Hi Francesco,

Not really, for the same reason we don't do permutation with fMRI time series.

Maybe if the sessions are well far apart from each other, but there is a substantial risk for invalid results.

That said, if the data has already been collected, it may well be worth running the analysis and, if the effects are really strong, consider presenting a case report, being open about this limitation.

All the best,

Anderson

On 9 March 2017 at 20:42, Francesco Sammartino <[log in to unmask]> wrote:

Hi

Would it be theoretically ok to run randomise on a single subject analysis where the subject has been scanned under multiple conditions?

I really have only one subject to study with a unique (for the location) brain implant.

Thanks