This paper examines the influence of three accent structures on the reproduction of simple musical tunes by 5-, 7-, 9-, and 11-year-old children and adult pianists. In the metric accent structure, periodically spaced beats are accented; in the melodic accent structure, events after a jump in pitch interval or after a turn in contour are accented; and in the rhythmic grouping accent structure, the first and last events of a rhythmic group are accented. Four themes were created in which all three accent structures coincided. Three variations were derived from each theme in which each accent structure in turn was moved out of synchrony with the other two. In a fourth variation, all three accent structures were out of synchrony with each other. Both the children's and the pianists' reproductions were the most accurate when the accent structures coincided and the least accurate when they conflicted. Intermediate results were obtained when the rhythmic grouping and melodic accent structures were displaced, but no deterioration in performance was observed when the metric accent structure was displaced, demonstrating the relatively unimportant role played by intensity variations in these conditions. These results were obtained only for the melody scores—no deteriorations were found in the rhythm scores. For the pianists, fewer errors were made on the accented notes than on the others. So, the juxtaposition of the three accent structures does indeed influence the perception of musical sequences. When attention is drawn to the same point in time by several different accent structures, the segmentation of the sequence into smaller units is efficient and reproductions are good. However, when attention is drawn to too many points in time, segmentation is less efficient and reproductions are poorer.