Permutation tests for time-frequency representations: Within trial experiments

[Eric Maris]

Hi Martin,







I want to compute the permutation tests for time-frequency representations:
within trial.

In the tutorial it it stated that i have to select equal length intervals.

'To perform this comparison by means of a permutation test, we have to
select equal-length non-overlapping time intervals in the baseline and the
activation period.'



However, my pre-stimulus and post-stimulus periods have a different length.
My baseline is only 500 msec, while the post stimulus period is 1.5 and I am
interested in the complete post-stimulus period.

Are there possibilities to use unequal intervals?

Would it make sense to split up the post stimulus period into 3 parts and
compute the statistics for each post stimulus period?



Yes, this is the way you should proceed. And, if you want to be safe on the
statistical side, use Bonferroni correction (per-comparison alpha-level =
0.05/3) to accommodate for the fact that you run three tests instead of one.





Good luck,



Eric Maris











Thanks

Best,



Martin

----------------------------------

The aim of this list is to facilitate the discussion between users of the
FieldTrip toolbox, to share experiences and to discuss new ideas for MEG and
EEG analysis.

http://listserv.surfnet.nl/archives/fieldtrip.html

http://www.ru.nl/fcdonders/fieldtrip/


----------------------------------
The aim of this list is to facilitate the discussion between users of the FieldTrip  toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/fcdonders/fieldtrip.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20090122/f654752e/attachment-0001.html>