Gesture Interaction with Spatial Audio Displays: Effects of Target Size and Inter-Target Separation

This paper presents the results of an experiment comparing two spatial audio display segmentation techniques by investigating the relative salience of target width versus distance to target in a gesture based spatial audio selection task. The first technique, MINIMAL, occupies as little of the display area as possible with sounds placed as close to each other as possible. The second technique, MAXIMAL, occupies all the available display area and sounds are placed as far apart as possible and the associated display area assigned to each sound is allowed to grow. Ratios of distance to target to target width were kept constant in both displays to investigate the relative salience of distance to target versus target width in the sound selection task. Participants performed an orientation based pointing task to select an audio display element in the presence of distracting sounds. Results show that the maximal strategy results in faster and more accurate interaction. Target width was found to have significantly more impact on time ratings than distance to target. Time and accuracy ratings indicate that deictic gesture interaction with a spatial audio display is a robust and efficient interaction technique.