Vibrotactile stimuli are defined in terms of their amplitude, frequency, waveform and temporal profile all of which have been varied to create tactons. A number of approaches have been adopted to design tactons including multidimensional scaling, iterative empirical methods and using perceptual processing models. The objective of the present set of experiments was to create sets of tactons based on the properties of the dimensions of vibrotactile stimuli. An absolute identification paradigm was used in which each of nine tactons was presented eight times using a tactor mounted on either the index finger or forearm. It was found that tactons created by varying the frequency, amplitude and temporal profile of the vibrotactile stimuli were correctly identified on 73-83% of the trials, with a mean information transfer of 2.41 bits. The latter metric indicates that for these sets of nine tactons between five and six could be reliably identified. The vibrotactile stimuli delivered in the experiments were identified as consistently on the forearm as the hand and the IT values were similar at the two locations. This suggests that sites other than the hand can be used effectively in tactile communication systems and that it is channel capacity that ultimately determines performance on this type of task.
user-centered design, Haptic I/O, human information processing, input devices and strategies
Lynette Jones, "Evaluating Vibrotactile Dimensions for the Design of Tactons", IEEE Transactions on Haptics, vol. , no. , pp. 0, 5555, doi:10.1109/TOH.2013.2296051