[From Bruce Abbott (950308.1900 EST)]
Bill Powers (950308.1325 MST)
My correlations were about 0.94 for two runs, with an RMS error of about
37. When I looked at the display, I realized that there is a
considerable lag in switching to the new target. So I went into your
program and changed the fitting program to use a delayed version of the
state marker. The altered function is listed at the end.
Here's my results for run 008 (I'm getting better at it...):
delay k r rms
0 0.0264 0.968 29.45
25 0.0518 0.989 17.51
30 0.0693 0.990 16.63
35 0.1006 0.988 17.90
40 0.1631 0.983 21.59
Looks like our performances are comparable: I get a best lag of 30/60 and
yours was 37/60. There's something lurking in memory somewhere that tells
me that this 1/2 second lag has been found in other kinds of studies
involving discriminated reaction times and that people do not perceive the
lag. Perhaps someone else remembers the study; at the moment I sure can't.
To investigate the higher systems in a little more detail, we should
probably use about 5 targets and 5 colors of the cursor. If the
relationship of color to target is randomized at the beginning of a
series of runs, we could see the process by which the person learns
which position goes with which color. There should be enough memory to
store a 5-minute run (that's 36000 bytes per array). You may want to
switch to storing the binary data for such long runs; the ASCII files
would tend to get pretty big.
This is a great idea. Binary storage is no problem; the use of ASCII was
simply to permit direct import into other analysis programs such as Minitab,
and that's not important here. (Besides, if ASCII were needed, the
conversion is trivial.)
Before we get too carried away by this experiment, how are "our" rats
doing? Any problems getting the apparatus set up? Are any of them still
alive?
Er, well, I've had trouble finding the time to get things set up, but it's
on my "to-do" list. (Right now I'm busy organizing a certain talk I'm
scheduled to give in Michigan in just over two weeks.) The problem is that
I don't have any slaves (graduate students) to do some of this set-up work
for me. (And I thought the phrase "n of 1" research referred to the number
of _subjects_.) I sometimes feel like the guy in the airline skit who takes
the tickets, loads the baggage onto the plane, then puts on his captain's
hat and flies the plane.
If there are any of you who want to run these programs, I think that
Bruce Abbott can be persuaded to put them on his server and repeat the
instructions for downloading them to a PC.
I'd be happy to. If you'd like an executable copy of any of the programs,
please inquire direct and I'll tell you how to do it via FTP. The programs
will run on an IBM-compatible PC with a VGA screen display and a mouse.
Bill, you should be receiving some material on stimulus control in the mail
in a couple of days.
ยทยทยท
-------------------------------------------------------------------
Samuel Saunders (950307:1310 EST)
Sam, thanks for the comments about conditional discrimination. Just one
minor point:
Bruce Abbott (950306.1505 EST) --
A pigeon has been trained to peck while a vertical line is projected on the
key and not to peck when a horizontal line is projected there. What happens
at the logic level when a line at 45 degrees is projected?What the pigeon actually DOES (at least initially) is peck at an
intermediate rate.While this is the majority view in EAB, there is a sizeable minority that
would argue that intermediate stimuli produce a weighted combination of
the behaviors appropriate to the end-points, and averaging over a stimulus
presentation gives the appearance of intermediate responding. Research on
this issue has been inconclusive.
You left out the sentences immediately following the last one you quoted:
Does the logic-level system try to set positive references to TWO
lower-level systems (keypeck rate and, say "exploration"? Or should
one conceive of the logic-level system as setting a single reference
(to keypeck rate) to positive or zero depending on the S-D?
I believe these questions imply the two views you describe. My bias at the
moment is toward the weighted-combination view.
Regards,
Bruce