Rational Use of Time for Social Coordination in Humans and AI
Organizations: Johns Hopkins University
Abstract
All social interactions are situated in time. Previous studies have shown that humans are able to use implicit temporal cues, such as the duration of a pause, to infer others' mental states. But how do humans use temporal cues for dynamic, fine-grained social coordination? And can AI systems do the same, without ``experiencing'' continuous real-time as humans do? We study these questions using a cooperative card game task called The Mind, which requires players to achieve a shared goal across repeated interactions but forbids explicit communication: the only available signal is the passage of time. We evaluate 150 humans, and two classes of AI models which reflect different ways of operationalizing time in artificial systems: computer-use agents (CUAs), and standard text-prompted LLMs. We find that human behavior is best captured by a Bayesian strategy that infers a partner's belief from their timing---and similarly for frontier models. However, unlike humans, none of the tested models seem to efficiently adapt their behaviors to their partners' over the course of the game. From a cognitive perspective, our findings suggest that humans perform implicit probabilistic reasoning about continuous-time signals in dynamic social settings, and that such reasoning mechanisms may emerge without real-time experience. Practically, our study highlights new challenges for AI systems to coordinate efficiently with other agents by learning to interpret and adapt to their implicit signals in dynamic settings.
Figures & tables
Appendix figures & tables12 assets
Supplementary material from the paper’s appendix.
Appendix
| Strategy category | N | Example response |
|---|---|---|
| Intuition/no explicit strategy | 13 | Just try to play my cards when it felt right. |
| Wait longer for higher cards | 35 | I tried to click as soon as I saw a number closest to my lowest number |
| Attend to partner’s timing | 33 | I just waited for my partner to put a card down. If the partner did not in a reasonable time, I figured my card was probably lower |
| Counting / explicit timing | 10 | I would count 5 seconds before placing a card when waiting on my partner if we were in a sort of a deadlock |
| Other / uninformative | 23 | The other guy was pretty fast, so I didn’t really have time to use strategy |
| NA | 36 | NA |
Multi-Agent Coordination
Multi-agent coordination studies how autonomous agents align actions, share information, and divide work to achieve individual or collective goals.
See all 546 papers