For many years, voltage sensitive dye imaging (VSDI) has enabled the fruitful analysis of neuronal transmission by monitoring the spreading of neuronal signals. Although useful, the display of diffusion of neuronal depolarization provides insufficient information in the quest for a greater understanding of neuronal computation in brain function. Here, we propose the optimal mass transportation theory as a model to describe the dynamics of neuronal activity. More precisely, we use the solution of an -Monge–Kantorovich problem to model VSDI data, to extract the velocity and overall orientation of depolarization spreading in anatomically defined brain areas. The main advantage of this approach over earlier models (e.g. optical flow) is that the solution does not rely on intrinsic approximations or on additional arbitrary parameters, as shown from simple signal propagation examples. As proof of concept application of our model, we found that in the mouse hippocampal CA1 network, increasing Schaffers collaterals stimulation intensity leads to an increased VSDI-recorded depolarization associated with dramatic decreases in velocity and divergence of signal spreading. In addition, the pharmacological activation of cannabinoid type 1 receptors (CB1) leads to slight but significant decreases in neuronal depolarization and velocity of signal spreading in a region-specific manner within the CA1, indicating the reliability of the approach to identify subtle changes in circuit activity. Overall, our study introduces a novel approach for the analysis of optical imaging data, potentially highlighting new region-specific features of neuronal networks dynamics.< Réduire