test
Search publications, data, projects and authors

Thesis

English

ID: <

10670/1.dbh0iq

>

Where these data come from
Lateral connectivity : propagation of network belief and hallucinatory-like states in the primary visual cortex

Abstract

In the primary visual cortex (V1), we examined the functional impact of centripetal apparent motion sequences originating from the far periphery and converging towards the receptive field of cortical cells along their preferred orientation axis. At high saccadic speed, the anisotropic congruency of elementary stimuli composing a coherent motion is crucial in the diffusion and lateral integration of contextual information. At the electrophysiological level, those results correspond to a latency advance and an amplitude gain of sub and suprathreshold responses, indicating the existence of a dynamic association field where form and motion are already bound in V1. Restricting the apparent motion to the silent periphery result in an invasion of the receptive field by predictive activity. This latter suggests the existence of a mechanism of lateral diffusion intrinsic to V1 that allows to solve the motion extrapolation problem. Second, we posit that geometric hallucinations reflect a long-distance spatial opponency of horizontal connectivity that structure the self organization of V1 ongoing activity, expressing itself through a model of interacting hypercolumns resulting in the formation of neural stripes on V1 surface. We designed visual stimuli in which perturbation by a 1/fα noise of a network highly adapted to geometric inducers result in perception of opponent planforms. Our results suggest that those dynamic percepts correspond to propagating waves of synaptic activity that are detectable at the level of V1 cells under the form of oscillations compatible with the local geometry and the dynamic of the induced percepts.

Your Feedback

Please give us your feedback and help us make GoTriple better.
Fill in our satisfaction questionnaire and tell us what you like about GoTriple!