ALE meta-analysis: controlling the false discovery rate and performing statistical contrasts

Abstract

Activation likelihood estimation (ALE) has greatly advanced voxel-based meta-analysis research in the field of functional neuroimaging. We present two improvements to the ALE method. First, we evaluate the feasibility of two techniques for correcting for multiple comparisons: the single threshold test and a procedure that controls the false discovery rate (FDR). To test these techniques, foci from four different topics within the literature were analyzed: overt speech in stuttering subjects, the color-word Stroop task, picture-naming tasks, and painful stimulation. In addition, the performance of each thresholding method was tested on randomly generated foci. We found that the FDR method more effectively controls the rate of false positives in meta-analyses of small or large numbers of foci. Second, we propose a technique for making statistical comparisons of ALE meta-analyses and investigate its efficacy on different groups of foci divided by task or response type and random groups of similarly obtained foci. We then give an example of how comparisons of this sort may lead to advanced designs in future meta-analytic research.

Figure 1

Comparison of different thresholding techniques. Meta‐analyses of overt speech in stuttering subjects (a), the Stroop task (b), picture naming (c), and painful stimulation were carried out (d) and the resultant ALE maps were thresholded in three different ways at P < 0.05: (i) using uncorrected P values; (ii) using P values that were corrected using the single threshold test that generates the null distribution for the maximal ALE statistic; and (iii) FDR‐corrected P values. Three contiguous slices are presented for each meta‐analysis result: z = 28–32 mm for stuttering; z = 40–44 mm for Stroop; z = −16 to −12 mm for picture naming; and z = 6–10 mm for painful stimulation (n, number of foci)

Figure 2

Statistical comparisons of ALE maps. ALE meta‐analyses (P < 0.05) for different groups of foci were carried out and are presented here as overlays of separate meta‐analyses (i) or as comparison meta‐analyses computed as a single ALE image (ii). Meta‐analyses carried out were for overt speech (a) in stuttering subjects (yellow) and controls (blue), the Stroop task (b) with a verbal (yellow) and manual (blue) responses, picture‐naming tasks (c) with a silent (yellow) or speaking (blue) baselines, and painful stimulation to the right (d) (yellow) and left (blue). Green indicates overlap between the individual meta‐analyses (blue + yellow). Three contiguous slices are presented for each meta‐analysis result: z = 28–32 mm for stuttering; z = 40–44 mm for Stroop; z = −16 to −12 mm for picture naming; and z = 6–10 mm for painful stimulation. (n, number of foci).