[Eeglablist] toolboxes for multiple testing of time-frequency data

[Archana Singh]

Dear All,

>From time to time, I see people asking for options to correct for multiple
comparison of statistics from high-dimensional time/frequency data, such as
ERSP. A part of my research concerns the very issue of controlling false
positives in EEG functional connectivity studies (see the references
below). Detecting significance from  a huge data in 3D (sensors, time, and
frequency intervals) is like searching for a needle in the hay. Most
multiple testing methods, including FDR, tend to control the false
positives at the cost of hiding true positives - so  we often end up
loosing out on significance when it should in fact be evident.

I would like to share some toolboxes that offer alternative options to
maximizes the power without compromising on control of false positives.
 These options, called hierarchical FDR control and Optimal Discovery
Procedure, are adopted from the approaches developed by  Yekutieli and
Storey, respectively,  for their application to EEG data. See the
references below for details.

These toolboxes are in matlab and R, if you like to try them, send me an
email.


1) Yekutieli, D., 2008. Hierarchical false discovery rate controlling
methodology. Journal of the American Statistical Association 103 (481),
309–316.
2) Storey, J.D., 2007. The optimal discovery procedure: a new approach to
simultaneous significance testing. Journal of the Royal Statistical
Society: Series B (Statistical Methodology) 69 (3), 347–368.
3) Phillips, S., Takeda, Y., & Singh, A. K. Visual feature integration
indicated by phase-locked frontal-parietal EEG signals. PLoS ONE 7(3):
e32502, 2012
4) Singh A. K., Asoh H.,  and  Phillips S., Optimal detection of functional
connectivity from high-dimensional EEG synchrony data, NeuroImage, 58(1),
148-156, 2011.
5) Singh A. K. and  Phillips S.,  Hierarchical control of false discovery
rate for phase locking measures of EEG synchrony, NeuroImage, 50(1), 40-47,
2010

Archana

_________________________________________

Archana K Singh,
ERATO Researcher,
Functional and Chemosensory brain imaging group
Department of Applied Biological Chemistry
Graduate School of Agricultural and Life Sciences
The University of Tokyo, Tokyo 113-8657
(tel)+81-3-5841-5590, (fax)+81-3-5841-8024
(email) archana at mail.ecc.u-tokyo.ac.jp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20131221/ea37cdf4/attachment.html>