Static network structure cannot stabilize cooperation among large language model agents.
Large language models (LLMs) are increasingly used to model human social behavior, with recent research exploring their ability to simulate social dynamics. Here, we test whether LLMs mirror human behavior in social dilemmas, where individual and collective interests conflict. Humans generally coope...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2025-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0320094 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849325348460691456 |
|---|---|
| author | Jin Han Balaraju Battu Ivan Romić Talal Rahwan Petter Holme |
| author_facet | Jin Han Balaraju Battu Ivan Romić Talal Rahwan Petter Holme |
| author_sort | Jin Han |
| collection | DOAJ |
| description | Large language models (LLMs) are increasingly used to model human social behavior, with recent research exploring their ability to simulate social dynamics. Here, we test whether LLMs mirror human behavior in social dilemmas, where individual and collective interests conflict. Humans generally cooperate more than expected in laboratory settings, showing less cooperation in well-mixed populations but more in fixed networks. In contrast, LLMs tend to exhibit greater cooperation in well-mixed settings. This raises a key question: Are LLMs about to emulate human behavior in cooperative dilemmas on networks? In this study, we examine networked interactions where agents repeatedly engage in the Prisoner's Dilemma within both well-mixed and structured network configurations, aiming to identify parallels in cooperative behavior between LLMs and humans. Our findings indicate critical distinctions: while humans tend to cooperate more within structured networks, LLMs display increased cooperation mainly in well-mixed environments, with limited adjustment to networked contexts. Notably, LLM cooperation also varies across model types, illustrating the complexities of replicating human-like social adaptability in artificial agents. These results highlight a crucial gap: LLMs struggle to emulate the nuanced, adaptive social strategies humans deploy in fixed networks. Unlike human participants, LLMs do not alter their cooperative behavior in response to network structures or evolving social contexts, missing the reciprocity norms that humans adaptively employ. This limitation points to a fundamental need in future LLM design-to integrate a deeper comprehension of social norms, enabling more authentic modeling of human-like cooperation and adaptability in networked environments. |
| format | Article |
| id | doaj-art-4e5911ad0c034978b57c1cb65eef2ff8 |
| institution | Kabale University |
| issn | 1932-6203 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj-art-4e5911ad0c034978b57c1cb65eef2ff82025-08-20T03:48:27ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01205e032009410.1371/journal.pone.0320094Static network structure cannot stabilize cooperation among large language model agents.Jin HanBalaraju BattuIvan RomićTalal RahwanPetter HolmeLarge language models (LLMs) are increasingly used to model human social behavior, with recent research exploring their ability to simulate social dynamics. Here, we test whether LLMs mirror human behavior in social dilemmas, where individual and collective interests conflict. Humans generally cooperate more than expected in laboratory settings, showing less cooperation in well-mixed populations but more in fixed networks. In contrast, LLMs tend to exhibit greater cooperation in well-mixed settings. This raises a key question: Are LLMs about to emulate human behavior in cooperative dilemmas on networks? In this study, we examine networked interactions where agents repeatedly engage in the Prisoner's Dilemma within both well-mixed and structured network configurations, aiming to identify parallels in cooperative behavior between LLMs and humans. Our findings indicate critical distinctions: while humans tend to cooperate more within structured networks, LLMs display increased cooperation mainly in well-mixed environments, with limited adjustment to networked contexts. Notably, LLM cooperation also varies across model types, illustrating the complexities of replicating human-like social adaptability in artificial agents. These results highlight a crucial gap: LLMs struggle to emulate the nuanced, adaptive social strategies humans deploy in fixed networks. Unlike human participants, LLMs do not alter their cooperative behavior in response to network structures or evolving social contexts, missing the reciprocity norms that humans adaptively employ. This limitation points to a fundamental need in future LLM design-to integrate a deeper comprehension of social norms, enabling more authentic modeling of human-like cooperation and adaptability in networked environments.https://doi.org/10.1371/journal.pone.0320094 |
| spellingShingle | Jin Han Balaraju Battu Ivan Romić Talal Rahwan Petter Holme Static network structure cannot stabilize cooperation among large language model agents. PLoS ONE |
| title | Static network structure cannot stabilize cooperation among large language model agents. |
| title_full | Static network structure cannot stabilize cooperation among large language model agents. |
| title_fullStr | Static network structure cannot stabilize cooperation among large language model agents. |
| title_full_unstemmed | Static network structure cannot stabilize cooperation among large language model agents. |
| title_short | Static network structure cannot stabilize cooperation among large language model agents. |
| title_sort | static network structure cannot stabilize cooperation among large language model agents |
| url | https://doi.org/10.1371/journal.pone.0320094 |
| work_keys_str_mv | AT jinhan staticnetworkstructurecannotstabilizecooperationamonglargelanguagemodelagents AT balarajubattu staticnetworkstructurecannotstabilizecooperationamonglargelanguagemodelagents AT ivanromic staticnetworkstructurecannotstabilizecooperationamonglargelanguagemodelagents AT talalrahwan staticnetworkstructurecannotstabilizecooperationamonglargelanguagemodelagents AT petterholme staticnetworkstructurecannotstabilizecooperationamonglargelanguagemodelagents |