Exploring a Gamified Personality Assessment Method through Interaction with Multi-Personality LLM Agents

Abstract: The execution of effective and imperceptible personality assessments is receiving increasing attention in psychology and human-computer interaction fields. This study explores an interactive approach for personality assessment, focusing on the multiplicity of personality representation. We propose a framework of gamified personality assessment through multi-personality representations (Multi-PR GPA). The framework leverages LLMs to empower virtual agents with diverse personalities. These agents elicit multifaceted human personality representations through engaging in interactive games. Drawing upon the multi-type textual data generated throughout the interaction, it achieves two ways of personality assessments (i.e., Direct Assessment and Que-based Assessment) and provides interpretable insights. Grounded in the classic Big Five theory, we implemented a prototype system and conducted a user study to assess the efficacy of Multi-PR GPA. The results underscore the effectiveness of our approach in personality assessment and demonstrate that it achieves superior performance when considering the multiplicity of personality representation.