Best practices for repeated measures ANOVAs of ERP data: Reference, regional channels, and robust ANOVAs

Published on Jan 1, 2017in International Journal of Psychophysiology2.41
· DOI :10.1016/j.ijpsycho.2016.09.006
Joseph Dien14
Estimated H-index: 14
(UMD: University of Maryland, College Park)
Abstract Analysis of variance (ANOVA) is a fundamental procedure for event-related potential (ERP) research and yet there is very little guidance for best practices. It is important for the field to develop evidence-based best practices: 1) to minimize the Type II error rate by maximizing statistical power, 2) to minimize the Type I error rate by reducing the latitude for varying procedures, and 3) to identify areas for further methodological improvements. While generic treatments of ANOVA methodology are available, ERP datasets have many unique characteristics that must be considered. In the present report, a novelty oddball dataset was utilized as a test case to determine whether three aspects of ANOVA procedures as applied to ERPs make a real-world difference: the effects of reference site, regional channels, and robust ANOVAs. Recommendations are provided for best practices in each of these areas.
  • References (68)
  • Citations (14)
Published on Jun 1, 2015in Psychophysiology3.38
Scott A. Baldwin30
Estimated H-index: 30
(BYU: Brigham Young University),
Michael J. Larson34
Estimated H-index: 34
(BYU: Brigham Young University),
Peter E. Clayson18
Estimated H-index: 18
(UCLA: University of California, Los Angeles)
Psychometric studies of the ERN, CRN, Pe, and Pc ERPs are increasing. Coherent integration of these results is difficult with classical test theory because the definition of error depends on the measure of reliability. This study used generalizability theory, which extends the ideas of classical test theory, as a framework for evaluating the influence of psychopathology and number of trials on dependability of measurement. Participants included 34 people meeting criteria for major depression, 29...
Published on May 1, 2014in Physiology & Behavior2.63
Renske Huffmeijer10
Estimated H-index: 10
(LEI: Leiden University),
Marian J. Bakermans-Kranenburg77
Estimated H-index: 77
(LEI: Leiden University)
+ 1 AuthorsMarinus H. van IJzendoorn93
Estimated H-index: 93
(LEI: Leiden University)
Abstract The reliability of event-related potentials (ERPs) is an important factor determining the value of studies relating ERP components to individual differences. However, studies examining the reliability of ERPs are surprisingly scarce. The current study examines the test–retest reliability of ERP components (VPP, N170, MFN, FRN, P3, and LPP) in response to feedback stimuli combining performance feedback with emotional faces in a sample of healthy female adults. In general, ERP amplitudes ...
Published on Jan 1, 2013in Frontiers in Psychology2.13
Daniel Lakens23
Estimated H-index: 23
(TU/e: Eindhoven University of Technology)
Effect sizes are the most important outcome of empirical studies. Most articles on effect sizes highlight their importance to communicate the practical significance of results. For scientists themselves, effect sizes are most useful because they facilitate cumulative science. Effect sizes can be used to determine the sample size for follow-up studies, or examining effects across studies. This article aims to provide a practical primer on how to calculate and report effect sizes for t-tests and A...
Published on Aug 1, 2012in Developmental Neuropsychology1.34
Joseph Dien14
Estimated H-index: 14
(UMD: University of Maryland, College Park)
Principal components analysis (PCA) has attracted increasing interest as a tool for facilitating analysis of high-density event-related potential (ERP) data. While every researcher is exposed to this statistical procedure in graduate school, its complexities are rarely covered in depth and hence researchers are often not conversant with its subtleties. Furthermore, application to ERP datasets involves unique aspects that would not be covered in a general statistics course. This tutorial seeks to...
Published on Jan 1, 2010
Hans Fischer1
Estimated H-index: 1
This book provides a comprehensive history of one of the most important theorems of probability theory and mathematical statistics from about 1810 to ca. 1950. Taking problems surrounding the central limit theorem as characteristic examples, the development of probability theory from its classical form to its modern, and even postmodern, shape is illuminated. The main topics of the book include: Laplace's normal approximations and its applications to natural sciences; Poisson's modifications; Di...
Published on Mar 1, 2010in Journal of Neuroscience Methods2.79
Joseph Dien17
Estimated H-index: 17
(UMD: University of Maryland, College Park)
This article presents an open source Matlab program, the ERP PCA (EP) Toolkit, for facilitating the multivariate decomposition and analysis of event-related potential data. This program is intended to supplement existing ERP analysis programs by providing functions for conducting artifact correction, robust averaging, referencing and baseline correction, data editing and visualization, principal components analysis, and robust inferential statistical analysis. This program subserves three major ...
Published on Jan 1, 2010in Psychophysiology3.38
Joseph Dien17
Estimated H-index: 17
(University of Louisville)
Principal components analysis (PCA) can facilitate analysis of event-related potential (ERP) components. Geomin, Oblimin, Varimax, Promax, and Infomax (independent components analysis) were compared using a simulated data set. Kappa settings for Oblimin and Promax were also systematically compared. Finally, the rotations were also analyzed in a two-step PCA procedure, including a contrast between spatiotemporal and temporospatial procedures. Promax was found to give the best overall results for ...
Published on Sep 1, 2009in Contemporary Clinical Trials2.28
Morten W. Fagerland25
Estimated H-index: 25
(Oslo University Hospital),
Leiv Sandvik56
Estimated H-index: 56
(Oslo University Hospital)
Tests for comparing the locations of two independent populations are associated with different null hypotheses, but results are often interpreted as evidence for or against equality of means or medians. We examine the appropriateness of this practice by investigating the performance of five frequently used tests: the two-sample T test, the Welch U test, the Yuen–Welch test, the Wilcoxon–Mann–Whitney test, and the Brunner–Munzel test. Under combined violations of normality and variance homogeneit...
Published on Jan 1, 2009in Biological Psychology2.63
Joseph Dien17
Estimated H-index: 17
(KU: University of Kansas)
This paper first provides a brief review of the functional neuroanatomy of reading single words, focusing on the lexical and phonological routes. Next, early (defined as peaking prior to 350 ms) reading event-related potential components are summarized. A comprehensive effort is made to organize existing observations into a coherent scheme and commentary is made on terminology. Proposals are made regarding the cognitive function reflected by each ERP component and the associated generator sites....
Published on Nov 1, 2008in Psychophysiology3.38
J Peterrosenfeld29
Estimated H-index: 29
(NU: Northwestern University),
Elena Labkovsky8
Estimated H-index: 8
(NU: Northwestern University)
+ 3 AuthorsErica Chedid1
Estimated H-index: 1
(NU: Northwestern University)
A new P300-based concealed information test is described. A rare probe or frequent irrelevant stimulus appears in the same trial in which a target or nontarget later appears. One response follows the first stimulus and uses the same button press regardless of stimulus type. A later second stimulus then appears: target or nontarget. The subject presses one button for a target, another for a nontarget. A P300 to the first stimulus indicates probe recognition. One group was tested in 3 weeks for de...
Cited By14
Published on Sep 1, 2019in Biological Psychology2.63
Fang Cui4
Estimated H-index: 4
(SZU: Shenzhen University),
Chengyao Wang1
Estimated H-index: 1
(SZU: Shenzhen University)
+ 1 AuthorsCan Jiao (SZU: Shenzhen University)
Abstract Third-party punishment refers to the punishment imposed by a party whose benefit is not directly affected by any norm violations. The present study using a modified Dictator Game explored how social statuses of the involved parties modulate punishment decisions of a third-party. Systematic manipulation of the dictators’ and recipients’ status revealed that: higher recipient status correlated with more severe punishment, but no effect of the dictator’s. We further focused on the neural u...
Published on 2019in Psychophysiology3.38
Samantha Pegg (Vandy: Vanderbilt University), Lindsay Dickey (Vandy: Vanderbilt University)+ 3 AuthorsAutumn Kujawa19
Estimated H-index: 19
(Vandy: Vanderbilt University)
Published on Jul 19, 2019in Psychophysiology3.38
Peter E. Clayson18
Estimated H-index: 18
(UCLA: University of California, Los Angeles),
Kaylie A. Carbine5
Estimated H-index: 5
(BYU: Brigham Young University)
+ 1 AuthorsMichael J. Larson34
Estimated H-index: 34
(BYU: Brigham Young University)
Published on Dec 7, 2018in Psychophysiology3.38
Florian Scharf1
Estimated H-index: 1
(Leipzig University),
Steffen Nestler18
Estimated H-index: 18
(Leipzig University)
Published on Dec 4, 2018in Social Cognitive and Affective Neuroscience3.66
Daniela M Pfabigan (McGovern Institute for Brain Research), Anna M. Wucherer2
Estimated H-index: 2
(University of Vienna)
+ 3 AuthorsShihui Han45
Estimated H-index: 45
(McGovern Institute for Brain Research)
Published on Nov 1, 2018in Journal of Neuroscience Methods2.79
Audrey Bürki9
Estimated H-index: 9
(University of Geneva),
Jaromil Frossard1
Estimated H-index: 1
(University of Geneva),
Olivier Renaud13
Estimated H-index: 13
(University of Geneva)
Abstract Background Event-related potentials (ERPs) are increasingly used in cognitive science. With their high temporal resolution, they offer a unique window into cognitive processes and their time course. In this paper, we focus on ERP experiments whose designs involve selecting participants and stimuli amongst many. Recently, Westfall et al. (2017) highlighted the drastic consequences of not considering stimuli as a random variable in fMRI studies with such designs. Most ERP studies in cogni...
Published on Oct 10, 2018in Journal of Advanced Transportation1.98
Weiwei Qi1
Estimated H-index: 1
(SCUT: South China University of Technology),
Zhexuan Wang1
Estimated H-index: 1
(SCUT: South China University of Technology)
+ 1 AuthorsLinhong Wang1
Estimated H-index: 1
(JLU: Jilin University)
Drivers’ mistakes may cause some traffic accidents, and such accidents can be avoided if prompt advice could be given to drivers. So, how to detect driving risk is the key factor. Firstly, the selected parameters of vehicle movement are reaction time, acceleration, initial speed, final speed, and velocity difference. The ANOVA results show that the velocity difference is not significant in different driving states, and the other four parameters can be used as input variables of neural network mo...
Published on Oct 1, 2018in International Journal of Psychophysiology2.41
Paige Ethridge2
Estimated H-index: 2
(McGill University),
Anna Weinberg26
Estimated H-index: 26
(McGill University)
Abstract Reward-related event-related potentials (ERPs) are often used to index individual differences that signal the presence or predict the onset of psychopathology. However, relatively little research has explored the psychometric properties of reward-related ERPs. Without understanding their psychometric properties, the value of using ERPs as biomarkers for psychopathology is limited. The present study, therefore, sought to establish the internal consistency reliability and convergent valid...
Published on Jul 1, 2018in Behavioural Brain Research2.77
Allan Campopiano2
Estimated H-index: 2
(Brock University),
Stefon J.R. van Noordt8
Estimated H-index: 8
(Yale University),
Sidney J. Segalowitz44
Estimated H-index: 44
(Brock University)
Abstract Research on robust statistics during the past half century provides concrete evidence that classical hypothesis tests that rely on the sample mean and variance are problematic. Even seemingly minor departures from normality are now known to create major problems in terms of increased error rates and decreased power. Fortunately, numerous robust estimation techniques have been developed that circumvent the need for strict assumptions of normality and equal variances, leading to increased...
View next paperAnalysis of Variance: The Fundamental Concepts