To answer Stephen's question, I think you would need at least one more group:
Group 3: PRF --> EXT --> EXT
And, according to Stephen, the measure of trials or time to an extinction
criterion would begin at the end of the first phase in each case.
BTW, something about Stephen's original post made me think that the partial
reinforcement extinction effect, or at least explanations for it, mirrors an
effect seen in aversive learning. "Vicious circle" behavior can be seen during
extinction of an escape response. The standard procedure involves a long runway
with an electrified floor. (The shock does not have to be intense.) It works
best if the rats "dropped" into the start box using a trap-door arrangement. The
rats must run to the other end of the runway to escape the shock. The escape
response will extinguish fairly rapidly if shock is turned off in the entire
runway, but will persist indefinitely if just the middle portion of the runway
remains electrified. That is, the rats will actually run some distance without
being shocked, cross the electrified portion, and then continue running to the
safe box.
The usual explanation is similar to Amsel's explanation of PREE. Supposedly, the
frustration encountered during extinction is similar to the frustration
encountered during partial reinforcement, so the response persists under these
similar cues. In vicious circle, the explanation that is frequently given is that
the discomfort/anxiety experienced during "punished" extinction is similar to the
discomfort/anxiety experienced during acquisition of the escape response, so the
response persists under these similar cues.
Kirsten Rewey wrote:
> Stephen -
>
> I'm not an expert in learning, but I think Domjan (1998) has answered the
> question for you. In his text he states that Jenkins (1962) and Theios (1962)
> independently tested the effects of CRF after partial reinforcement (PRF) on
> time to extinguish a behavior. Here are the basics of the design:
>
> Group 1: PRF -> CRF -> Extinction
>
> Group 2: CRF -> CRF -> Extinction
>
> Both Jenkins and Theios found that extinction (to criterion) took longer in
> Group 1 (with early experience with PRF and recent experience with CRF) than
> in Group 2 (with no PRF experience).
>
--
********* http://www.coe.uca.edu/psych/scoles/index.html ********
* Mike Scoles * [EMAIL PROTECTED] *
* Department of Psychology * voice: (501) 450-5418 *
* University of Central Arkansas * fax: (501) 450-5424 *
* Conway, AR 72035-0001 * *
*****************************************************************
