Acquiring Receptive Morphology: A Connectionist Model

TL;DR

This paper presents a modular connectionist model that learns various receptive morphological processes from phonetic inputs, demonstrating the ability to acquire complex inflectional rules and constraints.

Contribution

It introduces a novel connectionist architecture capable of learning multiple morphological processes directly from phonetic inputs, including complex and reduplication rules.

Findings

The model successfully learns suffixation, prefixation, infixation, and circumfixation.

It can acquire mutation, template, and deletion rules.

It incorporates constraints against association-line crossing.

Abstract

This paper describes a modular connectionist model of the acquisition of receptive inflectional morphology. The model takes inputs in the form of phones one at a time and outputs the associated roots and inflections. Simulations using artificial language stimuli demonstrate the capacity of the model to learn suffixation, prefixation, infixation, circumfixation, mutation, template, and deletion rules. Separate network modules responsible for syllables enable to the network to learn simple reduplication rules as well. The model also embodies constraints against association-line crossing.