The geographical distributions of the tamarins Saguinus mystax, S. imperator and S. lahiatus do not overlap, but each of these species is sympatric with S. fuscicollis, and mixed-species associations are formed. One way to examine the costs and benefits of association is to compare mixed-species tamarin groups and single-species tamarin groups in the field. However, the stability and permanency of Saguinus associations mainly preclude such opportunities and comparisons between single- and mixed-species groups from different sites are confounded by a number of factors. However, predictions derived from field observations can be tested in captivity under controlled experimental conditions. In this paper, we evaluate the usefulness of testing hypotheses in captivity and describe an experimental study of vigilance in which the behaviour of single-species groups and mixed-species groups of S. fuscicollis weddelli and S. I. lahiatus was compared in order to determine the influence of each species on the other. The results demonstrate that the patterns of vigilance are both different and complementary. As such, they provide support for improved detection of predators as being one advantage of forming mixed-species groups. The discussion focuses on how the combined approach of systematic testing of hypotheses in captivity, together with field studies, can be employed to examine the advantages and disadvantages of Saguinus associations.