Table 6.

Experiment 4: mean percentages of thresholded task-related GLM maps that are covered by the thresholded subject-specific FC maps generated with each hybrid method below. The right-most column gives percentages after adjusting the z-score used to threshold the FC maps so that number of false positives equals number of false negatives. In contrast with Experiment 3, the results for the SV method were significantly lower than those for the FV, MV, and DRS methods. The difference between Experiments 3 and 4 was that Experiment 4 was based on a group ICA that produced 5 components rather than 14. As explained in the text, the differences in results between these two experiments can be accounted for by subtle differences between the seed-source ICs, which led to markedly different brain locations for the single-voxel seed.

Method of Generating Maps	Group ICA Source	Percent of Task-Related Map Covered (95% Confidence Interval)
		Before Threshold Adjustment	After Threshold Adjustment
Single-voxel seed (SV)	PICA	15 (6–24)	25 (15–35)
9-voxel ROI seed (FV)	PICA	47 (35–59)	54 (43–65)
40-voxel ROI seed (MV)	PICA	57 (45–69)	57 (46–68)
Dual regression, 1 spatial map (DRS)	PICA	46 (33–60)	44 (32–56)
Dual regression, all spatial maps (DRA)	PICA	33 (22–44)	40 (29–50)