BEAM: Brainwave Empathy Assessment Model for Early Childhood.

Area of Science:

• Neuroscience
• Developmental Psychology
• Machine Learning

Background:

• Empathy is vital for children's social-emotional growth but difficult to predict accurately.
• Current assessment methods (self-report, observation) lack objectivity and miss dynamic empathy processes.
• Electroencephalography (EEG) offers objective measurement, but existing techniques overlook temporal dynamics.

Purpose of the Study:

• To introduce a novel deep learning framework, the Brainwave Empathy Assessment Model (BEAM).
• To predict empathy levels in children aged 4-6 years using multi-view EEG signals.
• To capture both cognitive and emotional aspects of empathy for a comprehensive assessment.

Main Methods:

• Developed BEAM, a deep learning framework utilizing multi-view EEG.
• Employed a LaBraM-based encoder for spatio-temporal feature extraction.
• Integrated a feature fusion module and a contrastive learning module for enhanced class separation.

Main Results:

• BEAM demonstrated superior performance compared to existing methods on the CBCP dataset.
• The model achieved high accuracy across multiple evaluation metrics.
• Results indicate BEAM's effectiveness in objective empathy assessment.

Conclusions:

• BEAM offers a promising, objective approach to assessing empathy in young children.
• The framework provides insights into early interventions for prosocial development.
• This method overcomes limitations of traditional subjective assessment techniques.