Removing an intersubject variance component in a general linear model improves multiway factoring of event-related spectral perturbations in group EEG studies

Jeffrey S. Spence, Matthew R. Brier, John Hart, Thomas C. Ferree

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Linear statistical models are used very effectively to assess task-related differences in EEG power spectral analyses. Mixed models, in particular, accommodate more than one variance component in a multisubject study, where many trials of each condition of interest are measured on each subject. Generally, intra- and intersubject variances are both important to determine correct standard errors for inference on functions of model parameters, but it is often assumed that intersubject variance is the most important consideration in a group study. In this article, we show that, under common assumptions, estimates of some functions of model parameters, including estimates of task-related differences, are properly tested relative to the intrasubject variance component only. A substantial gain in statistical power can arise from the proper separation of variance components when there is more than one source of variability. We first develop this result analytically, then show how it benefits a multiway factoring of spectral, spatial, and temporal components from EEG data acquired in a group of healthy subjects performing a well-studied response inhibition task. Hum Brain Mapp, 2013.

Original languageEnglish (US)
Pages (from-to)651-664
Number of pages14
JournalHuman Brain Mapping
Volume34
Issue number3
DOIs
StatePublished - Mar 2013

Fingerprint

Electroencephalography
Linear Models
Statistical Models
Healthy Volunteers
Brain

Keywords

  • Principal components
  • Statistical power
  • Time-frequency analysis

ASJC Scopus subject areas

  • Clinical Neurology
  • Anatomy
  • Neurology
  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology

Cite this

Removing an intersubject variance component in a general linear model improves multiway factoring of event-related spectral perturbations in group EEG studies. / Spence, Jeffrey S.; Brier, Matthew R.; Hart, John; Ferree, Thomas C.

In: Human Brain Mapping, Vol. 34, No. 3, 03.2013, p. 651-664.

Research output: Contribution to journalArticle

@article{29c2c95a7c7348caa248c46c833a7724,
title = "Removing an intersubject variance component in a general linear model improves multiway factoring of event-related spectral perturbations in group EEG studies",
abstract = "Linear statistical models are used very effectively to assess task-related differences in EEG power spectral analyses. Mixed models, in particular, accommodate more than one variance component in a multisubject study, where many trials of each condition of interest are measured on each subject. Generally, intra- and intersubject variances are both important to determine correct standard errors for inference on functions of model parameters, but it is often assumed that intersubject variance is the most important consideration in a group study. In this article, we show that, under common assumptions, estimates of some functions of model parameters, including estimates of task-related differences, are properly tested relative to the intrasubject variance component only. A substantial gain in statistical power can arise from the proper separation of variance components when there is more than one source of variability. We first develop this result analytically, then show how it benefits a multiway factoring of spectral, spatial, and temporal components from EEG data acquired in a group of healthy subjects performing a well-studied response inhibition task. Hum Brain Mapp, 2013.",
keywords = "Principal components, Statistical power, Time-frequency analysis",
author = "Spence, {Jeffrey S.} and Brier, {Matthew R.} and John Hart and Ferree, {Thomas C.}",
year = "2013",
month = "3",
doi = "10.1002/hbm.21462",
language = "English (US)",
volume = "34",
pages = "651--664",
journal = "Human Brain Mapping",
issn = "1065-9471",
publisher = "Wiley-Liss Inc.",
number = "3",

}

TY - JOUR

T1 - Removing an intersubject variance component in a general linear model improves multiway factoring of event-related spectral perturbations in group EEG studies

AU - Spence, Jeffrey S.

AU - Brier, Matthew R.

AU - Hart, John

AU - Ferree, Thomas C.

PY - 2013/3

Y1 - 2013/3

N2 - Linear statistical models are used very effectively to assess task-related differences in EEG power spectral analyses. Mixed models, in particular, accommodate more than one variance component in a multisubject study, where many trials of each condition of interest are measured on each subject. Generally, intra- and intersubject variances are both important to determine correct standard errors for inference on functions of model parameters, but it is often assumed that intersubject variance is the most important consideration in a group study. In this article, we show that, under common assumptions, estimates of some functions of model parameters, including estimates of task-related differences, are properly tested relative to the intrasubject variance component only. A substantial gain in statistical power can arise from the proper separation of variance components when there is more than one source of variability. We first develop this result analytically, then show how it benefits a multiway factoring of spectral, spatial, and temporal components from EEG data acquired in a group of healthy subjects performing a well-studied response inhibition task. Hum Brain Mapp, 2013.

AB - Linear statistical models are used very effectively to assess task-related differences in EEG power spectral analyses. Mixed models, in particular, accommodate more than one variance component in a multisubject study, where many trials of each condition of interest are measured on each subject. Generally, intra- and intersubject variances are both important to determine correct standard errors for inference on functions of model parameters, but it is often assumed that intersubject variance is the most important consideration in a group study. In this article, we show that, under common assumptions, estimates of some functions of model parameters, including estimates of task-related differences, are properly tested relative to the intrasubject variance component only. A substantial gain in statistical power can arise from the proper separation of variance components when there is more than one source of variability. We first develop this result analytically, then show how it benefits a multiway factoring of spectral, spatial, and temporal components from EEG data acquired in a group of healthy subjects performing a well-studied response inhibition task. Hum Brain Mapp, 2013.

KW - Principal components

KW - Statistical power

KW - Time-frequency analysis

UR - http://www.scopus.com/inward/record.url?scp=84873440349&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84873440349&partnerID=8YFLogxK

U2 - 10.1002/hbm.21462

DO - 10.1002/hbm.21462

M3 - Article

C2 - 22102426

AN - SCOPUS:84873440349

VL - 34

SP - 651

EP - 664

JO - Human Brain Mapping

JF - Human Brain Mapping

SN - 1065-9471

IS - 3

ER -