Semantic Interpretation as Computation in Nonmonotonic Logic: The Real Meaning of the Suppression Task

Abstract

Interpretation is the process whereby a hearer reasons to an interpretation of a speaker's discourse. The hearer normally adopts a credulous attitude to the discourse, at least for the purposes of interpreting it. That is to say the hearer tries to accommodate the truth of all the speaker's utterances in deriving an intended model. We present a nonmonotonic logical model of this process which defines unique minimal preferred models and efficiently simulates a kind of closed-world reasoning of particular interest for human cognition.Byrne's "suppression" data (Byrne, 1989) are used to illustrate how variants on this logic can capture and motivate subtly different interpretative stances which different subjects adopt, thus indicating where more fine-grained empirical data are required to understand what subjects are doing in this task.We then show that this logical competence model can be implemented in spreading activation network models. A one pass process interprets the textual input by constructing a network which then computes minimal preferred models for (3-valued) valuations of the set of propositions of the text. The neural implementation distinguishes easy forward reasoning from more complex backward reasoning in a way that may be useful in explaining directionality in human reasoning.

Reasoning About Uncertain Conditionals

“…This was demonstrated experimentally by the suppression task paradigm [7,48]. The data show that previously endorsed inferences can be suppressed by adding conditionals to the premise sets.…”

The logical response to a noisy world

“…Our aim in this paper is to motivate such a logical account, by showing that probability cannot provide the kind of defeasible framework within which interpretations can be established. Our positive logical account has appeared elsewhere [23,25]; here we concentrate on the shortcomings of probabilistic analyses of defeasible reasoning. We start by discussing a number of faulty assumptions about logic that have hindered its application to actual human reasoning.…”

The Logical Way to Be Artificially Intelligent

“…One of the earliest of these investigations, by Holldobler and Kalinke [15], studied the problem of translating normal logic programs into neural networks. More recently, Stenning and van Lambalgen [16] have argued that the implementation of logic programs by neural networks shows that logic can model intuitive thinking in the dual process model. D'Avila Garcez, Broda and Gabbay [14], on the other hand, studied the problem of extracting higher-level, "meaningful" logic programs from neural networks.…”

“…As Stenning and van Lambalgen [16] argue, natural language communications need to be interpreted to determine their intended logical form, even when those communications are already expressed in logical form. They argue that the gap between the surface logical structure of sentences and the deeper logical structure of their intended meanings helps to explain and refute certain psychological experiments that suggest that people are not logical.…”