Type I and Type II error concerns in fMRI research: re-balancing the scale - PubMed (original) (raw)
Type I and Type II error concerns in fMRI research: re-balancing the scale
Matthew D Lieberman et al. Soc Cogn Affect Neurosci. 2009 Dec.
Abstract
Statistical thresholding (i.e. P-values) in fMRI research has become increasingly conservative over the past decade in an attempt to diminish Type I errors (i.e. false alarms) to a level traditionally allowed in behavioral science research. In this article, we examine the unintended negative consequences of this single-minded devotion to Type I errors: increased Type II errors (i.e. missing true effects), a bias toward studying large rather than small effects, a bias toward observing sensory and motor processes rather than complex cognitive and affective processes and deficient meta-analyses. Power analyses indicate that the reductions in acceptable P-values over time are producing dramatic increases in the Type II error rate. Moreover, the push for a mapwide false discovery rate (FDR) of 0.05 is based on the assumption that this is the FDR in most behavioral research; however, this is an inaccurate assessment of the conventions in actual behavioral research. We report simulations demonstrating that combined intensity and cluster size thresholds such as P < 0.005 with a 10 voxel extent produce a desirable balance between Types I and II error rates. This joint threshold produces high but acceptable Type II error rates and produces a FDR that is comparable to the effective FDR in typical behavioral science articles (while a 20 voxel extent threshold produces an actual FDR of 0.05 with relatively common imaging parameters). We recommend a greater focus on replication and meta-analysis rather than emphasizing single studies as the unit of analysis for establishing scientific truth. From this perspective, Type I errors are self-erasing because they will not replicate, thus allowing for more lenient thresholding to avoid Type II errors.
Similar articles
- Evaluating methods of correcting for multiple comparisons implemented in SPM12 in social neuroscience fMRI studies: an example from moral psychology.
Han H, Glenn AL. Han H, et al. Soc Neurosci. 2018 Jun;13(3):257-267. doi: 10.1080/17470919.2017.1324521. Epub 2017 May 15. Soc Neurosci. 2018. PMID: 28446105 - Cluster-extent based thresholding in fMRI analyses: pitfalls and recommendations.
Woo CW, Krishnan A, Wager TD. Woo CW, et al. Neuroimage. 2014 May 1;91:412-9. doi: 10.1016/j.neuroimage.2013.12.058. Epub 2014 Jan 8. Neuroimage. 2014. PMID: 24412399 Free PMC article. - Balancing Type I and Type II error concerns in fMRI through compartmentalized analysis.
Cunningham WA, Koscik TR. Cunningham WA, et al. Cogn Neurosci. 2017 Jul;8(3):147-149. doi: 10.1080/17588928.2017.1299122. Epub 2017 Mar 13. Cogn Neurosci. 2017. PMID: 28285552 - Apparently conclusive meta-analyses on interventions in critical care may be inconclusive-a meta-epidemiological study.
Koster TM, Wetterslev J, Gluud C, Jakobsen JC, Kaufmann T, Eck RJ, Koster G, Hiemstra B, van der Horst ICC, Keus E. Koster TM, et al. J Clin Epidemiol. 2019 Oct;114:1-10. doi: 10.1016/j.jclinepi.2019.05.011. Epub 2019 Jun 11. J Clin Epidemiol. 2019. PMID: 31200004 Review. - Comparing methods of analyzing fMRI statistical parametric maps.
Marchini J, Presanis A. Marchini J, et al. Neuroimage. 2004 Jul;22(3):1203-13. doi: 10.1016/j.neuroimage.2004.03.030. Neuroimage. 2004. PMID: 15219592 Review.
Cited by
- Sex differences in the response to emotional distraction: an event-related fMRI investigation.
Iordan AD, Dolcos S, Denkova E, Dolcos F. Iordan AD, et al. Cogn Affect Behav Neurosci. 2013 Mar;13(1):116-34. doi: 10.3758/s13415-012-0134-6. Cogn Affect Behav Neurosci. 2013. PMID: 23293019 - Disrupted insula-based neural circuit organization and conflict interference in trauma-exposed youth.
Marusak HA, Etkin A, Thomason ME. Marusak HA, et al. Neuroimage Clin. 2015 Apr 10;8:516-25. doi: 10.1016/j.nicl.2015.04.007. eCollection 2015. Neuroimage Clin. 2015. PMID: 26199869 Free PMC article. - Altered spontaneous brain activity in lumbar disc herniation patients: insights from an ALE meta-analysis of neuroimaging data.
Qiu Z, Zhong X, Yang Q, Shi X, He L, Zhou H, Xu X. Qiu Z, et al. Front Neurosci. 2024 Feb 6;18:1349512. doi: 10.3389/fnins.2024.1349512. eCollection 2024. Front Neurosci. 2024. PMID: 38379762 Free PMC article. - The neural components of empathy: predicting daily prosocial behavior.
Morelli SA, Rameson LT, Lieberman MD. Morelli SA, et al. Soc Cogn Affect Neurosci. 2014 Jan;9(1):39-47. doi: 10.1093/scan/nss088. Epub 2012 Aug 9. Soc Cogn Affect Neurosci. 2014. PMID: 22887480 Free PMC article. - Bootstrapping fMRI Data: Dealing with Misspecification.
Roels SP, Moerkerke B, Loeys T. Roels SP, et al. Neuroinformatics. 2015 Jul;13(3):337-52. doi: 10.1007/s12021-015-9261-x. Neuroinformatics. 2015. PMID: 25672877
References
- Cox RW. AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages. Computers and Biomedical Research. 1996;29:162–73. - PubMed
- Fisher RA. The arrangement of field experiments. Journal of the Ministry of Agriculture of Great Britain. 1926;33:503–13.
- Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC. Improved assessment of significant activation in functional magnetic resonance imaging (fMRI): use of a cluster-size threshold. Magnetic Resonance in Medicine. 1995;33:636–47. - PubMed
- Genovese CR, Lazar NA, Nichols TE. Thresholding of statistical maps in functional neuroimaging using the false discovery rate. Neuroimage. 2002;15:870–8. - PubMed
- Griffin DW, Ross L. Subjective construal, social inference, and human misunderstanding. Advances in Experimental Social Psychology. 1991;24:319–59.
MeSH terms
Substances
LinkOut - more resources
Full Text Sources
Medical
Miscellaneous