The Personality Trap: How LLMs Embed Bias When Generating Human-Like Personas
Authors:
- Jacopo AmideiJAJacopo Amidei
- Gregorio FerreiraGFGregorio Ferreira
- Mario Muñoz SerranoMSMario Muñoz Serrano
- Rubén NietoRNRubén Nieto
- Andreas KaltenbrunnerAKAndreas Kaltenbrunner
Preprint
Abstract
This paper examines biases in large language models (LLMs) when generating synthetic populations from responses to personality questionnaires. Using five LLMs, we first assess the representativeness and potential biases in the sociodemographic attributes of the generated personas, as well as their alignment with the intended personality traits. While LLMs successfully reproduce known correlations between personality and sociodemographic variables, all models exhibit pronounced WEIRD (western, educated, industrialized, rich and democratic) biases, favoring young, educated, white, heterosexual, Western individuals with centrist or progressive political views and secular or Christian beliefs. In a second analysis, we manipulate input traits to maximize Neuroticism and Psychoticism scores. Notably, when Psychoticism is maximized, several models produce an overrepresentation of non-binary and LGBTQ+ identities, raising concerns about stereotyping and the potential pathologization of marginalized groups. Our findings highlight both the potential and the risks of using LLMs to generate psychologically grounded synthetic populations.Keywords
All Related Versions
This publication has 36 references indexed in Scilit:
- Can AI serve as a substitute for human subjects in software engineering research?Automated Software Engineering, 2024
- AI language models cannot replace human research participantsAI & SOCIETY, 2023
- Can AI language models replace human participants?Trends in Cognitive Sciences, 2023
- The association between personality traits and third-party moral judgment: A preregistered studyActa Psychologica, 2021
- Screening for Dark PersonalitiesEuropean Journal of Psychological Assessment, 2021
- Big-Five personality and political orientation: Results from four panel studies with representative German samplesJournal of Research in Personality, 2019
- A comparison between three and five factor models of Pakistani personality dataPersonality and Individual Differences, 2001
- The Big Three or the Big Five? A replication studyPersonality and Individual Differences, 2000
- Gender differences on the EPQ-R and attitudes to workPersonality and Individual Differences, 1998
- The development of an abbreviated form of the revised Eysenck personality questionnaire (EPQR-A): Its use among students in England, Canada, the U.S.A. and AustraliaPersonality and Individual Differences, 1992