Table 2:

Statistical table for significance contours in Figures 3, 4, and 7

IdentifierMinimum number of observationsMaximum number of observationsLargest cluster size (pixels)
d125 sessions25 sessions1137
d223 sessions23 sessions643
d325 sessions25 sessions55
d713 sessions13 sessions123
d822 sessions25 sessions612
d925 sessions26 sessions378
d1023 sessions23 sessions214
d1126 sessions26 sessions217
d1225 sessions27 sessions250
d1326 sessions27 sessions363
p70 cells114 cells3
  • With the exception of entry p (corresponding to Fig. 7D), all contours are calculated on the basis of one-sample t tests at an initial threshold of p < 0.001, with cluster correction at a FWER of p < 0.01. Within a significance contour, the number of sessions or cells used to calculate the mean effect at any given pixel may be less than the total number of sessions or cells in the study. This can occur because data in some pixels in some sessions are removed (set to nan), as a quality control measure (see Materials and Methods). For each contour region, the columns for the Maximum and Minimum number of observations therefore give the number of observations for the pixel with the smallest and largest number used, respectively