DiGyt / NBP_Hyperscanning

Code for the IKW Osnabrück's 2019/20 Hyperscanning study project.

NBP_Hyperscanning

Code for the IKW Osnabrücks's 2019/20 Hyperscanning study project.

This repository contains all the relevant code for our study project. It is mainly separated into two parts: The experiment implementation, and the analysis implementation.

Data Analysis

This part of the repository was mainly concerned with data analysis.

The purpose of each file is explained in its header.

Since there were multiple approaches and drawbacks, there are many files seemingly fullfilling the same purpose. This is due to many switches in implementation: For example main_preprocessing.py implements manual preprocessing as a python function, aimed to work when run from a local python install. main_preprocesssing.ipynb implements manual remote preprocessing, working as a jupyter notebook which can be run through the browser, while using the IKW computers for calculations. main_preprocessing_auto.ipynb (which was used for the final analysis) implements remote preprocessing, but only manually marking ICA components, while data cleaning is performed by autoreject.

The files used for the final analysis are:

• All the functions_ files, providing functions to work with.
• main_preprocessing_auto.ipynb to preprocess the data and annotate bad channels, segments, and ICs.
• main_phases.py, to apply the preprocessing and calculate the phase vectors.
• main_ispc.py to calculate the ISPCs from the phase vectors.
• main_swi.py to calculate the Small World Index from the ISPCs.

Statistics and visualisation were performed in main_statistics.ipynb and the plot_ files.

Experiment

The code to run our experiment.