Does Conceptual Representation Require Embodiment? Insights From Large Language Models

TL;DR

This study compares human and large language model representations of concepts, revealing that models excel in non-sensorimotor areas but lack embodied sensory and motor understanding, with multimodal input improving their human-likeness.

Contribution

The paper provides empirical evidence on the limitations of language-only models in representing embodied concepts and shows how multimodal learning enhances their human-like understanding.

Findings

Models align with humans in emotion and salience but lag in sensory/motor domains.

GPT-4 outperforms GPT-3.5, especially with visual learning.

Multimodal inputs improve models' representation of embodied concepts.

Abstract

To what extent can language alone give rise to complex concepts, or is embodied experience essential? Recent advancements in large language models (LLMs) offer fresh perspectives on this question. Although LLMs are trained on restricted modalities, they exhibit human-like performance in diverse psychological tasks. Our study compared representations of 4,442 lexical concepts between humans and ChatGPTs (GPT-3.5 and GPT-4) across multiple dimensions, including five key domains: emotion, salience, mental visualization, sensory, and motor experience. We identify two main findings: 1) Both models strongly align with human representations in non-sensorimotor domains but lag in sensory and motor areas, with GPT-4 outperforming GPT-3.5; 2) GPT-4's gains are associated with its additional visual learning, which also appears to benefit related dimensions like haptics and imageability. These results highlight the limitations of language in isolation, and that the integration of diverse modalities of inputs leads to a more human-like conceptual representation.