A Pipeline for fNIRS Data Analysis: MATLAB Code for Naturalistic Hyperscanning Studies

MATLAB analysis code accompanying the paper 'A Pipeline for fNIRS Data Analysis in Naturalistic Paradigms.' The code implements three analyses: heart-rate estimation from raw fNIRS, a cardiac signal-to-noise ratio framework for evaluating preprocessing pipelines, and interpersonal neural synchronization. It was applied to two naturalistic hyperscanning datasets named Long and ToMCAT.

Use Cases

• Benchmarking motion-correction schemes for fNIRS data based on the described cardiac SNR framework
• Estimating heart rate from fNIRS signals based on the method validated against simultaneous ECG
• Analyzing interpersonal neural synchronization using wavelet transform coherence as described in the pipeline
• Evaluating the impact of preprocessing steps, including spline interpolation and PCA, on signal quality

Strengths

• Implements a validated method for heart-rate estimation from fNIRS, cross-referenced with ECG data
• Benchmarks seven specific motion-correction schemes and a global-PCA threshold, providing a comparative framework
• Includes a control analysis using a 0.1-Hz notch filter for the neural synchronization study

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download
• Column-level documentation is absent; field semantics must be inferred after download

Provenance

Source

Harvard Dataverse, authored by Ye, ChuJie

Collection Method

Analysis code developed for a specific research paper's methodology.

Freshness

Last updated 2026-06-16 03:32:57; freshness should be verified