MindShift: Analyzing Language Models' Reactions to Psychological Prompts
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912772407689216 |
|---|---|
| author | Vasiliuk, Anton Abdullaeva, Irina Druzhinina, Polina Razzhigaev, Anton Kuznetsov, Andrey |
| author_facet | Vasiliuk, Anton Abdullaeva, Irina Druzhinina, Polina Razzhigaev, Anton Kuznetsov, Andrey |
| contents | Large language models (LLMs) hold the potential to absorb and reflect personality traits and attitudes specified by users. In our study, we investigated this potential using robust psychometric measures. We adapted the most studied test in psychological literature, namely Minnesota Multiphasic Personality Inventory (MMPI) and examined LLMs' behavior to identify traits. To asses the sensitivity of LLMs' prompts and psychological biases we created personality-oriented prompts, crafting a detailed set of personas that vary in trait intensity. This enables us to measure how well LLMs follow these roles. Our study introduces MindShift, a benchmark for evaluating LLMs' psychological adaptability. The results highlight a consistent improvement in LLMs' role perception, attributed to advancements in training datasets and alignment techniques. Additionally, we observe significant differences in responses to psychometric assessments across different model types and families, suggesting variability in their ability to emulate human-like personality traits. MindShift prompts and code for LLM evaluation will be publicly available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_09149 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MindShift: Analyzing Language Models' Reactions to Psychological Prompts Vasiliuk, Anton Abdullaeva, Irina Druzhinina, Polina Razzhigaev, Anton Kuznetsov, Andrey Computation and Language Artificial Intelligence Large language models (LLMs) hold the potential to absorb and reflect personality traits and attitudes specified by users. In our study, we investigated this potential using robust psychometric measures. We adapted the most studied test in psychological literature, namely Minnesota Multiphasic Personality Inventory (MMPI) and examined LLMs' behavior to identify traits. To asses the sensitivity of LLMs' prompts and psychological biases we created personality-oriented prompts, crafting a detailed set of personas that vary in trait intensity. This enables us to measure how well LLMs follow these roles. Our study introduces MindShift, a benchmark for evaluating LLMs' psychological adaptability. The results highlight a consistent improvement in LLMs' role perception, attributed to advancements in training datasets and alignment techniques. Additionally, we observe significant differences in responses to psychometric assessments across different model types and families, suggesting variability in their ability to emulate human-like personality traits. MindShift prompts and code for LLM evaluation will be publicly available. |
| title | MindShift: Analyzing Language Models' Reactions to Psychological Prompts |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2512.09149 |