A Novel Dual And Triple RSVP Paradigm For P300 Speller

AbstractObjectiveA speller system enables disabled people, specifically those with spinal cord injuries, to visually select and spell characters. A problem of primary speller systems is that they are gaze shift dependent. To overcome this problem, a single RSVP paradigm was introduced in which characters are displayed one by one at the center of a screen. In this paper, two new protocols named Dual and Triple RSVP paradigms are introduced and their results are compared against the single paradigm.MethodsIn the Dual and Triple paradigms, two and three characters are displayed at the center of the screen simultaneously, therefore holding the advantage of displaying the target character twice and three times respectively, compared to the one-time appearance in the single paradigm. Subsequently, by reducing the number of repetitions in the Dual and Triple paradigms, it is expected that ITR decreases. To compare the results of these three paradigms, three subjects participated in experiments using all three paradigms.ResultsThe offline results demonstrate an average character detection accuracy of 97% for the single and double protocols, and 80% for the Triple paradigm. In addition, average ITR is calculated to be 5.45, 7.62 and 7.90 bit/min for the single, Dual and Triple paradigms respectively. Results demonstrate an equally good character detection accuracy for the single and Dual paradigms, and a significant increase in ITR in the Dual paradigm compared to the single. The Triple RSVP paradigm demonstrates an almost equal ITR to that of the Dual paradigm, while decreasing character detection accuracy significantly.ConclusionsResults demonstrate that the Dual RSVP paradigm can be recognized as the most suitable approach, by providing the best balance between ITR and character detection accuracy.SignificanceThis research demonstrates the improved performance of a newly proposed speller system (the Dual RSVP paradigm). By replacing existing methods with this new approach, the performance of speller matrices will be enhanced, and in addition the gaze dependency issue that caused limitations for users suffering from unimpaired oculomotor control will be overcome.