[ACT-R-users] Spreading Activation and Retrieval Chunks

[Wolfgang Schoppek]

I'm wondering why the term "working memory" has never been mentioned 
in this discussion. Up to ACT-R 4 and still, the goal chunk was used
as working memory rather than as description of the goal. Currently,
many researchers are exploring the benefits of having multiple buffers
(see e.g. the imaginal buffer in the paper "An integrated theory of
the mind" by Anderson, Byrne, and Lebiere to be found at the ACT-R 
web site). I'm experimenting with a free structure that can keep
a number of chunks active without binding them to specific slots. I 
think this is appropriate for tasks where subjects cannot make sense
of the material from the beginning but rather have to form their own
structures in a learning process.
 I hope that all these efforts will result in an ACT-R theory of 
working memory. In the meantime, it would be helpful to have an 
experimental version of ACT-R where the behavior of the buffers can be
felxibly controlled by parameters (e.g. turning on and off source 
activation for all buffers like with the :ga parameter for the
goal chunk).
-- Wolfgang
-----------------------------------------------------------------
Dr. Wolfgang Schoppek          Universitaet Bayreuth
 Tel.: +49 921 554140
 http://www.uni-bayreuth.de/departments/psychologie/wolfgang.html
-----------------------------------------------------------------