In the table below I check if the "reliable" peaks that were identified by Casey et al (Casey, B.J., Cohen, J.D., Craven, K.O., Davidson, R., Nelson, C.A., Noll, D.C., Hu, X., Lowe, M., Rosen, B., Truwit, C., & Turski, P. (1998). Reproducibility of fMRI results across four institutions using a spatial working memory task. NeuroImage, 8, 249-261.) are really reliable. For that, I list the peaks on the left, and for each peak find in each of the individual sites the closest peak that was found in this site for this condition.
In each line, the first three numbers are the coordinate s (X,Y,Z) of a "reliable" peak from table 1. The rest of line contains the closest peak from each site, followed by the distance of the peak from the "reliable" peak (rounded to integers). All numbers are in mm. Each column is headed by the table from which the numbers were taken (tables 4c, 4d do not appear in the paper).
To be regarded as the same peak, the distance needs to be less than a cm (i.e. less than 10). As can be seen in the table, only one "reliable" peak is close to be reproduced by two of the sites (marked with '*'). All the other peaks are reproduced by either one site or none at all.
Even if you allow peaks to diverge by 2 cm, which is a huge distance in the cortex, only four peaks are reproduced by three out of four of the sites (marked with '+'). Thus even at the resolution of two cm these peaks are not reliable at all. The correlation between each site and the "reliable" peaks seem to be close to 0.
Note that we should expect quite significant correlations between the individual sites and the "reliable" peaks, because some of the tests they use are known to be quite localized in the cortex (visual input in the visual areas, Motor action in the motor areas).
Memory vs Motor
Table 1 Table 2a Table 2b Table 2c Table 2d
12,57,2 32,43,13 [27] -37,38,21 [56] 7,63 ,15 [15] 3 ,50 ,0 [12] -
-2,58,4 32,43,13 [20] -37,38,21 [44] 7,63 ,15 [15] 3 ,50 ,0 [10]
4,57,17 32,43,13 [32] -37,38,21 [45] 7,63 ,15 [7] 3 ,50 ,0 [18]
38,25,35 42,29,17 [19] 26,-2,46 [32] 23,19 ,34 [16] 32 ,15 ,56 [24]
-44,16,41 -42,18,48 [8] -31,3,46 [19] -25,19 ,41 [19] -30 ,-6 ,43 [26] +
35,15,7 36,18,1 [7] 26,-2,46 [43] 39,-13 ,14 [29] 42 ,6 ,5 [12]
53,-27,36 48,-29,48 [13] 26,-2,46 [38] 39,-13 ,14 [30] 57 ,-51 ,34 [24]
0,-43,28 5,-41,56 [29] 2,-43,24 [4] -1,-33 ,29 [10] -4 ,-50 ,10 [20] *
36,-46,62 19,-56,54 [21] -11,-51,45 [50] 23,-39 ,76 [20] 36 ,-58 ,59 [12]
-24,-62,38 -39,-40,60 [35] -11,-51,45 [18] -49,-60 ,29 [27] -25 ,-60 ,54 [16]
-15,-66,41 5,-41,56 [35] -11,-51,45 [16] -10,-47 ,62 [29] -25 ,-60 ,54 [17]
Motor vs Rest
Table 1 Table 3a Table 3b Table 3c Table 3d
2,31,38 -19,13,40 [28] 2,-7,62 [41] -1,25 ,17 [22] 24 ,37 ,17 [31]
-3,30,40 -19,13,40 [23] 18,-2,62 [41] -1,25 ,17 [24] 24 ,37 ,17 [32]
-1,4,41 -19,13,40 [20] 2,-7,57 [20] -1,25 ,17 [32] 9 ,-1 ,42 [11] +
50,-7,30 22,-12,58 [40] 37,-13,52 [26] 55,11 ,42 [22] 43 ,14 ,10 [30]
-44,-10,46 -19,13,40 [34] -43,-23,37 [16] -1,25 ,17 [63] -49 ,-13 ,33 [14]
5,-12,60 22,-12,58 [17] 2,-7,57 [6] 23,-14 ,63 [18] 9 ,-1 ,42 [21] +
-3,-4,53 12,3,58 [17] 2,-7,57 [7] 23,-14 ,63 [30] 9 ,-1 ,42 [17] +
17,-43,69 33,-34,60 [20] 37,-13,52 [40] 39,-35 ,61 [25] 39 ,-44 ,50 [29]
-25,-37,70 22,-12,58 [55] -37,-42,53 [23] 23,-14 ,63 [54] -22 ,-35 ,73 [5]
-4,-57,0 2,-69,6 [15] -20,-48,-10 [21] -17,-55 ,-25 [28] -1 ,-53 ,-25 [25]
-28,-83,-4 -22,-74,-12 [12] -20,-48,-10 [36] -17,-55 ,-25 [37] -10 ,-78 ,31 [40]
Visual vs Rest
Table 1 Table 4a Table 4b
39,-78,8 -5,-70,9 [45] 39,-80,10 [3]
-25,-81 ,1 -5,-70 ,9 [24] -10,-90, 34 [26]
12,-83,26 -11,-74,33 [26] 7,-88,31 [9]
-6,-92 ,1 -5, -70 ,9 [23] -10,-90 ,34 [20]
4,-67,2 -5,-70,9 [12] 13,-96,-2 [31]