The objective of this experiment is to find whether different eye gaze directions : averted, direct, and natural, affects the inter-brain synchrony.
Explain the task more details here later !
Note : A difference between Hyperscanning2 and Hyperscanning2-redesign is the later adding questionnaire inside VR and also improved the UNITY that is used for the experiment.
In this experiment, there are three main data that will be analyzed :
Note : The storage refers to HPC of ABI
-
Separate EEG between baseline & experimetal data using this code (main branch)
This will extract EEG for both baseline and experimental data and save into
Baseline
/hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/raw_baseline_data/
Experimental
****` /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/raw_experimental_data/ `** **
2. Combine pre and post data (for each baseline and experimental), for all eye conditions, using this code (EEG-pre-processing branch). The result will be saved in the following folder
Baseline
**** /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/raw_baseline_data/raw_combined_baseline_data/
`` Experimental
**** /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/raw_experimental_data/raw_combined_experimental_data
3. Clean the above-(point no.2)-combined-EEG data for both baseline and experimental data using this code.
This will result in 3 files
1. list_deleted_epoch_indices_averted_baseline_post.pkl
2. list_circular_correlation_scores_all_pairs_averted_post_no_filter.pkl (ignore this kind of file. Not used in further analysis)
The 1st and 2nd files are located here
/hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/pre-processed_eeg_data/
3. preprocessed epoched files : for both baseline and experimental
`**`For epoched baseline data is located here`**`
/hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/pre-processed_eeg_data/raw_preproc_baseline_epoched_data/
``` **For experimental data is located here`**
NOTE : The following paths are only for EXPERIMENTAL data
averted_pre : /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/pre-processed_eeg_data/raw_preproc_experiment_epoched_data/averted_pre/
_ averted_post :_ /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/pre-processed_eeg_data/raw_preproc_experiment_epoched_data/averted_post/
`` direct_pre : /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/pre-processed_eeg_data/raw_preproc_experiment_epoched_data/direct_pre/
`` direct_post: /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/pre-processed_eeg_data/raw_preproc_experiment_epoched_data/direct_post/
`` natural_pre : /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/pre-processed_eeg_data/raw_preproc_experiment_epoched_data/natural_pre/
`` natural_post: /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/pre-processed_eeg_data/raw_preproc_experiment_epoched_data/natural_post/
- TODO : Update bad channels, in case the data has increased / updated
- Statistical analysis to check if the connection is significant or not. It saves significant connections as well as the actual score of such significant connection by using this code
For now, the number of permutation is 150. The higher, the longer time to take to process the data !
This step will find which connection that is statistically significant for each pair (out of 256 possible connections) within four different frequencies : theta, alpha, beta, and gamma.
**** This will result in 2 pkl files for each pair
1. *_Significant connection.pkl
2. *_Actual score of that significant connection along with the label of connection, eg. FP1 - F7
The files will be stored in various folders that are available in six forms.
NOTE : This is for EXPERIMENTAL data that has been permuted
averted_pre : /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/significant_connections/averted_pre/
_ averted_post :_ /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/significant_connections/averted_post/
`` direct_pre : /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/significant_connections/direct_pre/
`` direct_post: /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/significant_connections/direct_post/
`` natural_pre : /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/significant_connections/natural_pre/
`` natural_post: /hpc/igum002/codes/Hyperscanning2-redesign/data/EEG/significant_connections/natural_post/
TODO: It still needs to be moved to main branch. Once it is done, then change the commit hash that is located in the main branch
NOTE : 1.2. Add leading zero to subject number from 1 - 9 using a function of add_leading_zero which is available in this code so that it would make easier to sort out the data later on.
2. Count significant connection for each eye condition which is divided into different frequencies (theta, alpha, beta, and gamma) and algorithms (ccorr, coh, and plv). Us a function of total_significant_connections which is available in this code.
3. ANCOVA for all participants once the above step is completed. ANCOVA which compares the number of significant connections between eye condition within a specific frequency.
4. Use this code to count average significant actual score of specific connections out of all pairs (from dictionary), which have key (out of all participants).
NOTE : We need to populate into one container first (e.g. list) which has average score of each eye condition within a specific frequency and algorithm. This is still in progress here
Extract (separate data of baseline and experimental) using file of using extract_eye_tracker_new.py, combine raw files using combine_2_csv_files_eye_tracker_data.py, and clean up the raw eye tracker files using file cleaning_eye_tracker.py All of those codes are available in this repo. NOTE : Ignore file of extract_eye_tracker.py
Code construction in progress 🎉
- Calculation total score of each sub-scale of SPGQ
- Calculation total score of SPGQ
- Calculation total score of Co-Presence
- ANCOVA for total score of SPGQ
- ANCOVA for total score of Co-Presence
All the above stuff can be done via this code