Cluster-extent based thresholding in fMRI analyses: Pitfalls and recommendations

By Choong-Wan Woo, Anjali Krishnan, Tor D. Wager. NeuroImage, 2014

1,264 citations recorded by the source corpus.

Cluster-extent based thresholding is currently the most popular method for multiple comparisons correction of statistical maps in neuroimaging studies, due to its high sensitivity to weak and diffuse signals. However, cluster-extent based thresholding provides low spatial specificity; researchers can only infer that there is signal somewhere within a significant cluster and cannot make inferences about the statistical significance of specific locations within the cluster. This poses a particular problem when one uses a liberal cluster-defining primary threshold (i.e., higher p-values), which often produces large clusters spanning multiple anatomical regions. In such cases, it is impossible to reliably infer which anatomical regions show true effects. From a survey of 814 functional magnetic resonance imaging (fMRI) studies published in 2010 and 2011, we show that the use of liberal primary thresholds (e.g., p < .01) is endemic, and that the largest determinant of the primary threshold level is the default option in the software used. We illustrate the problems with liberal primary thresholds using an fMRI dataset from our laboratory (N = 33), and present simulations demonstrating the detrimental effects of liberal primary thresholds on false positives, localization, and interpretation of fMRI findings. To avoid these pitfalls, we recommend several analysis and reporting procedures, including 1) setting primary p < .001 as a default lower limit; 2) using more stringent primary thresholds or voxel-wise correction methods for highly powered studies; and 3) adopting reporting practices that make the level of spatial precision transparent to readers. We also suggest alternative and supplementary analysis methods.

Subfields: Cognitive Neuroscience, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/76fa9d010a94151202b0aa7ee3cfdc76065711f4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates (2015, positive): “We note that other authors have found RFT clusterwise inference to be invalid in certain settings under stationarity (21, 30) and nonstationarity (13, 33).”
The brain’s conversation with itself: neural substrates of dialogic inner speech (2015, methodology): “As the latter criterion has been recommended to avoid false positives (Woo et al., 2014), the results reported later are all significant at P 0.05 (corrected) based on a voxel-wise error of P 0.001, unless otherwise stated.”
Ten simple rules for neuroimaging meta-analysis (2017, background): “, 2013) and pushes researchers towards analyses and thresholding procedures that increase false positives (Eklund et al., 2016; Wager et al., 2007; Wager et al., 2009; Woo et al., 2014).”
The relation between statistical power and inference in fMRI (2017, methodology): “In addition, a common approach is to apply cluster-extend thresholds which is in general a more sensitive method than these voxel-wise threshold methods (see for instance threshold free cluster enhancement [47]), though certain common practices have recently also been debated [48,49].”
DPABI: Data Processing & Analysis for (Resting-State) Brain Imaging (2016, positive): “001 (Eklund et al. 2015; Woo et al. 2014), we have set voxel-level Z> 3.”
Behavior, Sensitivity, and power of activation likelihood estimation characterized by massive empirical simulation (2016, neutral): “Searches for studies investigating major brain disorders such as schizophrenia, depression, dementia, stroke or Parkinson’s disease likewise revealed many hundred clinical neuroimaging papers.”