NOTE :
Before we run analysis,

• we take average score of questionnaire for pre and post
• then substract post and pre

6ef7b5b

Correlation between SPGQ total & Direct

• Significant correlations in :

• coherence - theta : (0.6189193814333278, p-val : 0.03189176226829951
• coherence - beta : (0.564254914717613, p-val : 0.05599109700636175)
• coherence - gamma : (0.6105006072150732, p-val : 0.034995765473106906)
• PLV - gamma : (0.6538593554639102, p-val : 0.021091591502908406)

Good reason why we may want to choose PLV, does not require stationary of EEG = the stability of core characteristics of the time series

https://academic.oup.com/scan/article/16/1-2/72/5919711?login=false

from hyperscanning2-redesign.

ToDo :

• - Do this with average score
• - The actual score is too small
• Not continuing

from hyperscanning2-redesign.

from hyperscanning2-redesign.

NOTE :
Before we run analysis,

• we take average score of questionnaire for pre and post
• then substract post and pre

7064715 use this commit to see
Significant correlation between Behavioral SPGQ & Direct

total_sig_plv_gamma_connections
(0.5986361869534904, p-val : 0.039728851739450406)

from hyperscanning2-redesign.

1718bcb
Sig correlation SPGQ for averted & direct

• Substract post & pre for both the number of interbrain connections and SPGQ
• Make it absolute (np.abs) in order to remove the minus
• Negative correlation : averted & SPGQ
• Postive correlation : direct & SPGQ (beta and gamma)
• For averted & SPGQ , it is significantly correlated in beta

Particularly, run this sections, in the file above

Find difference of number of connections between pre and post
NOTE : The variable name is exactly the same with section 2.1 . Make sure you run the function of total_significant_connections first !!.

IMPORTANT : Change no. 9 to whatever condition that you want to test. See the multiple output of total_significant_connections function

Combine SPGQ Total score of subject1 & 2, etc..
NOTE : With subtraction of post and pre

Correlation SPGQ and Averted *
(r = -0.7081740226371446, p-val : 0.00995639735361368)

Sig. Correlation SPGQ and Direct *

Beta
(r = 0.6613483937034923, p-val = 0.019179026955168273)

Gamma
(r = 0.7016103719784926, p-val = 0.010991581361881438)

Table of total inter-brain connections and SPGQ Total score (after being substracted pre and post)

from hyperscanning2-redesign.

The above result is rejected because the permutation is wrong.

See below significant results with correct permutation

d042347 Use that commit

Added other analysis

• Number of connections (after reducing post and pre for number of interbrain connections)
• Significant connections (subtracted post - pre) in direct and SPGQ
• Sig connections (subtracted post - pre) , negative feelings, empathy, & behavioral. Plus CoPresence. It is all in theta frequency
• Added friedmans test but no significant difference
• Added correlation test between inter-brain connections (combine all frequencies but considering reducing post and pre) and SPGQ total

from hyperscanning2-redesign.

Sig. Correlation SPGQ and Direct * - Subtracted post and pre

(0.6086911100596132, 0.027262246738311523)

from hyperscanning2-redesign.

Sig. Correlation Empathy SPGQ and Direct * - Subtracted post and pre

(0.7364941834340334, 0.004090001679791294)

from hyperscanning2-redesign.

Correlations of inter-brain connections and subscales of SPGQ as well as CoPresence

Negative feelings SPGQ and Direct - Subtracted post and pre

(-0.7057681021024668, 0.007026657759570824)

Behavioural SPGQ and Direct - Subtracted post and pre

(0.7800821906829145, 0.001657724392669428)

CoPresence and Direct - Subtracted post and pre

(-0.7057681021024668, 0.007026657759570824)

from hyperscanning2-redesign.

Send number of connections to Amit and Mark. Don't forget to give explanation the difference (post and pre) that we calculated for correlation as well as friedmans test

from hyperscanning2-redesign.

Create a function for correlation between EEG connections and SPGQ (Put it under stats.py). Please refer to this code

• Find difference between EEG connections for both post and pre (for all eye conditions)

  • Calculate the number of connections with count_sig_connections (phd_codes/EEG/stats.py/Connections)

  • Extract only the first 4 since, they are using Ccor algorithm

  • Calculate the difference between post and pre. See this line

  • See this b46837c
    (corr-EEG-SPGQ) Difference of connections between post and pre(EEG)

    It extracts only ccor algorithm for this correlational score

• Find difference between SPGQ

  • Combine score between a pair. See this line first and then this line
  • Calculate the difference between pre and post. See this line
  • See this 927100d

• Correlation between EEG connection (diff) and SPGQ (diff) See this line

• See this 927100d
  (corr-EEG-SPGQ) Pearson correlation EEG connections & SPGQ questionnaire

  • Do the correlation for the subscales . See this 4cb0f30
    Make it as a function that returns the correlational score of total score of SPGQ and its subscale with respect to eye condition along with frequency. See this 4cb0f30

• Visualization of correlation. See this 6620a33 or scroll down

from hyperscanning2-redesign.

4cb0f30

Added funct calculate correlation EEG & spgq total
Added funct calculate correlation EEG & spgq subscales

from hyperscanning2-redesign.

6620a33 (use this since it has visualization correlation function)
Added functions to extract raw scores of EEG & SPGQ

 - Extract raw scores of a number of EEG connections
- SPGQ total scores, SPGQ subscale scores
- Implementation of calling those functions within diff_connections_pre_post_gaze.py
- Note that in order to use visualization of correlation we need to use raw score eeg spgq first. We need to use its result as input of vis correlation function

from hyperscanning2-redesign.

5fcac1b

Implementing visualization of sig correlations (EEG & SPGQ)

Added two figures of sig correlations:
- EEG connections & SPGQ Total score (Direct-Theta)
- EEG connections & Behaviour (SPGQ subscale) (Direct-Theta)

from hyperscanning2-redesign.