Cockroach escape reorganization

From Greg Williams (920324-3)

[from Gary Cziko 920324.1320]

How can the cockroach change his behavior adaptively in this situation if
it doesn't care where its going (e.g., make sure ground speed is faster
than air speed, that means I going downwind, away from the air disturber)?
Please, somebody explain to me how this can happen if the cockroach's
escape behavior is purely ballistic.

The precalibrated settings for ballistic escape might be reset on the basis of
wind direction/intensity AFTER the escape was completed. No feedback control
would be needed during the escape for this to happen.

Greg

[from Gary Cziko 920325.1350]

In reply to my (920324.1320) saying:

How can the cockroach change his behavior adaptively in this situation if
it doesn't care where its going (e.g., make sure ground speed is faster
than air speed, that means I going downwind, away from the air disturber)?
Please, somebody explain to me how this can happen if the cockroach's
escape behavior is purely ballistic.

Greg Williams (920324-3) says:

The precalibrated settings for ballistic escape might be reset on the basis of
wind direction/intensity AFTER the escape was completed. No feedback control
would be needed during the escape for this to happen.

Yes, I suppose that this is possible, but I don't like it very much. If
this is what's happening, the cockroach has to sense something after the
escape routine is finished and then realize "Hey, this isn't where I was
supposed to end up. I better try a new twist to my escape routine the next
time I feel that puff of air. I hope I remember. And I hope that what I
try doesn't put me in a worse position than I'm in now (e.g., under the
cockroach stomper's boot heel)."

It seems to me that the reorganization would seem a lot easier and faster
if the cockroach is getting sensory feedback WHILE it is escaping. Then he
could reorganize successfully (at least to some degree) in one trial.

If we saw the cockroach with the newly waxed cerci getting closer to the
right direction of escape toward the END of the escape move, this would be
good evidence that the move is under feedback control of some type. Here
is where some good cockroach data would be of use. I think I'll send this
to Randy Beer and see what he has to say.--Gary

P.S. That reminds to remind CSGnetters once again that Randy Beer is NOT
on CSGnet, although he sends his replies to Powers, Marken, et al. to
CSGnet. If you want to have your message sent to Beer you have to add his
address after the CSGnet address. This is beer@cthulu.ces.cwru.edu (Tom
Bourbon take special note).

ยทยทยท

------------------------------------------------------------------
Gary A. Cziko Telephone: (217) 333-4382
Educational Psychology FAX: (217) 244-0538
University of Illinois Internet: g-cziko@uiuc.edu
1310 S. Sixth Street Radio: N9MJZ
210 Education Building
Champaign, Illinois 61820-6990
USA
------------------------------------------------------------------