Evaluating Personality Traits in Large Language Models: Insights from Psychological Questionnaires

TL;DR

This study applies psychological questionnaires to large language models to assess their personality traits, revealing that LLMs exhibit diverse and distinct personality profiles across five core dimensions.

Contribution

It introduces a novel approach of using established psychological tools to evaluate and profile the personality traits of large language models.

Findings

LLMs show unique dominant personality traits.

Personality profiles vary significantly even within the same model family.

LLMs exhibit diverse characteristics across the Big Five dimensions.

Abstract

Psychological assessment tools have long helped humans understand behavioural patterns. While Large Language Models (LLMs) can generate content comparable to that of humans, we explore whether they exhibit personality traits. To this end, this work applies psychological tools to LLMs in diverse scenarios to generate personality profiles. Using established trait-based questionnaires such as the Big Five Inventory and by addressing the possibility of training data contamination, we examine the dimensional variability and dominance of LLMs across five core personality dimensions: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. Our findings reveal that LLMs exhibit unique dominant traits, varying characteristics, and distinct personality profiles even within the same family of models.