Berends, Sanne (2025) The Evolution of Theory of Mind in a Simulated Population: a Mixed-Motive Setting. Master's Thesis / Essay, Artificial Intelligence.
|
Text
mAI2025BerendsS.pdf Download (6MB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
Toestemming.pdf Restricted to Registered users only Download (172kB) |
Abstract
The mixed-motive interaction hypothesis suggests that mixed-motive social situations like negotiations drove the evolution of (higher-order) theory of mind in humans. Simulations of pairwise interactions have shown that agents with (higher-order) theory of mind perform better than those without it, supporting this hypothesis. However, pairwise interactions alone do not capture the complexity of social group settings. Whether theory of mind also provides advantages on a population level remains unexplored. We constructed a simulated evolutionary process where agents with various orders of theory of mind (zero, one, or two) trade resources to survive. The negotiations required for successful trade form the mixed-motive aspect of the environment. Two experiments were conducted: one where agents did not distinguish between trading partners and another where agents remembered partner-specific information. In the latter, agents rejected negotiations with incompatible partners based on prior interactions. In both experiments, the survival of agents per type was tracked. The results showed that while theory of mind benefited agents in pairwise negotiations, agents without it outlasted others. Negotiations involving theory of mind took longer, leading to resource thresholds often being unmet. These findings shed new light on the possible evolutionary advantages of higher-order theory of mind in mixed-motive settings.
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Supervisor name: | Weerd, H.A. de and Verbrugge, L.C. |
| Degree programme: | Artificial Intelligence |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 28 Jan 2025 09:36 |
| Last Modified: | 28 Jan 2025 09:36 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/34577 |
Actions (login required)
 |
View Item |