search

Centre-IRM-INT / GT-MVPA-nilearn

GT MVPA nilearn from Marseille
3 stars 3 forks source link

MVPA & inter-subject approach #6

Open JeanLucAnton opened 3 years ago

JeanLucAnton commented 3 years ago

Q : How big a role does inter-individual variability play in a classifier’s ability to categorise activations? Is the classification done on an individual basis, or on a group basis? A : on an individual basis.

JeanLucAnton commented 3 years ago

Q : Could we potentially train a classifier on data from one group of participants and then test it on data from a separate group of participants? A : Sylvain says yes, it's possible. Also, can combine data from several subjects for the classifier - even though it's noisier because of inter-subject variability, the fact that you've increased the number of data-points makes it worth it.

SylvainTakerkart commented 3 years ago

As with most of these questions, there's no strict universal answers... This is just a matter of compromises between different factors that pull in different directions... In our Neuroimage paper (https://doi.org/10.1016/j.neuroimage.2019.116205), we show that we can explain a fair amount of what we observe from two main factors: the size of the effect and the amount of variability. But since these two can vary depending on the brain region and on the task performed, it's difficult to make a general claim on which is the most powerful between inter- and within-subject decoding.

But yes, overall, in this paper, we demonstrate that it seems feasible to perform inter-subject decoding without any fancy method (such as hyper-alignement)!