Language models emulate certain cognitive profiles: An investigation of how predictability measures interact with individual differences

TL;DR

This study investigates how language models' predictability measures relate to individual differences in reading, revealing that models tend to emulate readers with lower verbal intelligence and that incorporating cognitive scores improves prediction accuracy.

Contribution

The paper introduces a method to incorporate individual cognitive capacities into language model predictions of reading times, highlighting biases in models' emulation of different cognitive profiles.

Findings

Incorporating cognitive scores improves prediction accuracy.

Models tend to emulate readers with lower verbal intelligence.

High verbal intelligence correlates with lower sensitivity to predictability effects.

Abstract

To date, most investigations on surprisal and entropy effects in reading have been conducted on the group level, disregarding individual differences. In this work, we revisit the predictive power of surprisal and entropy measures estimated from a range of language models (LMs) on data of human reading times as a measure of processing effort by incorporating information of language users' cognitive capacities. To do so, we assess the predictive power of surprisal and entropy estimated from generative LMs on reading data obtained from individuals who also completed a wide range of psychometric tests. Specifically, we investigate if modulating surprisal and entropy relative to cognitive scores increases prediction accuracy of reading times, and we examine whether LMs exhibit systematic biases in the prediction of reading times for cognitively high- or low-performing groups, revealing what type of psycholinguistic subject a given LM emulates. Our study finds that in most cases, incorporating cognitive capacities increases predictive power of surprisal and entropy on reading times, and that generally, high performance in the psychometric tests is associated with lower sensitivity to predictability effects. Finally, our results suggest that the analyzed LMs emulate readers with lower verbal intelligence, suggesting that for a given target group (i.e., individuals with high verbal intelligence), these LMs provide less accurate predictability estimates.