Neurobiosensory and Visual Patterns of Attentional and Emotional Functioning in Children Using Machine Learning:...

Area of Science:

• Child psychology and neuroscience
• Machine learning applications in healthcare
• Developmental disorders screening

Background:

• Childhood attention is influenced by cognitive, behavioral, and emotional factors.
• Executive function deficits are common in children with attentional vulnerabilities.
• Integrated screening is needed to address multidimensional child functioning.

Purpose of the Study:

• To create an automated, real-time, multimodal screening method for children's attentional profiles.
• To integrate oculomotor, emotional, and neuropsychological data for early detection.
• To support personalized interventions by identifying individualized behavior patterns.

Main Methods:

• A cross-sectional study of 260 children (aged 7-15) using neuropsychological tests, eye-tracking, and facial emotion recognition.
• Simultaneous, real-time capture of oculomotor, emotional, and cognitive data.
• Machine learning (Random Forest) analyzed data to differentiate attentional performance levels.

Main Results:

• The Random Forest classifier achieved high accuracy (94.2%) and AUC (0.9527) in identifying attentional performance.
• Children with lower attention showed higher emotional reactivity and altered visual exploration patterns.
• Emotional stimuli (anger, fear) were key in differentiating attentional performance.

Conclusions:

• Simultaneous multimodal data capture with machine learning aids in identifying behavioral patterns linked to attention.
• This approach offers ecologically valid, behavior-centered indicators for screening and intervention.
• The findings support enhanced screening and personalized intervention strategies in clinical and educational settings.