In the case of paired-associate learning of elements that start with
spaces of equal size, have no overlap and, therefore, no associative probability on each other, we assume that contiguity leads to overlap and
that overt responding of the "response" item B establishes a larger space
for it than for the "stimulus" item A. This would account for the relatively
weaker B-A association. Another possibility is that the functional A may
be different from the nominal A (in Underwood's terminology). In particular, this may be the case when A is not pronounced, and it would
lead to the same prediction since the overlap produced by contiguity
is not between A and B but between some A′ and B, and the probability
of A (given B) would necessarily be reduced. Finally, it is of interest
to note that, if we assume that paired-associate learning of "nonsense"
CVCs produces a spatial arrangement such that B is not only larger but
also that all of A is contained in B, we have established a point of contact
with the logical model described earlier. In this case (A contained
in B) the probability model is isomorphic with the logical concept "If
A then B" or "All A are B" which was one of the input transformations
suggested for the paired-associate situation.Returning to the mediation studies and the trend to make them into
concept formation experiments, it is obvious that the logical and quasilogical transformations go even further in that direction. I am suggesting certain habitual conceptual transformations of input in the case
of adult subjects. If the mediation phenomena are "conceptual" (either in
associative or logical terms) one of the problems with the four-stage
paradigm might be that it overtaxes the informational capacity of the
organism. The amount of information to be used by the time the last
stage is reached might be more than the organism can "hold." The success
of the Russell and Storms experiment ( 1955) with preestablished normative associations may be a function of the easier availability of the first
stages which have been overlearned and well established prior to the
experimental procedure. In any case, Professor Jenkins's suggested further
research should shed some light on this problem regardless of whether
subjects are "associating' or "concept forming."
| Kjeldergaard P. M., &
Horton D. L. ( 1960) An experimental analysis of
associative factors in stimulus equivalence, response equivalence and
chaining paradigms. Studies in verbal behavior. Rep. No. 3, NSF
Grant, University of Minnesota.|
| Mandler G., &
Cowan P. A. ( 1962) Learning of simple structures. J. Exp.
Psychol., 64, 177-183.|
| Russell W. A., &
Storms L. H. ( 1955) Implicit verbal chaining in paired-
associate learning. J. Exp. Psychol., 49, 287-293.|
Questia, a part of Gale, Cengage Learning. www.questia.com
Book title: Verbal Behavior and Learning: Problems and Processes:Proceedings.
Contributors: Charles N. Cofer - Editor, Barbara S. Musgrave - Editor.
Place of publication: New York.
Publication year: 1963.
Page number: 252.
This material is protected by copyright and, with the exception of fair use, may
not be further copied, distributed or transmitted in any form or by any means.