Comparative Evaluation of Facial Paradigms for Devanagari Script-Based P300 Speller: Preliminary Findings

Abstract

Display paradigm design and brain signal decoding are two important stages in design of any brain computer interface (BCI) device including speller system. Conventional grey to white intensification in display paradigm limits the usability of existing Devanagari-script based P300 speller (DSP3S). Few studies suggest that the use of familiar face improves visual evoked potential. Thus, the present study aimed to investigate the efficacy of facial paradigms on further improving the usability of DSP3S. Authors investigated three facial paradigms based on the smiley-face, famous face and family face for DSP3S. The three facial paradigms are evaluated based on their effectiveness, efficiency and satisfaction. Effectiveness was measured using classification performance and amplitude difference waveform. Whereas efficiency and satisfaction were measured using NASA-TLX, VAS-F scale and SUS respectively. The obtained results showed that using the proposed family face intensification, efficiency and satisfaction could be improved. However no significant difference was observed for effectiveness among three display paradigms. In this study, apart from investigating usability of three display paradigm authors also propose brain signal decoding using morphological filtering and extreme gradient boosting (XGBoost). Finally, the study concludes that the design of display paradigms using different intensification pattern affect usability of speller system and should be taken into consideration.

Received: 31 July 2023 | Revised: 20 December 2023 | Accepted: 29 December 2023

Conflicts of Interest

Narendra D. Londhe is an editorial board member for Journal of Data Science and Intelligent Systems, and was not involved in the editorial review or the decision to publish this article. The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The data that support the findings of this study are not publicly available due to privacy concerns. However, anonymous data are available on reasonable request. Requests should be made to Narendra D. Londhe (nlondhe.ele@nitrr.ac.in) and should include a brief description of the intended use of the data.