Date of Award

6-2022

Document Type

Open Access

Degree Name

Bachelor of Science

Department

Neuroscience

First Advisor

Stephen Romero

Abstract

A brain-computer interface (BCI) is a novel technology that creates direct assistive communication between the brain and a computer. While numerous electroencephalogram (EEG) based BCI-speller applications have been used for communication by adults with physical disabilities; few BCI studies have included children, and none using BCI spellers. A pilot study of a developmentally-appropriate EEG-based speller-storybook interface that relied on steady-state visual evoked potentials (SSVEPs) by two pediatric users with quadriplegic cerebral palsy showed limited speller reliability (E. Floreani, personal communication, September 30, 2021). In the pilot study, the alphabet was parsed between three boxes, each flashing at a different rate (6Hz, 7.5Hz, 10Hz). The users attended to the box containing the required letter, and the BCI interpreted the resulting fluctuations in the EEG to make the selection. The present study sought to improve BCI speller-storybook reliability by improving stimulus timing and by adding auditory feedback. Speller performance was directly correlated with stimuli reliability but there was no significant difference in the average selection time or accuracy for the auditory-visual versus visual conditions. Nevertheless , auditory feedback may still yield an important addition for impaired participants. The results also suggest the speller is more reliable since participants could complete all the trials. Future work will involve testing the auditory-visual feedback condition for impaired participants. An updated speller-storybook interface with improved reliability still may provide a new educational tool to acquire literacy skills for pediatric users with complex communication disorders.

Download

Share

COinS
 

Rights Statement

In Copyright - Educational Use Permitted.