Working memory and connectionist models of parsing: A reply to MacDonald and Christiansen (2002)

The authors take issue with 3 claims made by M. C. MacDonald and M. H. Christiansen (2002): (a) that connectionist models of syntactic processing do not have components that function in the same way as limitations on working memory (WM) in production system models of psychological processes; (b) that individual differences in WM are related to the efficiency of syntactic processing because both are affected by reading experience; and (c) that individual differences in the nature of phonological representations affect the efficiency of syntactic processing. The authors argue that the concept of WM cannot be eliminated from models of language processing and that the literature supports the view that syntactic processing involves a specialized WM system.