What Drives the Use of Metaphorical Language? Negative Insights from Abstractness, Affect, Discourse Coherence and Contextualized Word Representations

TL;DR

This study investigates whether linguistic and cognitive properties like frequency, abstractness, affect, coherence, and contextualized representations can predict metaphorical language use, finding these features alone are insufficient for accurate prediction.

Contribution

The paper introduces five simple models based on linguistic and cognitive properties to predict metaphorical versus literal language use in context.

Findings

Models based on selected properties do not fully explain metaphorical language choices.

Linguistic and cognitive features alone are insufficient for accurate metaphor prediction.

Human judgments reveal limitations of current models in capturing metaphorical language use.

Abstract

Given a specific discourse, which discourse properties trigger the use of metaphorical language, rather than using literal alternatives? For example, what drives people to say "grasp the meaning" rather than "understand the meaning" within a specific context? Many NLP approaches to metaphorical language rely on cognitive and (psycho-)linguistic insights and have successfully defined models of discourse coherence, abstractness and affect. In this work, we build five simple models relying on established cognitive and linguistic properties -- frequency, abstractness, affect, discourse coherence and contextualized word representations -- to predict the use of a metaphorical vs. synonymous literal expression in context. By comparing the models' outputs to human judgments, our study indicates that our selected properties are not sufficient to systematically explain metaphorical vs. literal language choices.