We propose a biological cortical column model, at an intermediate mesoscopic scale, in order to better understand and interpret biological sources of voltage-sensitive dye imaging signal (VSD signal). To perform a quantitative analysis of the relative contributions to the VSD signal, a detailed compartmental model was developed at a scale corresponding to one pixel of optical imaging. The generated model was used to solve the VSD direct problem, i.e. generate a VSD signal, given the neural substrate parameters and activities. Here, we confirm and quantify the fact that the VSD signal is the result of an average from multiple components. Not surprisingly, the compartments that mostly contribute to the signal are the upper layer dendrites of excitatory neurons. However, our model suggests that inhibitory cells, spiking activity and deep layers contributions are also significant and, more unexpected, are dynamically modulated with time and response strength.
Copyright 2010 Elsevier Inc. All rights reserved.