The approach I take to modeling sentence
processing and other cognitive tasks draws on work on general
architectures in cognitive science (Anderson, 1983,1993; Newell,
1972, 1973, 1990; Pylyshyn, 1984). All cognitive modeling seeks
to characterize the information processing that underlies task
performance, but architectural modeling also has the goal of
explicitly characterizing the fixed structure that supports cognition
across a wide range of tasks---the memories, processes, and control
structure.
Building a model within a cognitive architecture has a number
of methodological and theoretical advantages. The model is constrained
by independently motivated assumptions which play a role
in explaining other kinds of behaviors or functionality; the
model is computationally complete, in that it explicitly
specifies the memories, processes, and control structures that
underlie cognition; and the model is integrated in that
all the pieces and components mesh together to yield cognitive
behavior. For example, with respect to working memory theories,
architectures provide convenient vehicles for exploring the functionality
of specific proposals for components embodying limitations of
some kind.
I've constructed a sentence processing architecture based on Soar called NL-Soar, and am currently working on a model of sentence processing in ACT-R .
I also build architecturally-based models of human-device interaction---for example, learning to use an automated-teller machine (Vera, Lewis & Lerch, 1993).
Here are some relevant publications and presentations:
- Lewis, R.L. (in preparation) Cognitive and Computational Foundations of Sentence Processing. Oxford: Oxford University Press.
- Lewis, R. L. (to appear). Computational Psycholinguistics. In Encyclopedia of Cognitive Science. London: Macmillon (Nature Publishing Group). [PDF].
- "Toward Detailed Models of Sentence Processing Constrained by ERP Data." Workshop on Analyzing and Modelling Event-Related Brain Potentials: Cognitive and Neural Approaches. Potsdam University, Germany, November 2001. [PDF]
- "Does the Mind Need a Bottleneck? Toward a functional analysis of bottlenecks, executive processes, and control structure." NASA Ames Research Center, Cognitive Group, November 2001. [PDF]
- Lewis, R.L. (2001, in preparation) Cognitive theory, Soar. In International Encylopedia of the Social and Behavioral Sciences. Amsterdam: Pergamon (Elsevier Science). [PDF,]
- Lewis, R.L. (1999) Cognitive modeling, symbolic. In Wilson, R. and Keil, F. (eds.), The MIT Encyclopedia of the Cognitive Sciences. Cambridge, MA: MIT Press. [PDF]
Lewis, R. L. (1999). Specifying architectures for language processing: Process, control, and memory in parsing and interpretation. In Crocker, M., Pickering, M., and Clifton, C. Jr., editors, Architectures and Mechanisms for Language Processing. Cambridge University Press. [postscript]
Lehman, J. F., Lewis, R. L., and Newell, A. (1998). Architectural influences on language comprehension. In Pylyshyn, Z., editor, Cognitive Architecture. Ablex, Norwood, NJ.
Lewis, R. L. (1996). Architecture Matters: What Soar has to say about modularity. In Steier, D. and Mitchell, T., editors, Mind Matters: Contributions to Cognitive and Computer Science in Honor of Allen Newell. Erlbaum, Hillsdale, NJ. [postscript]
Lewis, R. L. (1993). An Architecturally-based theory of Human Sentence Comprehension. PhD thesis, Carnegie Mellon University. [PDF] Also available as Computer Science Tech Report CMU-CS-93-226; send requests to reports@cs.cmu.edu.
Vera, A. H., Lewis, R. L., and Lerch, F. J. (1993). Situated decision-making and recognition-based learning: Applying symbolic theories to interactive tasks. In Proceedings of the Fifteenth Annual Conference of the Cognitive Science Society. [PDF]
[Home | Top of page]