[FieldTrip] correlating EEG Data with behavioural data

[tim.redepenning at uni-bielefeld.de]

Dear list,

I have a variable 'X' with 29 × 1 values that tells me how strong the
desired behavioural effect is.

 

And I have a matrix 'ITC_values' with 29 × 62 × 12 × 200 with EEG (ITC)
values.

 

29 participants

62 channels

12 frequencies

200 time points

 

Could anyone suggest me an approach to find significant correlations between
my EEG data and my behavioural data?

 

Addition:

I had the idea to calculate Mutual information

So, I calculated the Mutual information between

X and ITC_values(: , 1 , 1 , 1)

and

X  and ITC_ values (: , 1, 1, 2)

and so on. 

In other words: I calculated the relationship between the hysteresis values
and each timepoint at each frequency on each channel.

So, my result is a matrix with the size of 62 × 12 × 200 that holds all
Mutual Information values.

Is this a valid approach?

Cause now I want to check the significance with a cluster permutation test.
But that normally works only across subjects. But the MI values are already
the result of aggregating participants.

 

 

Best wishes

Tim

 

Tim Redepenning

 

PhD Student | “Motor Control and Training” – Research Group

Faculty of Psychology and Sports Science

Bielefeld University | Cognitive Interaction Technology – Center of
Excellence (CITEC)

33501 Bielefeld

Germany

 

Room: CITEC-1.411 | Phone: +49-(0)521-106-5129

E-Mail:  <mailto:tim.redepenning at uni-bielefeld.de>
tim.redepenning at uni-bielefeld.de

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20250323/2426b0cb/attachment.htm>