This paper examines the relationship between intelligence and cooperation in two 20-period repeated public goods games. Study 1, set in an equal endowment environment, finds that higher intelligence individuals contribute less to the public good and hold lower expectations about others’ contributions. Study 2 tests for the robustness of this relationship in a public good game with unequal endowments and finds that this pattern disappears. A closer analysis reveals that high intelligence rich individuals sustain more stable contributions over time than their lower intelligence counterparts. Consistently, high intelligence individuals —regardless of their own endowment— hold more stable beliefs about the rich individuals’ contributions over time. These findings underscore the complex interplay between cognitive ability, resource distribution, and belief formation in shaping cooperative behavior in the long run.