# SDM exploration of the results

## Multimodal meta-analyses

This procedure allows the combination of two meta-analyses of studies conducted in different modalities, e.g. VBM and fMRI.

### To conduct a multimodal meta-analysis

Conduct the two unimodal (i.e. standard) meta-analyses separately.

Press the button [Multimodal]

*or*

Select [Multimodal] in the *Statistics* menu, to open the following dialog:

*or:*

### Command-line and batch usage

multi *probability map1*,
*probability map2*,

. . . *maximum voxel probability*,
*maximum peak probability*,
*minimum number of voxels*,

. . . *assume error (0 or 1)*,
*best scenario (0 or 2)*,
*tolerable meta-analytic probabilities*

Assuming error ('1') makes the analysis conservative, for what you may consider adjusting the p-values to the best scenario ('2'). This latter option means that the p-values will be more liberal, but the analysis is STILL an overlap (please see Radua et al 2013 for details). To avoid being too liberal, specify the tolerable p-values in the unimodal meta-analyses.

*Example:*

multi meta1, meta2, 0.00250, 0.00025, 10, 1, 2, 0.10000

Please note that you should have previously creted a folder and copied there "meta1" (a map of probabilities from the first meta-analysis) and "meta2" (a map of probabilities from the second meta-analysis).

### References

(multimodal meta-analyses):
Radua J, Romeo M, Mataix-Cols D *and* Fusar-Poli P. A general approach for
combining voxel-based meta-analyses conducted in different neuroimaging
modalities.
*Curr Med Chem* 2013; 20:462-466.
.