Subject: Re: NIDCD plan From: Harry Erwin <herwin(at)GMU.EDU> Date: Sat, 30 Dec 1995 19:45:29 -0500Harry Erwin here, responding to Al Bregman <bregman(at)HEBB.PSYCH.MCGILL.CA>. I was at the bat session at the ASZ conference last week, and what I heard indicates to me that bats quite possibly use some of the same mechanisms that man-made radars use to track objects in clutter. One of these mechanisms is the use of a narrow (in time) tracking gate to avoid hearing echoes from clutter and insect jamming. In a neural system this would basically involve some biological counterpart to Kalman filtering to establish the timing of when to open the gate and what az and el to be expecting the target to appear at. There is evidence from studies of olfaction (Freeman 1979-1995) that primary cells can be brought nearly to spiking by reafference, held there by opposing excitatory and inhibitory signals, and then released to respond _immediately_ to the first afferent stimulus thereafter. That general approach can be applied to a lot of problems that need the types of top-down processing that Al Bregman speaks of. Hopefully, I'll have more to say on this in a year and a half after I've finished modeling it. Harry Erwin Internet: herwin(at)gmu.edu Web Page: http://osf1.gmu.edu/~herwin (contains 'indecent' academic material) PhD student in comp neurosci: "Glitches happen" & "Meaning is emotional" Lecturer for CS 211 (advanced C++)