English
 
User Manual Privacy Policy Disclaimer Contact us
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Slow adaptation facilitates excitation-inhibition balance in the presence of structural heterogeneity

Landau, I., Egger, R., Dercksen, V., Oberlaender, M., & Sompolinsky, H. (2016). Slow adaptation facilitates excitation-inhibition balance in the presence of structural heterogeneity. In Computational and Systems Neuroscience Meeting (COSYNE 2016) (pp. 34-35).

Item is

Basic

show hide
Item Permalink: http://hdl.handle.net/21.11116/0000-0000-7D1E-4 Version Permalink: http://hdl.handle.net/21.11116/0000-0003-CC9B-9
Genre: Meeting Abstract

Files

show Files

Locators

show
hide
Locator:
Link (Any fulltext)
Description:
-

Creators

show
hide
 Creators:
Landau, I, Author
Egger, R1, 2, Author              
Dercksen, VJ, Author
Oberlaender, M1, 2, Author              
Sompolinsky, H, Author
Affiliations:
1Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497794              
2Former Research Group Computational Neuroanatomy, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_2528698              

Content

show
hide
Free keywords: -
 Abstract: Traditional analysis of cortical network dynamics has most commonly treated simple random graph structure. We present anatomy-based estimates of connectivity statistics within local circuits of the rat barrel cortex. We observe that the in-degree from within a single cell-type is significantly broader than expected from a simple random graph, and that in-degrees from different cell-types are substantially correlated. Simulations of LIF networks with connectivity structure from data reveal unbalanced networks in which a large majority of neurons are totally silent and those cells that fire do so at high rates and with temporal regularity. Analytically, we study a generic model of networks with broad and correlated in-degrees. We show that in general, networks with broad in-degree distributions cannot maintain the dynamic balance of excitation and inhibition, and the dynamics are mean driven. Correlated in-degrees can mitigate this effect and enable the recovery of balance and fluctuation-driven irregular firing. We analytically determine the structural boundary for maintaining balance and find that the connectivity estimates from anatomy fall outside of the balance regime. We present a novel dynamical state in which a slow adaptation current corrects for the structural imbalance locally and facilitates the global emergence of balance. We find that moderate adaptation currents, of the same order of magnitude as those observed in both excitatory and inhibitory cortical neurons, are sufficient to significantly mitigate the impact of structural imbalance. Finally, we explore the relationship between connectivity and activity that emerges in the adaptation-facilitated balanced state. Population activity is primarily distributed along a single dimension of the underlying connectivity structure, and this dimension is determined by the excitation-inhibition balance rather than by the structure of network connectivity.

Details

show
hide
Language(s):
 Dates: 2016-02
 Publication Status: Published in print
 Pages: -
 Publishing info: -
 Table of Contents: -
 Rev. Method: -
 Identifiers: BibTex Citekey: LandauEDOS2016
 Degree: -

Event

show
hide
Title: Computational and Systems Neuroscience Meeting (COSYNE 2016)
Place of Event: Salt Lake City, UT, USA
Start-/End Date: 2016-02-25 - 2016-02-28

Legal Case

show

Project information

show

Source 1

show
hide
Title: Computational and Systems Neuroscience Meeting (COSYNE 2016)
Source Genre: Proceedings
 Creator(s):
Affiliations:
Publ. Info: -
Pages: - Volume / Issue: - Sequence Number: T-18 Start / End Page: 34 - 35 Identifier: -