Gerard Kempen’s Website

Text

Gerard A.M. Kempen (1943) is Research Associate at the Max Planck Institute for Psycholinguistics in Nijmegen since 1999, and Professor of Cognitive Psychology at Leiden University since 1992 (emeritus since 2008). From 1976 to 1992 he was Professor of Psycholinguistics at the Radboud University in Nijmegen, where he had received his PhD in 1970.

His psycholinguistic work mainly concerns the grammatical aspects of human sentence processing during language production and comprehension: human grammatical coding. He is studying this topic through a combination of experimental-psychological, linguistic, and computational methods. His contributions include:

1. •the division of labor and the interaction between conceptualization and formulation processes in sentence production (1977)
   

2. •the concept of incrementality in language production (1982)
   

3. •the distinction between lemmas and lexemes in the Mental Lexicon for language production (1983; with Pieter Huijbers)
   

4. •an early computational model of human sentence production (1987; with Eduard Hoenkamp)
   

5. •the (neuro)cognitive model of parsing called Unification Space (1989-2009; with Theo Vosse)
   

6. •the (neuro)cognitively motivated Performance Grammar formalism (PG; 1991-2003; with Koenraad De Smedt and Karin Harbusch; see Harbusch’s  special PG page here)
   

7. •a grammatical theory of clausal coordinate ellipsis based on similarities with speech error repairs (2009)
   

8. •experimental and theoretical arguments for major overlap between the grammatical processing mechanisms underlying sentence comprehension and those underlying sentence production (1997-2011; with Nomi Olsthoorn and Simone Sprenger)
   

9. •a new model of self-monitoring for grammatical and lexical speech errors, which does not require the “perceptual loop” (2011).
   

For a categorized list of Kempen’s papers on human grammatical coding (1977-present), see this page.

Since 1980, Kempen initiated and supervised various theoretical and applied research projects dealing with the computational treatment of Dutch, among other things, for visual-interactive teaching of grammatical structures (sentence analysis) in secondary education.

In the course of the latter project, he (re)discovered (in 1993 or thereabout):

An imaginative an unusual visualization of syntactic structure: Jac. van Ginneken's living grammar from De Roman van een Kleuter ('A Toddler's Novel') (1917). Here is the original Dutch text accompanying the visualizations in De Roman; the German translation by Berry Claus is here.

Recently finished project (2004-2010; funded by NWO as part of the Cognition program): PLUS: A neurocomputational model for the Processing of Linguistic Utterances based on the Unification-Space architecture. With Peter Hagoort, Jos van Berkum, Theo Vosse, and Tineke Snijders.

Last update: February 2, 2012