Meeting Abstract
Social networks are formed through the actions of individuals and the structure that emerges is central to processes occurring at several biological levels of organization. How individual actions modify social networks, however, remains an important and open question. In our study system, brown-headed cowbirds (Molothrus ater), females are known to influence male courtship behavior. The specific mechanisms behind this phenomenon have been difficult to discover, however, because interactions (e.g. body motions and vocalizations) between multiple pairs of individuals occur simultaneously and are difficult to classify without quantitative measurements. We used an array of eight cameras and 24 microphones followed by a computer vision pipeline to continuously record the position, posture, and vocalizations of cowbirds over the entire breeding season. With these data, we investigate how moment-to-moment interactions drive changes in the social network over months. Our data also provide a difficult setting containing multiple camera views, background motion, shadows, and changes in lighting, in which to test animal tracking, pose estimation, and re-identification algorithms.
