Re: Sniffy

[Robert Herdegen]

At 02:06 PM 6/29/2002 -0500, you wrote:
>At 3:27 PM -0500 6/28/02, Kirsten Rewey wrote:
>>At 2:21 PM -0500 6/28/02, Paul Brandon wrote:
>>>At 2:11 PM -0500 6/28/02, Kirsten Rewey wrote:
>>>\>I used Sniffy to teach both classical and operant conditioning in a
>>>>learning class. It worked great! As I recall, students did some of
>>>>the work individually and some in groups. Let me know if you want
>>>>more info.
>>>>
>>>
>>>Have you tried shaping Sniffy as if it were a live rat?
>>>What happened?
>
>I've worked with live lab rats for 40+ years, as well as with a number of
>simulations.
>If you try to shape Sniffy as if he were a live rat (_without_ reading the
>instructions) you'll get nowhere; his behavior does not obey the same rules.
>
>I'd strongly recommend the CyberRat program (see
>http://www.psych-ai.com/CyberRat/CyberRat.html) which should be ready this
>Fall.
>I'm somewhat biased; I've been working (pro bono) with CyberRat's programmers.
>Partly this is because I've used Sniffy and found it unsatisfactory for
>teaching operant conditioning.
>

I would have to second Paul's concerns. I also have worked with live lab rats for instructional purposes for a loooooong time, even back when I was an undergraduate. This last spring, due to equipment breakdowns, insufficient funding for replacement/repair of the equipment, and the inability to find operant conditioning equipment I could borrow, I resorted to Sniffy for use in my Learning lab. Sniffy wasn't nearly as effective as live rats for instruction in operant conditioning. First, as Paul suggested, his (whoops! maybe Sniffy is a she? How do you determine the sex of a virtual rat??) behavior doesn't obey the same rules as a live rat, though the authors of the program admit that up front. (For example, with a little patience and not that much time, one student got Sniffy trained up to a FR5000 schedule. Try doing that with your live Wistar rat!) Second, shaping the response of a virtual rat simply doesn't have the impact of seeing the lever press response acquired by a live rat. Third, there is an option for Sniffy that allows you to "enclose" Sniffy in a sound-attenuating chamber and to "accelerate" the learning process by speeding up time. My students consistently chose this option (I couldn't blame them...who could stand watching a virtual rat wander around a virtual cage and occasionally press a virtual response lever??). But that even further removed them from the experience of seeing the conditioning process occur, making it even less meaningful. Fourth, the instructions were at times confusing for the students (though they just may not have been sufficiently patient to work through them to figure out how to work the program). Etc., etc.

I had the opportunity for a naturalistic experiment in this particular lab course. Two of the students in the lab were repeaters, taking it for the second time (why they were is another story). The first time they went through the lab they used live rats for the operant conditioning portion of lab, and this time used Sniffy. Both said that the live rat was far better than Sniffy, and that Sniffy was something of a waste of time. Also, my 3rd grade daughter used the lab's one working operant chamber, and trained a rat to lever-press for food and perform a light-on/light-off discrimination for a science fair project. I daresay she got far more out of the experience of working with the live rat than my students did out of working with Sniffy.

If you have absolutely no opportunity to use live rats, one of the virtual rats may be better than nothing. Also, the virtual rats may be the only option if you have very large numbers of students to go through the exercise (such as for an introductory-level course). Sniffy may be adequate under such circumstances. But I'd recommend first checking out CyberRat, as recommended by Paul. Sniffy just underwhelmed me.

Bob


****************************************************************
Robert T. Herdegen III
Elliott Professor of Psychology
Department of Psychology
Hampden-Sydney College
Hampden-Sydney, VA 23943
434-223-6166
****************************************************************
--- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED]