From Monolingual to Bilingual: Investigating Language Conditioning in Large Language Models for Psycholinguistic Tasks

TL;DR

This study explores how large language models encode psycholinguistic knowledge across languages, revealing that language identity influences their outputs and internal representations, with implications for cross-linguistic cognition modeling.

Contribution

It demonstrates that language prompts affect both behavior and internal representations in LLMs, advancing understanding of multilingual psycholinguistic processing in AI models.

Findings

Models adjust outputs based on language prompts.

Chinese prompts produce stronger valence signals.

Deeper layers encode more decodable psycholinguistic signals.

Abstract

Large Language Models (LLMs) exhibit strong linguistic capabilities, but little is known about how they encode psycholinguistic knowledge across languages. We investigate whether and how LLMs exhibit human-like psycholinguistic responses under different linguistic identities using two tasks: sound symbolism and word valence. We evaluate two models, Llama-3.3-70B-Instruct and Qwen2.5-72B-Instruct, under monolingual and bilingual prompting in English, Dutch, and Chinese. Behaviorally, both models adjust their outputs based on prompted language identity, with Qwen showing greater sensitivity and sharper distinctions between Dutch and Chinese. Probing analysis reveals that psycholinguistic signals become more decodable in deeper layers, with Chinese prompts yielding stronger and more stable valence representations than Dutch. Our results demonstrate that language identity conditions both output behavior and internal representations in LLMs, providing new insights into their application as models of cross-linguistic cognition.