The Representational Alignment between Humans and Language Models is implicitly driven by a Concreteness Effect

TL;DR

This study reveals that the alignment between human and language model semantic representations is primarily driven by the concreteness of words, highlighting the importance of concreteness in semantic processing.

Contribution

It demonstrates that human and language model representations are aligned along the concreteness dimension, a novel insight into semantic similarity and cognitive modeling.

Findings

Human and model representations are significantly aligned.

Alignment is primarily driven by concreteness, not other word features.

Both spaces are implicitly aligned to explicit concreteness ratings.

Abstract

The nouns of our language refer to either concrete entities (like a table) or abstract concepts (like justice or love), and cognitive psychology has established that concreteness influences how words are processed. Accordingly, understanding how concreteness is represented in our mind and brain is a central question in psychology, neuroscience, and computational linguistics. While the advent of powerful language models has allowed for quantitative inquiries into the nature of semantic representations, it remains largely underexplored how they represent concreteness. Here, we used behavioral judgments to estimate semantic distances implicitly used by humans, for a set of carefully selected abstract and concrete nouns. Using Representational Similarity Analysis, we find that the implicit representational space of participants and the semantic representations of language models are significantly aligned. We also find that both representational spaces are implicitly aligned to an explicit representation of concreteness, which was obtained from our participants using an additional concreteness rating task. Importantly, using ablation experiments, we demonstrate that the human-to-model alignment is substantially driven by concreteness, but not by other important word characteristics established in psycholinguistics. These results indicate that humans and language models converge on the concreteness dimension, but not on other dimensions.