Tufted capuchin monkeys (Sapajus spp.) are the only Neotropical Primates that regularly use tools in the wild, but only one population of bearded capuchin monkeys (Sapajus libidinosus) is known to habitually use sticks as probes. In this population, males are typically the only sex to use stick tools, something unexpected, since there are no obvious physical constraints, and females do use stone tools in the wild and sticks in experimental conditions. We investigated the development of probe tool use in eight infants to clarify whether social influences on learning varied between the sexes, as tool use learning by capuchin monkeys is a socially biased process. We found that in the first 10 months of age, females manipulate sticks as much as males, but after 10-12 months of age, males begin to manipulate them at higher frequencies. We examined if social connections-as opportunities for social learning-could explain this difference and verified that, on close distance social networks, infant males and females have similar connections with older males. However, males observe probe tool use events more often than females when close to such events. The higher frequency of manipulation of sticks, as well as the higher rates of probe tool use observation, appear to be the key to understand why only males are probe tool users in this population. Since there are only male potential models of probe use, a sex motivational bias could explain the sex difference in observation; a bias in observation could explain the differences in manipulation-and manipulation rates would certainly influence the chances of individual, trial-and-error learning (a case of "local/stimulus enhancement").