Human and macaque pairs employ different coordination strategies in a transparent decision game

Abstract

Interactions of group-living primates with conspecifics range from cooperation to competition.Game theory allows testing the strategies that underlie such interactions, but in classical theory, agents act simultaneously or sequentially. Many real-world decisions, however, are made while directly observing partner's actions. To investigate social decision-making under conditions of face-to-face action visibility, we developed a setup where two agents observe each other and reach to targets on a shared transparent display, enabling naturalistic interactions we call "transparent games". Here we compared human and macaque pairs in the transparent version of the coordination game "Bach or Stravinsky", which rewards coordination but entails the conflict about which of the two individually-preferred coordinated options to choose. Most human pairs developed coordinated behavior, and 53% adopted dynamic coordination via turntaking to equalize the payoffs. All macaque pairs also converged on coordination, but in a simpler, static way: persistently selecting one of the two coordinated options or one of the two display sides. Two animals that underwent training with a turn-taking human confederate learned to coordinate dynamically. When tested as a pair, they mostly converged on the faster monkey's preferred option, and a dynamic coordination emerged as animals spontaneously took turns in leading to their respective preferred option and following to the other's. The observed choices were captured by modeling a probability to see the other's action before own movement. Importantly, such competitive turn-taking was unlike the benevolent turn-taking in humans, who equally often initiated switches to and from their preferred option. Our findings demonstrate that dynamic coordination is not restricted to humansalthough it serves a selfish motivation in macaquesand emphasize the importance of action visibility in the emergence and maintenance of coordination.

Learning of object-in-context sequences in freely-moving macaques

“…Consequently, the animal shows a regular outward positioning to react to screens, limiting the visibility of face and limbs during the task epochs. As such, this task structure is not optimized for assessing reach and eye movements, therefore other design strategies could be added for tasks designed to study reaching and facial movements (Womelsdorf et al, 2021; Ulbrich and Gail, 2021; Möller et al, 2020; Hayden et al, 2022) Our use of touches on the screen, however, was effective at ensuring spatiotemporal and physical (vision and reach) points of alignment, each funneling into lower degrees of freedom than the full continuous behaviors would offer. This suggests that for predictable goals, the range of movements and poses has far fewer degrees of freedom and is therefore more tractable to analyze than random foraging and exploration.…”

The cone method: Inferring decision times from single-trial 3D movement trajectories in choice behavior

“…Other paradigms allow subjects to sequentially make multiple decisions within a given trial while continuously moving around the task environment (Diamond et al, 2017;Michalski et al, 2020), effectively abolishing the one-decision-per-trial structure for more naturalistic choice sequences. Also, coordination of action with others, be it in a cooperative or a competitive setting, requires continuous integration of selection and control when the opponents' actions are mutually visible and leads to specific dyadic choice behavior, like leader-follower strategies in transparent games (Möller et al, 2020;Unakafov et al, 2020).…”

The Cone Method: Inferring Decision Times from Single-Trial 3D Movement Trajectories in Choice Behavior

“…Other paradigms allow subjects to sequentially make multiple decisions within a given trial while continuously moving around the task environment (Diamond et al, 2017;Michalski et al, 2020), effectively abolishing the onedecision-per-trial structure for more naturalistic choice sequences. Also, coordination of action with others, be it in a cooperative or a competitive setting, requires continuous integration of selection and control when the opponents' actions are mutually visible and leads to specific dyadic choice behavior, like leader-follower strategies in transparent games (Möller et al, 2020;Unakafov et al, 2020).…”