TraceLAB: A MATLAB Toolbox for Interindividual Synchrony Analysis of Facial Expression and Head Movement Data...

Area of Science:

• Psychology
• Neuroscience
• Computer Science

Background:

• Facial expressions reveal internal states and shared experiences during content viewing.
• Interindividual expression synchrony can indicate shared information processing and engagement.
• Acquiring and analyzing facial data presents ethical and technical hurdles for researchers.

Purpose of the Study:

• To introduce TraceLAB, a MATLAB toolbox for preprocessing and analyzing facial data collected via the 'Trace' media player.
• To provide researchers with tools for quantifying interindividual synchrony in facial expressions and head movements.
• To overcome limitations in current facial data analysis for understanding shared experiences.

Main Methods:

• Developed TraceLAB, a MATLAB-based toolbox for formatting, aligning, and filtering anonymized facial landmark data.
• Implemented Correlated Component Analysis for quantifying expression synchrony.
• Implemented Surrogate Synchrony for quantifying head movement synchrony.
• Utilized simulated and real datasets for demonstrating TraceLAB's functionality.

Main Results:

• TraceLAB successfully preprocesses anonymized facial landmark data.
• Novel analysis techniques within TraceLAB enable quantification of expression and head movement synchrony.
• The toolbox demonstrates utility on both simulated and real-world datasets.

Conclusions:

• TraceLAB offers a solution to ethical and technical challenges in facial expression analysis.
• The toolbox facilitates the study of interindividual synchrony as a measure of shared information processing and engagement.
• TraceLAB enhances the potential for understanding subjective experiences through objective facial data analysis.