Recent experimental psychological research on visual perception, auditory perception, and cross-modal perception has shed light on how these processes differ, and how the relations between visual and auditory stimuli shade our understanding of the events perceived. This work offers a possible way into considering the question of how music and dance "go together" or not, and particularly may shed light on the unusual twentieth-century human behavior of NOT having music and dance "go together." Our paper presents relevant research in perception, examines factors contributing to the separation of perceptual modalities that has often appeared in twentieth-century dance, and discusses the separation in terms of the specific behaviors of dancing and looking at dance.Keywords: perception; vision; audition; music; dance We come to this paper with a practical interest in what makes music and dance go together, or not, in that one of us choreographs and one of us composes, and we share a desire to make computer-interactive pieces. In such pieces the connection between movement and sound must be examined explicitly, and this has led us to explore the literature from experimental psychology relating sound to movement and attempt to connect it to the making and viewing of dances.We start by trying to distinguish between cases where dance and music are seen to "go together" and those where they are not. The latter appear to be unusual in human history. Indeed, one standard explanation for the evolution of music and dance hypothesizes that they contribute to group cohesion (McNeill, 1995;Cross, 1999;Brown, 2001;Hagen & Bryant, 2003), and perhaps the traditional cohesion of music and dance not only reflects