Abstract
We investigate the impact of decisions in the second-level (i.e., over subjects) inferential process in functional magnetic resonance imaging on (1) the balance between false positives and false negatives and on (2) the data-analytical stability, both proxies for the reproducibility of results. Second-level analysis based on a mass univariate approach typically consists of 3 phases. First, one proceeds via a general linear model for a test image that consists of pooled information from different subjects. We evaluate models that take into account first-level (within-subjects) variability and models that do not take into account this variability. Second, one proceeds via inference based on parametrical assumptions or via permutation-based inference. Third, we evaluate 3 commonly used procedures to address the multiple testing problem: familywise error rate correction, False Discovery Rate (FDR) correction, and a two-step procedure with minimal cluster size. Based on a simulation study and real data we find that the two-step procedure with minimal cluster size results in most stable results, followed by the familywise error rate correction. The FDR results in most variable results, for both permutation-based inference and parametrical inference. Modeling the subject-specific variability yields a better balance between false positives and false negatives when using parametric inference.
Highlights
In cognitive neurosciences, functional Magnetic Resonance Imaging plays an important role to localize brain regions and to study interactions among those regions The analysis of an fMRI time course in a single subject offers some insight into subject-specific brain functioning while group studies that aggregate results over individuals yield more generalizable results
We summarize different multiple testing strategies that are frequently exploited in the fMRI literature, such as approaches that control the familywise error rate, approaches for control of the False Discovery Rate, and a two-step procedure based on an uncorrected threshold but requiring a minimum cluster size
In this study we investigated both the balance between true positives (TP), true negatives (TN), false positives (FP), and false negatives (FN) and data-analytical stability of methodological choices in the second-level analysis of fMRI data
Summary
Functional Magnetic Resonance Imaging (fMRI) plays an important role to localize brain regions and to study interactions among those regions (resp., functional segregation and functional integration; see, e.g., [1]) The analysis of an fMRI time course in a single subject (first-level analysis) offers some insight into subject-specific brain functioning while group studies that aggregate results over individuals (second-level analysis) yield more generalizable results. We focus on the mass univariate approach in which the brain is divided in small volume units or voxels, alternatives exist (e.g., [2]). For each of these voxels, a general linear model (GLM) is used to model brain activation, at the first and the second level [3]. The selection of activated voxels can be viewed as a sequence of different phases [4]. For first-level analyses, Carp [5] demonstrated the large variation in the choices made in each of these different phases which impacts results. We consider the following phases in the analysis of group studies: (1) aggregation of data over subjects, (2) inference, and (3) correction for multiple testing
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
