Living in large groups is, for many species, an adaptive solution to survival and reproductive issues. It followed that in primates, and even more so humans, communication evolved into a complex signalling system that includes language, nonverbal vocalisations such as laughter, and facial expressions. A series of studies were designed to address the function of smiling and laughter through an analysis of context and consequences. First, naturalistic observations were conducted in areas where people could be watched interacting in stable social groups. Focal sampling of men and women allowed the recording of smiling and laughter frequencies, as well as other interpersonal aspects such as talking and listening time, and body contacts. Smiles were classified in two categories: spontaneous and forced. A test based on predictions derived from three hypotheses (mate choice, social competition, and cooperation) revealed that spontaneous smiling and laughter are likely to be involved in the formation of cooperative relationships. A closer examination of dyadic interactions revealed that smiling was related to talking and listening time, whereas female's vocalised laughter positively affected the partner's speech output. Finally, smiling and laughter rates increased the probability of observing affiliative body contacts between individuals. A second set of studies investigated the possibility that smiling could (1) advertise attributes relevant to the formation of social relationships, and (2) be a honest signal of altruistic dispositions. The assessment of various traits was examined through people's judgments of neutral and smiling photographs. Results showed that smiling faces were perceived as being significantly more attractive, more generous, healthier, more agreeable, more extroverted, and more open to experiences than their neutral counterparts. Interestingly, men were influenced by smiling in a much larger extent than women, particularly when smiling faces were female's. The rating study also revealed that people who displayed smiles involving an emotional component (Duchenne smiles) received higher scores on extroversion and generosity than people who did not, indicating that people's ratings of sociability and generosity are sensitive to facial movements that are not easy to produce on purpose. A final study investigated the effect of bargaining contexts on smiling and laughter rates between friends. Analysis of videotaped interactions showed that Duchenne smiling and vocalised laughter were displayed at significantly higher rates when people were involved in the sharing of material resources (as opposed to a control interaction). Moreover, data confirmed that Duchenne smiling could be a reliable signal of altruism, as its frequency of occurrence in the bargaining interaction was positively affected by measures of altruism. Finally, results showed that smiling and laughter could advertise personality traits as well as aspects of the relationship between sender and receiver. All in all, the present thesis indicates that smiling and laughter could be used adaptively to develop social alliances, and that this bonding process would entail the reliable advertisement of evolutionarily relevant attributes. The relevance of smiling to a behavioural style based on cooperation and prosocial activities is also discussed.