Visually-impaired individuals and sighted individuals who are conducting tasks in divided-attention situations, both benefit from using sound to display information typically communicated visually. Auditory displays of statistical graphs are one such tool and can be effective in these situations. However, it is not obvious how these graphs should be designed. In a series of experiments in which information was conveyed by sound, subjects were divided into two groups: those hearing graphs using integral (pitch and loudness) and separable (pitch and timing) dimensions of sound. The results showed that pitch alone produced the worst performance and timing the best. However, designs using pitch and loudness redundantly provided good results as well.
This paper costs $33 for non-members and is free for AES members and E-Library subscribers.