Towards Classifying Cognitive Performance by Sensing Electrodermal Activity in Children With Specific Learning Disorders

When children suffer from cognitive disorders, school performance and social environment are affected. Measuring changes in cognitive progress is essential for assessing the clinical follow-up of the patient's cognitive abilities. This process is considered as a challenge in ambulatory settings, where follow-ups should be non-invasive and continuous. Psychophysiological measures are an objective and unobtrusive evaluation alternative for recognizing cognitive changes. This paper aims to validate the relationship between cognition and the changes in physiological signals of children suffering from Specific Learning Disorders (SLD). This validation was carried out in an eHealth rehabilitation context (with the HapHop-Physio game). Electrodermal activity (EDA) signals were collected, processed, and analyzed through a machine learning approach. Obtained results were: a dataset built from wearable physiological data and a supervised classification model. The classification model can identify the children's cognitive performance (class) from the features of the tonic component of the EDA signal (attributes) with an accuracy of 79.95%. The presented results evidence that psychophysiological measures could allow for a highly objective follow-up for patients. They can also lead to creating a basis for further improvement of rehabilitation environments and developing neurofeedback applications.