Theory of Mind and Self-Attributions of Mentality are Dissociable in LLMs
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918418860474368 |
|---|---|
| author | Kim, Junsol Street, Winnie Rocca, Roberta Korngiebel, Daine M. Waytz, Adam Evans, James Keeling, Geoff |
| author_facet | Kim, Junsol Street, Winnie Rocca, Roberta Korngiebel, Daine M. Waytz, Adam Evans, James Keeling, Geoff |
| contents | Safety fine-tuning in Large Language Models (LLMs) seeks to suppress potentially harmful forms of mind-attribution such as models asserting their own consciousness or claiming to experience emotions. We investigate whether suppressing mind-attribution tendencies degrades intimately related socio-cognitive abilities such as Theory of Mind (ToM). Through safety ablation and mechanistic analyses of representational similarity, we demonstrate that LLM attributions of mind to themselves and to technological artefacts are behaviorally and mechanistically dissociable from ToM capabilities. Nevertheless, safety fine-tuned models under-attribute mind to non-human animals relative to human baselines and are less likely to exhibit spiritual belief, suppressing widely shared perspectives regarding the distribution and nature of non-human minds. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_28925 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Theory of Mind and Self-Attributions of Mentality are Dissociable in LLMs Kim, Junsol Street, Winnie Rocca, Roberta Korngiebel, Daine M. Waytz, Adam Evans, James Keeling, Geoff Computation and Language Artificial Intelligence Safety fine-tuning in Large Language Models (LLMs) seeks to suppress potentially harmful forms of mind-attribution such as models asserting their own consciousness or claiming to experience emotions. We investigate whether suppressing mind-attribution tendencies degrades intimately related socio-cognitive abilities such as Theory of Mind (ToM). Through safety ablation and mechanistic analyses of representational similarity, we demonstrate that LLM attributions of mind to themselves and to technological artefacts are behaviorally and mechanistically dissociable from ToM capabilities. Nevertheless, safety fine-tuned models under-attribute mind to non-human animals relative to human baselines and are less likely to exhibit spiritual belief, suppressing widely shared perspectives regarding the distribution and nature of non-human minds. |
| title | Theory of Mind and Self-Attributions of Mentality are Dissociable in LLMs |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2603.28925 |