Children’s imitation is not random, but depends on the context of the demonstration and imitation opportunity. For example, children are more likely to copy acts modeled by multiple people versus a single individual. In this study, I investigate the mechanisms underlying this phenomenon by manipulating the number of demonstrators and mode of presentation for a goal-directed task. Children saw either one or two adults demonstrate unnecessary target acts while opening boxes to retrieve toys, and demonstrations were presented either live or on video. Children imitated the target acts at equal rates across conditions. This may reflect children’s heightened attention to reproducing the salient goal (i.e., opening the box to retrieve a toy) as opposed to copying the acts used to achieve the goal. Future studies should manipulate children’s prior experiences, goal salience, and the majority influence to determine the relative importance of each of these factors in guiding social learning.