Oxford Journals
tools journals homepage advanced search contact help
Journal of Logic and Computation: Current Issue

OUP > Journals > Computing/Engineer. & Mathematics/Stats. > Journal of Logic and Computation

Journal of Logic and Computation

Volume 13, Issue 1, February 2003: pp. 147-155

Representation and Human Reasoning

Alice G. B. Ter Meulen1

1Center for Language and Cognition, University of Groningen, The Netherlands. E-mail: atm@let.rug.nl

Interpretation and reasoning are two sides of sharing information. Representations of the context and common ground must capture the rich content of what has been said, by linking to situations in the world as well as to what has been said before, common sense and to the presuppositions and entailments. This paper will address two strategies in modelling human information sharing: (1) rich representation/light inference and (2) light representation/rich inference. The first strategy is characteristic of computational modelling relying on unification as inference rule and allowing for underspecification. The second strategy is argued to be better suited to cognitive science interests in modelling human information processing. Memory overload is avoided by light representation, and the theory seeks empirical testing in the judgements we share on the validity of all kinds of inferences. An analysis of dynamic temporal reasoning in time serves to illustrate this account of modelling representation and reasoning.

Keywords: Representation; human reasoning; monmonotonic inference; cognitive science; information sharing; temporal reasoning

Table of Contents   Full-Text PDF (49 KB)

Oxford University Press
Published by Oxford University Press
Copyright ©Oxford University Press 2003
Print ISSN: 0955-792X  Online ISSN: 1465-363X.
Oxford University Press Privacy Policy and Legal Statement