Designing LLM-Agents with Personalities: A Psychometric Approach

This research introduces a novel methodology for assigning quantifiable, controllable andpsychometrically validated personalities to Large Language Models-Based Agents (Agents)using the Big Five personality framework. It seeks to overcome the constraints of humansubject studies, proposing Agents as an accessible tool for social science inquiry. Through aseries of four studies, this research demonstrates the feasibility of assigningpsychometrically valid personality traits to Agents, enabling them to replicate complexhuman-like behaviors. The first study establishes an understanding of personalityconstructs and personality tests within the semantic space of an LLM. Two subsequentstudies—using empirical and simulated data—illustrate the process of creating Agents andvalidate the results by showing strong correspondence between human and Agent answersto personality tests. The final study further corroborates this correspondence by usingAgents to replicate known human correlations between personality traits anddecision-making behaviors in scenarios involving risk-taking and ethical dilemmas, therebyvalidating the effectiveness of the psychometric approach to design Agents and itsapplicability to social and behavioral research.