew T Kaufman Adam Kohn Matthew A Smith J. A Movshon Nicholas J Priebe Steven G Lisberger Stephen I Ryu Krishna V Shenoy John P Cunningham Mark M Churchland PubMed Jeffrey S Seely Matthew T Kaufman Adam Kohn Matthew A Smith J. A Movshon Nicholas J Priebe Steven G Lisberger Stephen I Ryu Krishna V Shenoy John P Cunningham Mark M Churchland Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.
Referência(s)