This paper presents two improvements to the Activation Likelihood Estimation (ALE) method for voxel-based meta-analysis in functional neuroimaging. First, it evaluates two techniques for correcting multiple comparisons: the single threshold test and a procedure that controls the false discovery rate (FDR). The FDR method is found to more effectively control false positives in meta-analyses of both small and large numbers of foci. Second, the paper proposes a technique for statistically comparing ALE meta-analyses and investigates its effectiveness on different groups of foci divided by task or response type and random groups of similarly obtained foci. An example is given of how such comparisons may lead to advanced designs in future meta-analytic research. The ALE method pools 3-D coordinates from multiple studies and spatially normalizes them to a common template. Each reported coordinate is modeled by a 3-D Gaussian distribution, and the ALE statistic is calculated based on the probability of a focus being located in a given voxel. A permutation test is used to assess the significance of the results, generating a null distribution of the ALE statistic by randomly generating foci and computing their ALE values. The paper discusses two methods for correcting multiple comparisons: the single threshold test and the FDR method. The single threshold test uses a single threshold to control the family-wise error rate (FWE), while the FDR method controls the expected proportion of false positives among rejected voxels. The FDR method is found to be more powerful and less conservative than the single threshold test. The paper also presents a statistical comparison technique for ALE maps, which involves subtracting the ALE values of two groups of foci to measure the difference in convergence between the maps. This technique is used to compare different groups of foci and to assess the significance of differences between them. The study demonstrates that the FDR method is more effective than the single threshold test in controlling false positives and that the statistical comparison technique provides a reliable method for testing differences between ALE meta-analyses. The results show that the FDR method provides a better balance between controlling Type I error and maximizing sensitivity. The paper concludes that the ALE method, with its improvements in multiple comparison correction and statistical comparison, is a valuable tool for meta-analysis in functional neuroimaging.This paper presents two improvements to the Activation Likelihood Estimation (ALE) method for voxel-based meta-analysis in functional neuroimaging. First, it evaluates two techniques for correcting multiple comparisons: the single threshold test and a procedure that controls the false discovery rate (FDR). The FDR method is found to more effectively control false positives in meta-analyses of both small and large numbers of foci. Second, the paper proposes a technique for statistically comparing ALE meta-analyses and investigates its effectiveness on different groups of foci divided by task or response type and random groups of similarly obtained foci. An example is given of how such comparisons may lead to advanced designs in future meta-analytic research. The ALE method pools 3-D coordinates from multiple studies and spatially normalizes them to a common template. Each reported coordinate is modeled by a 3-D Gaussian distribution, and the ALE statistic is calculated based on the probability of a focus being located in a given voxel. A permutation test is used to assess the significance of the results, generating a null distribution of the ALE statistic by randomly generating foci and computing their ALE values. The paper discusses two methods for correcting multiple comparisons: the single threshold test and the FDR method. The single threshold test uses a single threshold to control the family-wise error rate (FWE), while the FDR method controls the expected proportion of false positives among rejected voxels. The FDR method is found to be more powerful and less conservative than the single threshold test. The paper also presents a statistical comparison technique for ALE maps, which involves subtracting the ALE values of two groups of foci to measure the difference in convergence between the maps. This technique is used to compare different groups of foci and to assess the significance of differences between them. The study demonstrates that the FDR method is more effective than the single threshold test in controlling false positives and that the statistical comparison technique provides a reliable method for testing differences between ALE meta-analyses. The results show that the FDR method provides a better balance between controlling Type I error and maximizing sensitivity. The paper concludes that the ALE method, with its improvements in multiple comparison correction and statistical comparison, is a valuable tool for meta-analysis in functional neuroimaging.