Home

About

Research

Events

Contact

print

Banbury Center Workshop Series

Neurocomputational Strategies: From Synapses to Behavior (1998)
Functional Organization of the Thalamus and Cortex and their Interactions (1999)
Toward Animal Models of Attention and Consciousness (2000)
Persistent Neural Activity (2000)
Can a Machine Be Conscious? (2001)
Communication in Brain Systems (2004)
Neurobiology of Decision-Making (2005)
Computational Approaches to Cortical Functions (2006)
New Frontiers in Studies of Nonconscious Processing (2007)
Theoretical and Experimental Approaches to Auditory and Visual Attention (2008)
Searching for Principles Underlying Memory in Biological Systems (2009)
Neuronal Response Variability and Cortical Computation (2011)
Connections and Communication in the Brain (2014)

Antoni Guillamon, David W. McLaughlin, John Rinzel

Estimation of synaptic conductances in conductance-based models

New York University

Direct intracellular measurements of membrane potentials (vm) over several periods of activity are important both for unveiling cortical mechanisms and for validating cortical models. Given measurements of membrane potential vm(t; Iapp) for different values of the injected current Iapp , estimations of total synaptic conductance (gT(t)), as well as its excitatory and inhibitory components, are conceivable. Recent examples of such measurements and estimates can be found in [Borg-Graham et al., Visual input evokes transient and strong shunting inhibition in visual cortical neurons, Nature 393: 369-73, 1998; Anderson et al., Orientation tuning of input conductance, excitation, and inhibition in cat primary visual cortex, J. Neurophysiol. 84. 909-929, 2000].

Computational models can be used to test the validity and accuracy of such methods of data analysis. Here we consider several conductance-based models of type

C dvm = - Ileak - INa - IK - Isyn + Iapp ,
dt

with Isyn = gexc (vm - Vexc ) + ginh (vm - Vinh ), where gexc = gexc (t ) and ginh = ginh (t ) represent, respectively, the excitatory and the inhibitory synaptic conductances. We drive the synaptic input to the model equations in two ways: (1) with a periodic prescribed drive, and (2) with data from the simulations of a computational network of the layer 4C? of the visual cortex, introduced in [McLaughlin et al., A neural network model of macaque primary visual cortex, Proc. Nat. Acad Sci. 2000].

We use these models to study ( i ) different methods for smoothing the temporal profiles of the membrance potential and for clipping spikes from these profiles, and (ii) a linear regression to extract the synaptic conductances, gr (t), gexc (t), and ginh (t), from the smoothed data. These estimates of the synaptic conductances are then compared their prescribed time courses in the computational model.

Our conclusions confirm that the linear estimations of synaptic conductances are reliable if the measurements are carried out in non- spiking regimes; however, the model establishes that qualitative inaccuracies arise in the presence of strong spiking - and clearly identifies the origin of these inaccuracies. Finally, consequences for the interpretation of the measurements of Anderson, et al. are discussed.