Towards a Brain-Computer Interface (BCI) for Improving Phonological Processing in Developmental Dyslexia: An Exploratory Study

Abstract

Brain-computer interfaces (BCIs) have immense potential regarding the provision of therapies for disorders of development, but to date have typically been created for non-linguistic disorders such as ADHD (attention deficit hyperactivity disorder). Here we present a BCI designed to improve linguistic phonological processing during natural speech listening, aimed at participants with developmental dyslexia. Phonological "deficits" are considered a core feature of dyslexia across languages and are present from infancy in those later diagnosed with dyslexia. Here a non-invasive EEG-BCI relying on auditory inputs and visual feedback was developed to optimise brain patterns related to phonological development in infants and children. The BCI focused on neural patterns identified using Temporal Sampling theory, which has found the theta/delta ratio during natural speech listening to be a key neural associate of impaired phonology. As a first step, the BCI was tested with adults. Adults with and without dyslexia played the BCI for 16 sessions, and received pre- and post-testing regarding phonological awareness and single word and nonword reading skills. Learning was assessed both online, as in most BCI studies, and offline, enabling removal of a greater number of potential EEG artifacts. Significant associations between offline BCI scores (a measure of BCI learning) and improvements in nonword reading were found for all participants who learned the BCI, both dyslexic and control. BCI learning also showed significant associations with improvements in real word reading and amplitude rise time discrimination. The data are interpreted with respect to Temporal Sampling theory.