EEG Based Functional Brain Network Analysis and Classification of Dyslexic Children During Sustained Attention Task

Area of Science:

• Neuroscience
• Cognitive Science
• Developmental Psychology

Background:

• Reading is a complex cognitive skill reliant on visual, attention, and linguistic abilities.
• Sustained attention is crucial for effective reading and learning.
• Dyslexia is a common learning disorder characterized by difficulties in reading acquisition.

Purpose of the Study:

• To investigate functional brain network connectivity during sustained attention in dyslexic children.
• To compare network properties between dyslexic and non-dyslexic children during a demanding cognitive task.
• To identify neural markers associated with attention deficits in dyslexia.

Main Methods:

• Electroencephalogram (EEG) signals were recorded from 15 dyslexic and 15 non-dyslexic children (ages 9-10).
• Participants performed a visual continuous performance task (VCPT) to assess sustained attention.
• Graph theory metrics (clustering coefficient, pathlength, efficiency) were analyzed to characterize brain network topology.

Main Results:

• Dyslexic children made significantly more omission and commission errors during the VCPT.
• The dyslexic group displayed altered network properties, including a lower clustering coefficient and longer characteristic pathlength.
• Reduced global and local efficiency in theta and alpha frequency bands was observed in dyslexic children's brain networks.
• A k-nearest neighbor (KNN) classifier achieved 96.7% accuracy in distinguishing between dyslexic and non-dyslexic groups based on EEG data.

Conclusions:

• Dyslexic children exhibit impaired functional segregation and disturbed information transfer within brain networks during sustained attention.
• These network alterations may underlie the attention deficits observed in dyslexia.
• EEG-based network analysis shows promise for identifying and potentially diagnosing dyslexia.