Click here to download document.
Background
In the 1984-85 academic year, preprints of the 2-volume Parallel Distributed Processing by Jay McClelland, Dave Rumelhart and the PDP Research Group (MIT Press, 1986) were being circulated and discussed among cognitive scientists. In Fall 1984, Jay McClelland moved from the University of California, San Diego (where the PDP books were mainly written) to the Psychology Department of Carnegie-Mellon University. A number of other prominent connectionists also joined the CMU faculty about the same time. This set up an intellectual convergence (perhaps more aptly, collision) between the new PDP movement and leading proponents of production system architectures for symbolic processing who were already on the CMU faculty (e.g., John Anderson, Allan Newell and Herbert Simon).
Walter Schneider and I were visitors at CMU during that exciting year. We co- organized a one time seminar to promote discussion of alternative architectures of cognition. This event was held in the CMU Department of Psychology, Baker Hall 336B, on Friday April 12, 1985, 3:30-5:30pm. Over the preceding week and a half, many people posted their thoughts on an online forum, which became known as the “PSPDP Bulletin Board” (“Production Systems and Parallel Distributed Processing”). The attached file was scanned from my own paper printout of that discussion. It includes comments from (in order), Holyoak, John Anderson, Holyoak, Walter Schneider, David Klahr, Jeff Shrager, Jay McClelland, Schneider, Anderson, McClelland.
–Keith Holyoak (7/18/2012)