Abstract
In sensory systems, a range of computational rules are presumed to be implemented by neuronal subpopulations with different tuning functions. For instance, in primate cortical area MT, different classes of direction-selective cells have been identified and related either to motion integration, segmentation or transparency. Still, how such different tuning properties are constructed is unclear. The dominant theoretical viewpoint based on a linear-nonlinear feed-forward cascade does not account for their complex temporal dynamics and their versatility when facing different input statistics. Here, we demonstrate that a recurrent network model of visual motion processing can reconcile these different properties. Using a ring network, we show how excitatory and inhibitory interactions can implement different computational rules such as vector averaging, winner-take-all or superposition. The model also captures ordered temporal transitions between these behaviors. In particular, depending on the inhibition regime the network can switch from motion integration to segmentation, thus being able to compute either a single pattern motion or to superpose multiple inputs as in motion transparency. We thus demonstrate that recurrent architectures can adaptively give rise to different cortical computational regimes depending upon the input statistics, from sensory flow integration to segmentation.
Highlights
Sensory inflows received by animals are highly complex and ambiguous with multiple local sensory events occurring simultaneously
We show that the different MT neuronal classes reported previously are steady-state solutions of the dynamical system, their strength and stability varying with both input statistics and excitationinhibition balance
We focused on the properties of the steady state solutions such as the shape of the direction tuning functions, their number of peaks at convergence and the peak positions with respect to the driving inputs
Summary
Sensory inflows received by animals are highly complex and ambiguous with multiple local sensory events occurring simultaneously. A challenging computational task faced by any sensory system is to either integrate or segment these local signals in order to encode behaviourallyrelevant information This is well illustrated by visual motion processing. When presented with two motion directions, the primate visual motion system can group them according to linear (i.e., vector average) or nonlinear (i.e., intersectionof-constraints) rules [1, 26]. It can segment them by either suppressing one of the two inputs (i.e., winner-take-all) or simultaneously representing both of them (i.e. superposition) as in the challenging case of motion transparency [45, 35, 47]
Sign-in/Register to view highlights & summary 
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call