Related Experiment Video
Updated: Aug 16, 2026
Infant Auditory Processing and Event-related Brain Oscillations
Published on: July 1, 2015
Predicting training outcomes for developmental dyslexia from EEG data
Giuseppe Di Dona1, Denisa Adina Zamfira2, Francesco De Benedetto3
1School of Psychology, Vita-Salute San Raffaele University, Via Olgettina 58, Milano, MI 20132, Italy; Division of Neuroscience, IRCCS San Raffaele Scientific Institute, Via Olgettina 60, Milano, MI 20132, Italy; Department of Psychology and Cognitive Science, University of Trento, Corso Bettini 31, Rovereto, TN 38068, Italy.
Machine learning applied to resting-state EEG can predict treatment success in adults with developmental dyslexia (DD). This may lead to personalized interventions by identifying neural markers for reading improvement.
Area of Science:
- Neuroscience
- Machine Learning
- Developmental Disorders
Background:
- Developmental dyslexia (DD) affects ~10% of individuals, causing reading difficulties and potential societal barriers.
- Effective interventions are needed, but outcomes vary due to individual differences.
- Predicting training outcomes could enable personalized treatment protocols for DD.
Purpose of the Study:
- To apply machine learning to resting-state EEG data.
- To predict longitudinal training outcomes in adults with DD.
- To identify neural markers associated with treatment responsiveness.
Main Methods:
- Adults with DD were enrolled in a randomized clinical trial with three training groups: visuo-attentional + tACS, visuo-attentional + sham, and phonological + sham.
- Resting-state EEG was recorded.
- Machine learning models were used to predict improvements in reading speed and pseudoword reading.
Main Results:
- Improvement in text reading speed correlated with spectral power in low-beta and individual alpha/beta frequencies (IAF/IBF).
- Improvement in pseudoword reading correlated with IBF.
- EEG markers show potential for predicting treatment responsiveness in DD.
Conclusions:
- Resting-state EEG, analyzed with machine learning, can identify neural markers predictive of treatment outcomes in developmental dyslexia.
- These findings support the development of personalized interventions for DD.
- Future research should focus on generalizability and specific marker-protocol associations.
More Related Videos
09:00Advancing Dyslexia Assessment in Children Through Computerized Testing
Published on: August 16, 2024
10:02Event Related Potentials (ERPs) and other EEG Based Methods for Extracting Biomarkers of Brain Dysfunction: Examples from Pediatric Attention Deficit/Hyperactivity Disorder (ADHD)
Published on: March 12, 2020