Conventional DMI design will often follow a unidirectional paradigm; sensors capture musical gestures and this data is then mapped to synthesis parameters. The reality of interaction with musical instruments is, of course, less straightforward. In reality, there is a coupling between the musician and the instrument that is bidirectional – the musician is affected by the instrument and its physicality as well as the musician affecting the instrument through gesture.

This PhD project aims to explore this idea of bidirectionality – where the boundary between the musician and the instrument in the tactile domain does not enforce a certain level of unidirectionality. Through the use of high bandwidth inputs and outputs (using audio signals rather than the frequently used approach of discretisation to MIDI notes, for example), a certain richness may be obtained from musical interactions.

Matthew has investigated this idea, and the effect on instrument designs, through a user study involving experienced DMI designers using the HaptiCoupler system. Published in CHI (Conference on Human Factors in Computing Systems), this study explored how the bidirectional audio-rate interaction has the potential to 'remutualise' audio and vibrotactile haptics, as well as changing interactions techniques and musical gestures.