A single sample t-test was conducted to indicate voxels responding to SWM relative to vigilance across all participants (n = 72). To control for Type I error, a combination of t-statistic magnitude and cluster volume thresholding was applied (Forman et al., 1995; Ward, 1997) by only interpreting clusters comprised of at least 22 contiguously activated voxels at a < 0.025 (≥943 μl in volume). Eleven clusters surpassed this threshold. The largest cluster (342,486 μl) was reduced into 9 smaller regions of interest defined using regional proximity, established correlates of brain structure to function (Heilman and Valenstein, 1972, 2003), and the direction of BOLD activation (i.e., areas activating preferentially to SWM were grouped separately from those activating to vigilance; see Fig. 1). For each of the resulting 19 clusters (see Table 3), average activation for each participant was converted into z-scores.
