Hi,

I am trying to do a correlation between some behavioural index and channel-wise beta values (I have got 52 channels). After I run the contrast using spm_fnirs for individual participant, I get a con_0001.mat file which seems to be the channel-wise contrast. However, these values seem to be very small (.00004 - .0001), although I see significant activations when looking at the brain map after I perform the contrast using spm_fnirs.

I am concerned whether I should actually be using these values to perform my correlation with behavioural index. I have tried using another toolbox (NIRS_SPM) which also outputs a ".txt" file with channel-wise beta values and those values (for the same participant) seem to be much larger (.02-.06), although the activations on the brain seem less vivid.

It is worth mentioning that I have mostly used the same kind of pre-processing in both toolboxes - DCT filtering and HRF. In spm-fnirs I have used the MARA motion correction.

I would like to know in spm-fnirs, what could explain such small values for beta for the participant - and whether the con_0001 file indeed gives channel-wise beta values. If it does not, where can I retrieve channel-wise beta for spm-fnirs. By the way, I have done a correlation between the spm_fnirs and NIRS_SPM bet values and they significantly NEGATIVELY correlated.

Any help is greatly appreciated. 

Thank you,
Avijit Chowdhury
PhD student, NUS