Large Language Models for Psycholinguistic Plausibility Pretesting

TL;DR

This paper explores the use of large language models, especially GPT-4, for pretesting linguistic materials in psycholinguistics by assessing their plausibility judgments compared to human evaluations.

Contribution

It demonstrates that GPT-4's plausibility judgments strongly correlate with human judgments and can replace humans for coarse-grained pretesting but not for fine-grained assessments.

Findings

GPT-4's plausibility judgments align closely with human judgments.

Other LMs perform well on common syntactic structures.

GPT-4 is effective for coarse but not fine-grained pretesting.

Abstract

In psycholinguistics, the creation of controlled materials is crucial to ensure that research outcomes are solely attributed to the intended manipulations and not influenced by extraneous factors. To achieve this, psycholinguists typically pretest linguistic materials, where a common pretest is to solicit plausibility judgments from human evaluators on specific sentences. In this work, we investigate whether Language Models (LMs) can be used to generate these plausibility judgements. We investigate a wide range of LMs across multiple linguistic structures and evaluate whether their plausibility judgements correlate with human judgements. We find that GPT-4 plausibility judgements highly correlate with human judgements across the structures we examine, whereas other LMs correlate well with humans on commonly used syntactic structures. We then test whether this correlation implies that LMs can be used instead of humans for pretesting. We find that when coarse-grained plausibility judgements are needed, this works well, but when fine-grained judgements are necessary, even GPT-4 does not provide satisfactory discriminative power.