Estimation of the number of "true" null hypotheses in multivariate analysis of neuroimaging data

Citation
Fe. Turkheimer et al., Estimation of the number of "true" null hypotheses in multivariate analysis of neuroimaging data, NEUROIMAGE, 13(5), 2001, pp. 920-930
Citations number
37
Language
INGLESE
art.tipo
Article
Categorie Soggetti
Neurosciences & Behavoir
Journal title
NEUROIMAGE
ISSN journal
1053-8119 → ACNP
Volume
13
Issue
5
Year of publication
2001
Pages
920 - 930
Database
ISI
SICI code
1053-8119(200105)13:5<920:EOTNO">2.0.ZU;2-I
Abstract
The repeated testing of a null univariate hypothesis in each of many sites (either regions of interest or voxels) is a common approach to the statisti cal analysis of brain functional images. Procedures, such as the Bonferroni , are available to maintain the Type I error of the set of tests at a speci fied level. An initial assumption of these methods is a "global null hypoth esis," i.e., the statistics computed on each site are assumed to be generat ed by null distributions. This framework may be too conservative when a sig nificant proportion of the sites is affected by the experimental manipulati on. This report presents the development of a rigorous statistical procedur e for use with a previously reported graphical method, the P plot, for esti mation of the number of "true" null hypotheses in the set. This estimate ca n then be used to sharpen existing multiple comparison procedures. Performa nce of the P plot method in the multiple comparison problem is investigated in simulation studies and in the analysis of autoradiographic data. (C) 20 01 Academic Press.