“Œ‹ž‘εŠwHŠw•”Œv”HŠw‰Θ@ƒVƒXƒeƒ€ξ•ρ’k˜b‰ο

“ϊŽžF2007”N9ŒŽ14“ϊi‰Ξj@11:00--12:30

κŠF“Œ‹ž‘εŠwHŠw•”6†ŠΩ@3ŠKƒZƒ~ƒi[ŽΊA i–{‹½ƒLƒƒƒ“ƒpƒXj@@@

(ƒAƒNƒZƒX) http://www.i.u-tokyo.ac.jp/map/index.shtml#hongo

uŽtFBijoy K. Ghosh ‹³Žφ
         (Texas Tech University)

‘θ–ځFControl and Signal Processing in Animal Vision

ŠT—vF In this problem we study how the visual circuit of an animal is able to track a mobile target in its visual space. The first problem we discuss is the gaze control problem wherein the rotational dynamics of the eye is introduced. The control objective is to ensure that the gaze of the eye is always directed towards a mobile target. Rotational motion of an eye satisfies a constraint known as the eListingfs Lawf. An optimal control is derived that ensures the eye movement between an initial and final gaze while satisfying the Listingfs constraint. The second problem we discuss is how a population of neurons encodes images on the retina. Specifically we show that visual images produce a wave of visual activity in a portion of the visual circuit known as the evisual cortex.f A dynamic model of the visual cortex with four distinct classes of neurons is introduced that produces a cortical wave. Statistical analysis of the wave shows that they encode position and velocity cues from the visual space. An animal can use the cortical activity patterns to decode these features from the visual space.

˜A—ζ: Œ΄ ’CŽŸ (shinji_hara (at) ipc.i.u-tokyo.ac.jp)
            ¬“‡ηΊ (chiaki_kojima (at) ipc.i.u-tokyo.ac.jp)