A functional imaging study explores the neural basis of cooperation in two-person reciprocal exchange. The study involved 12 right-handed subjects who played a trust game with either a human or computer counterpart in a functional MRI scanner. The results showed that seven subjects consistently cooperated with humans, while five did not. Prefrontal cortex activity was more pronounced when subjects played with humans compared to computers. This suggests that prefrontal regions are involved in integrating theory-of-mind processing with cooperative actions. The study hypothesizes that cooperation requires the binding of contingent information to evaluate the mental states of counterparts and commit to reward-motivated choices. The medial prefrontal cortex is proposed as a convergence zone for this process, enabling the integration of game and counterpart entities into mutual-gains events. The trust game involved two decision-makers, with the first player choosing to trust or not, and the second player reciprocating. The study found that cooperative subjects showed significant activation in the medial prefrontal cortex when interacting with humans, while noncooperators did not. An aggregate analysis suggested a common cortical network for cooperation, with activation differences observed in the occipital, parietal, and thalamic regions. The findings support the idea that cooperation requires an active convergence zone in the prefrontal cortex that binds joint attention to mutual gains and inhibits immediate reward gratification. The study concludes that cooperation involves complex cognitive processes that are supported by specific brain regions, particularly the prefrontal cortex. The results highlight the role of prefrontal cortex in integrating theory-of-mind processing with cooperative actions, and suggest that subjects learn game-dependent rules when playing with computers or noncooperative humans.A functional imaging study explores the neural basis of cooperation in two-person reciprocal exchange. The study involved 12 right-handed subjects who played a trust game with either a human or computer counterpart in a functional MRI scanner. The results showed that seven subjects consistently cooperated with humans, while five did not. Prefrontal cortex activity was more pronounced when subjects played with humans compared to computers. This suggests that prefrontal regions are involved in integrating theory-of-mind processing with cooperative actions. The study hypothesizes that cooperation requires the binding of contingent information to evaluate the mental states of counterparts and commit to reward-motivated choices. The medial prefrontal cortex is proposed as a convergence zone for this process, enabling the integration of game and counterpart entities into mutual-gains events. The trust game involved two decision-makers, with the first player choosing to trust or not, and the second player reciprocating. The study found that cooperative subjects showed significant activation in the medial prefrontal cortex when interacting with humans, while noncooperators did not. An aggregate analysis suggested a common cortical network for cooperation, with activation differences observed in the occipital, parietal, and thalamic regions. The findings support the idea that cooperation requires an active convergence zone in the prefrontal cortex that binds joint attention to mutual gains and inhibits immediate reward gratification. The study concludes that cooperation involves complex cognitive processes that are supported by specific brain regions, particularly the prefrontal cortex. The results highlight the role of prefrontal cortex in integrating theory-of-mind processing with cooperative actions, and suggest that subjects learn game-dependent rules when playing with computers or noncooperative humans.