Competition in Cross-situational Word Learning: A Computational Study

TL;DR

This paper presents a computational model demonstrating how mutual exclusivity bias and competitive mechanisms facilitate early word learning under uncertainty, emphasizing the importance of algorithmic strategies in cognitive modeling.

Contribution

It introduces a novel computational framework analyzing the role of mutual exclusivity and competition in cross-situational word learning, providing insights into mechanisms that reduce uncertainty.

Findings

Mutual exclusivity bias helps reduce ambiguity in word learning.

Competition among words and referents is crucial for successful learning.

Algorithmic analysis clarifies mechanisms underlying word learning processes.

Abstract

Children learn word meanings by tapping into the commonalities across different situations in which words are used and overcome the high level of uncertainty involved in early word learning experiences. We propose a modeling framework to investigate the role of mutual exclusivity bias - asserting one-to-one mappings between words and their meanings - in reducing uncertainty in word learning. In a set of computational studies, we show that to successfully learn word meanings in the face of uncertainty, a learner needs to use two types of competition: words competing for association to a referent when learning from an observation and referents competing for a word when the word is used. Our work highlights the importance of an algorithmic-level analysis to shed light on the utility of different mechanisms that can implement the same computational-level theory.