Investigating the feasibility of bci-based social media interaction

Abstract

Brain-computer Interfaces (BCIs) use the inherent brain activity of subjects in order to communicate with external devices. BCIs have been utilised by disabled users to navigate and interact with their environments using a host of application platforms such as mouse cursor controllers, virtual keyboards and mobile wheelchairs. These platform highlight the wide assistive potential of BCIs. These applications however are restricted to the subjects local environment and do not readily present an opportunity to communicate over long distances or allow messaging to multiple recipients. Social media platforms allow long distance communication over the internet and messaging to multiple recipients which can be leveraged for BCI-based communication. However, the integration of BCIs with social media platforms for communication is limited. This paper investigates the integration of a BCI with Twitter for the communication of public posts and private messages. An online P300-based BCI was developed which allowed for communication to Twitter. Three subjects participated in a programme of experimentation using the developed BCI. The BCI yielded an average classification accuracy of 86% for the identification of user messages. All user messages however were successfully communication to Twitter. This paper therefore highlighted social media as a viable communication platform which can further improve the assistive value of BCIs. Possible areas of future study include the investigation of other classification methodologies and the integration of the BCI with further social media platforms such as Facebook and LinkedIN.