Large language models can replicate cross-cultural differences in personality

TL;DR

This study demonstrates that GPT-4 can simulate cross-cultural personality differences between the US and South Korea, showing potential for aiding cross-cultural research despite some biases and limitations.

Contribution

It is the first large-scale experiment showing GPT-4's ability to replicate cross-cultural personality differences using the Big Five model.

Findings

GPT-4 replicated cross-cultural differences in Big Five traits.

Mean ratings showed upward bias and lower variation than human data.

GPT-4's simulated data had lower structural validity.

Abstract

We use a large-scale experiment (N=8000) to determine whether GPT-4 can replicate cross-cultural differences in the Big Five, measured using the Ten-Item Personality Inventory. We used the US and South Korea as the cultural pair, given that prior research suggests substantial personality differences between people from these two countries. We manipulated the target of the simulation (US vs. Korean), the language of the inventory (English vs. Korean), and the language model (GPT-4 vs. GPT-3.5). Our results show that GPT-4 replicated the cross-cultural differences for each factor. However, mean ratings had an upward bias and exhibited lower variation than in the human samples, as well as lower structural validity. We provide preliminary evidence that LLMs can aid cross-cultural researchers and practitioners.