1 These slides correspond to a short presentation

1 These slides correspond to a short presentation (5 minutes) given at Neurocomp 2008. Slides 10 — 14 are the main ones. FACETS

2 A Simple Mechanism to Reproduce the Neural Solution of the Aperture Problem in Monkey Area MT Ref. 4. 4 Escobar Pierre Kornprobst Maria-Jose INRIA Sophia Antipolis Odyssée project team Guillaume S. Masson INCM CNRS, Marseille

3 MT cell preferred direction?

4

5 MT cell preferred direction? Classically, one consider drifting gratings

6 MT cell preferred direction? Representation of direction tuning of the MT cell

7 But is it so simple?

Is the MT preferred direction invariant to input stimulus? 8

Is the MT preferred direction invariant to input stimulus? For example… ? ? ? ? Does the MT cell have the same preferred direction? 9

Is the MT preferred direction invariant to input stimulus? NO The MT PD evolves in time and depends on the input stimulus. It is also proportional on the barberpole aspect ratio early Contours late Terminators Pack C. , G artland A. and Born, R. Integration of Contour and Terminator Signals in V isual Area MT of Alert Macaque. The Journ al of Neuro science, 24(13): 326 8 -3280, 200 4. 10

11 Terminators are very important If the terminators are removed, the PD shifting and perception is lost. Contours Terminators MT PD shifting is related with the ratio terminators/contours But the dependence is not linear : If the input stimulus is 2 times bigger, the terminator effect becomes stronger.

The main focus of the poster will be… V 1 cells with surround suppression are more sensitive to terminators than contours. Surround suppression is stronger in V 1 layer 4 B Most of the information coming from V 1 to MT contains terminators info 90% of inputs from V 1 to MT come from 4 B 12

13 So… we propose a model… V 1 And we test it! MT

Results MT PD drifting grating barber pole aspect ratio 2: 1 barber pole aspect ratio 4: 1 early without V 1 surround inhibition late with V 1 surround inhibition 14