Parsing Model and a Rational Theory of Memory.

Jakub Dotlačil,Puck De Haan

doi:10.3389/fpsyg.2021.657705

Jakub Dotlačil, Puck De Haan

Open Access

https://doi.org/10.3389/fpsyg.2021.657705

Copy DOI

Abstract

This paper explores how the rational theory of memory summarized in Anderson (1991) can inform the computational psycholinguistic models of human parsing. It is shown that transition-based parsing is particularly suitable to be combined with Anderson's theory of memory systems. The combination of the rational theory of memory with the transition-based parsers results in a model of sentence processing that is data-driven and can be embedded in the cognitive architecture Adaptive Control of Thought-Rational (ACT-R). The predictions of the parser are tested against qualitative data (garden-path sentences) and a self-paced reading corpus (the Natural Stories corpus).

Highlights

In the rational theory of cognition, it is argued that cognitive functions are largely shaped by our adaptation to the environment
The structure of the paper is as follows: we briefly introduce the rational theory of memory as part of the cognitive architecture Adaptive Control of Thought-Rational (ACT-R)
We provided empirical evidence for the parsing model that is built on the assumptions of the rational theory of memory proposed in Anderson (1991) and embedded in ACT-R

Summary

Introduction

In the rational theory of cognition, it is argued that cognitive functions are largely shaped by our adaptation to the environment In this view, it is assumed that various aspects of our behavior can be explained as the result of the optimization to the structure of the environment. If retrieval is abandoned when the cost for retrieval exceeds some threshold, we expect the less needed an item is, the more likely it is that its recall fails. These predictions have been largely confirmed, see Anderson (1991). The fact that the NSC has a plethora of syntactic constructions and includes manually controlled PTB-compatible syntactic parses makes the corpus usable for the computational modeling of parsing

Objectives

Results

Conclusion