analog rat experiment

Tom Bourbon [941101.1620]

[From Bill Powers (941028.0845 MDT)]

Bill, your prof at Northwestern was Janet Taylor -- Janet Taylor Spence
began to chair the psych department at Univ. of Texas at Austin shortly
after I finished there.

Skipping to the end of your post, where you summarized your intent:

Some day I hope we can try out analog control as an alternative to the
"rate of behaving" measure which seems to be the only thing studied in

I'll second that idea. In your post you described a couple of possible

. . . In the one-dimensional case,
an example would be a long position-sensitive bar enabling a computer to
tell where on the bar the rat was pressing on it. The number of presses
required for a reinforcement would be a function of x, the distance from
some point x0 on the bar. So this would be an FR(|x-x0|) schedule.

The question to be answered is whether the rat would find the best place
to press by a systematic method, by a reinforcement-based method, or by
the E. coli method. A sub-question would be whether we could distinguish
a reinforcement-based method from an E. coli method.


Suppose we mount the bar on a lever attached to a potentiometer. The
lever (counterweighted) can be pulled down through an arc of perhaps 90
degrees against a mild spring. The potentiometer can be connected to the
game-paddle input of a computer, or to an A/D converter.

The gimmick is that the rat is rewarded for bringing the lever to a
certain angle within some small range and holding it there for some
length of time. If the reward dish is far enough from the lever the rat
will have to let go to collect the reinforcer, thus letting the lever
rise to its home position for the next trial.

with some additional details that I omit here.

I think this is an important idea to consider -- using analog control in
animal studies. Do you remember the work by Kavanaugh, in the 1960s, that I
have described a few times on the net? He worked with recently-captured
wild mice (Peromyscus). He showed that they would readily operate various
manipulanda (including analog devices like running wheels, and discrete
devices like repetitively-operated microswitches) (1) to consistently oppose
or reverse all kinds of environmental conditions (discrete or continuous)
imposed by the experimenter, and (2) to create their own environmental
conditions, such as turning the lights up and down to produce a light-dark
cycle like the one they had experienced in the wild. (They went to sleep,
and to work, on schedule -- AFTER they had set the lights to the appropriate
levels.) I think Kavanaugh's work shows that animals -- at least wild ones
-- can engage in all kinds of analog actions to control their environments
(their perceptions).

Then there is the work by Marwine and Collier and associates, reported by
Timberlake who re-plotted the original data in a way that showed how animals
varied their actions, under different ratio schedules, and kept constant the
amount of food and-or water they consumed each day. The animals performed
the required ratios and earned free access to food or water -- eg., press
the required number of times and food is available for as long as you stay
in the food bin; leave the bin and the door closes and it only opens again
after you press the required number of times. The evidence for control is
clear and crisp, but Timberlake tried to model it as control of behavior
(pressing and consuming), not control of the amount of food taken in, or
perceptions thereof.

A couple of my thesis students tried to create more-or-less analog
equivalents of the schedule studies reported by Timberlake: one of them
placed food at the end of a long ramp and recorded number of trips a rat
made along the ramp, time spent at the end, and amount of food consumed
during each visit to the end -- all as functions of the angle of inclination
of the ramp. Another did something similar, only the food was placed on a
small "island" in a pool of water, located at various distances away from
the home platform. Technical problems overwhelmed us in those studies --
adequate equipment was not available. Nonetheless, the animals performed
like control systems with refeences for perceptions of food and water levels
and for perceptions of effort expended to obtain food and water. The ramp
and island studies aren't quite like the possibilities you described, but
they were attempts to observe animals under conditions where we might learn
more about them as control systems.

I think it might be time well spent if we were to devise a set of
experiments in which animals could be studied as controllers, then make
them available to anyone on the net who has facilities to do the work.