Social interactions are unique from many other behaviors in that they require individuals to anticipate each other's intentions and how they may consequently respond to our own decisions. Primates are highly social animals that share many behaviors in common with humans and can serve as a potential model for understanding interactive social behavior. Here, we trained multiple pairs of Rhesus macaques to perform an iterated prisoner's-dilemma game that allowed them to make joint decisions which could mutually affect their shared outcome. Using this platform, we studied how neurons in the dorsal anterior cingulate cortex (dACC), an area through to be involved in social behavior, encode interactive decisions and how neural modulation of the dACC influence the animal's decisions. We observe that while many dACC neurons encoded the monkeys own decision based on prior selections, a predominant and largely distinct group of neurons predicted their opponent's decisions when they were yet unknown. At the population-level, mixed population predictions of the other were remarkably near-optimal when compared to behavioral decoders, while providing relatively little information about the primates own behavior or reward. Disrupting cingulate activity negatively biased mutually beneficial interactions between the monkeys whereas administration of oxytocin enhanced them. In the talk, I will discuss the relevance of these findings to our potential understanding of social behavior, future directions in studying interactive behavior in humans and future strategies for treating individuals with social behavioral disorders.