[FieldTrip] Choose minimun time-width of clusters for cluster based permutations on freq_statistics

Constantino Méndez Bértolo constantino.mendezbertolo at ctb.upm.es
Fri Jul 26 18:46:23 CEST 2013


We have Timelock data (LFPs). In order to run cluster based permutation
tests we generated a Template similar to a Freqdata structure, with

size(Template.powspctrm) = 9 1 1 320
Template.dimord = subj_chan_freq_time

We set cfg.neighbouts = [] and run ft_freqstatistics (method ='montecarlo';
correctm = 'cluster', etc)
This way we can run permutation tests and explore the output, hence,
clusters are made on time.

What we would like to do is:

- *Selecting a minimum amount of consecutive time-bins for a cluster to be
For example: Only if (say) 10 consecutive time points are above threshold
(whether 'maxsum' or 'wcm' criteria are being used) fieldtrip should
consider it a cluster.  If just one solely sample point is above
clusteralpha threshold, it should not be considered.

Is it possible? It would similar to cfg.minnbchan, but in time. Only a
cfg.minnbtbin does not exist! As we are already tricking freqstats to
believe is working with freqdata (with just one frequency) I am not sure if
this is possible. If someone has got some input it would be much appreciated

Constantino Méndez-Bértolo
Laboratorio de Neurociencia Clínica, Centro de Tecnología Biomédica (CTB)

Parque Científico y Tecnológico de la UPM, Campus de Montegancedo

28223 Pozuelo de Alarcón, Madrid, SPAIN
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20130726/935d37f4/attachment-0001.html>

More information about the fieldtrip mailing list