Stable Personas: Dual-Assessment of Temporal Stability in LLM-Based Human Simulation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916031555960832 |
|---|---|
| author | Gonnermann-Müller, Jana Haase, Jennifer Leins, Nicolas Kosch, Thomas Pokutta, Sebastian |
| author_facet | Gonnermann-Müller, Jana Haase, Jennifer Leins, Nicolas Kosch, Thomas Pokutta, Sebastian |
| contents | Large Language Models (LLMs) acting as artificial agents offer the potential for scalable behavioral research, yet their validity depends on whether LLMs can maintain stable personas across extended conversations. We address this point using a dual-assessment framework measuring both self-reported characteristics and observer-rated persona expression. Across two experiments testing four persona conditions (default, high, moderate, and low ADHD presentations), seven LLMs, and three semantically equivalent persona prompts, we examine between-conversation stability (3,473 conversations) and within-conversation stability (1,370 conversations and 18 turns). Self-reports remain highly stable both between and within conversations. However, observer ratings reveal a tendency for persona expressions to decline during extended conversations. These findings suggest that persona-instructed LLMs produce stable, persona-aligned self-reports, an important prerequisite for behavioral research, while identifying this regression tendency as a boundary condition for multi-agent social simulation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_22812 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Stable Personas: Dual-Assessment of Temporal Stability in LLM-Based Human Simulation Gonnermann-Müller, Jana Haase, Jennifer Leins, Nicolas Kosch, Thomas Pokutta, Sebastian Human-Computer Interaction Large Language Models (LLMs) acting as artificial agents offer the potential for scalable behavioral research, yet their validity depends on whether LLMs can maintain stable personas across extended conversations. We address this point using a dual-assessment framework measuring both self-reported characteristics and observer-rated persona expression. Across two experiments testing four persona conditions (default, high, moderate, and low ADHD presentations), seven LLMs, and three semantically equivalent persona prompts, we examine between-conversation stability (3,473 conversations) and within-conversation stability (1,370 conversations and 18 turns). Self-reports remain highly stable both between and within conversations. However, observer ratings reveal a tendency for persona expressions to decline during extended conversations. These findings suggest that persona-instructed LLMs produce stable, persona-aligned self-reports, an important prerequisite for behavioral research, while identifying this regression tendency as a boundary condition for multi-agent social simulation. |
| title | Stable Personas: Dual-Assessment of Temporal Stability in LLM-Based Human Simulation |
| topic | Human-Computer Interaction |
| url | https://arxiv.org/abs/2601.22812 |