Postdoctoral Research Associate
Department of Psychology
Program in Neural, Informational and Behavioral Sciences
University of Southern California

Ph.D. in Cognitive Science, Centre for Cognitive Science, University of Edinburgh

==================================================

Research Interests

Neural Network Models of Language Learning and Processing, especially of syntax and infant speech segmentation; Psycholinguistics; Computational Modeling of Cognition; Cognitive Science; Sequential Learning and Memory Processes; Cognitive Neuroscience.


Click here to see a research statement.

==================================================


Click on any of the topics to go there directly:


==================================================

Current Projects

My current research centers around the study of performance constraints on the learning and processing of language and other complex sequentially structured information. Below are some of the closely related experimental and a computational projects that I am currently working on (clicking on links will bring up an abstract):

I am currently applying a self-paced reading task (with on-line grammaticality judgments) to test predictions emerging from a connectionist model (Christiansen, in preparation) concerning the processing of sentences with multiple instances of the same recursive structure (Christiansen & MacDonald, in preparation).
I am working on a paper reappraising the poverty of stimulus argument(s) in the light of recent evidence from developmental psychology, probabilistic approaches to adult language processing, and new insights about learning in neural networks (Seidenberg, Allen, Christiansen & MacDonald, in preparation).
I am also continuing my work on the modeling of speech segmentation, extending previous work (Allen & Christiansen, 1996; Christiansen, Allen & Seidenberg, in press) to deal with noisy input (Christiansen & Allen, in preparation).
A series of simulations seeks to demonstrate that nonlinguistic limitations on learning in recurrent neural networks may explain why natural languages are predominately either head-first or head-last (Christiansen & Devlin, in preparation).
An experiment is in progress (motivated by the ideas outlined in Morrison & Christiansen, 1995), testing aphasic patients and normal controls on an artificial grammar learning task [collaborator: Richard Shillcock, University of Edinburgh].
I also plan to develop a neural network simulator that allows for the modeling of both comprehension and production within the same recurrent network (similar to the approach taken in Christiansen & Chater, in submission). This simulator would permit further computational studies of speech segmentation and syntactic processing, investigating ways in which production and comprehension may interact and influence each other both during acquisition and adult processing.
Go back to the top

==================================================


Publications

Refereed Articles

Go back to the top

Book Chapters


Work in Submission


Work in Preparation

Go back to the top

Book Reviews


PhD Thesis

==================================================


Address

Program in Neural, Informational and Behavioral Sciences
University of Southern California
University Park MC-2520
Los Angeles, CA 90089-2520
Phone: (213) 740-6299
Fax: (213) 740-5687
Email:
morten@gizmo.usc.edu


To the Language and Cognitive Neuroscience Lab home page

To the Christiansen-Govindjee Family Web Page


visitors so far.

Last modified: January 15, 1997.

rule.gif (155 bytes)

Home | People | Research | Links | Contact | Publications | Presentations
Cognitive Neuroscience Laboratory

Please email suggestions/errors to mhc27@cornell.edu