Fig. 1. Stimuli and tasks used to functionally localize the FEF (A) and MT+ (B) ROIs were presented in a block design during functional imaging runs. All functional runs began with 4 volumes of fixation immediately followed by 1 complete block of fixation. In addition, all functional runs ended with a block of fixation. Numbers under each line segment indicate the number of volumes within the block. Blocks in the FEF localization sequence were 8 volumes long and alternated between fixation and saccade. Within each run, there were 10 blocks of the fixation task and 8 blocks of the saccade task. Three types of blocks were used to localize MT+: fixation, still, and motion. Every block of the still and motion conditions lasted 10 volumes and was immediately followed by 1 block of fixation lasting 5 volumes. Blocks alternated between the still and motion conditions, with the fixation blocks interspersed between these.

Table 1. [Talairach and Tournoux, 1988] coordinates for centers of mass of individual and group functional ROIs

Note. Reported values were averaged across ROIs from separate localizer runs. Means are reported for the individual ROIs in atlas space with standard errors in parentheses.

Table 2. Proportion of overlapping voxels of functional ROIs, each defined with data from a single localization run

Note. Mean values and their standard errors (in parentheses) are reported for the FEF ROIs of 10 participants and MT+ ROIs of 9 participants. Individual data were removed if an ROI could not be identified in either run when the data were in either native or atlas space. Data removed due to a failure to find an ROI in native space were replaced for the individual to group comparison. This had a small effect on the mean and standard error of the atlas space FEF ROIs (M = 0.313, SE = 0.049) and MT+ ROIs (M = 0.219, SE = 0.177). The proportion of overlapping voxels was defined as the number of voxels common to both functional ROIs divided by the number of voxels in either ROI and reflects differences in both the location and extent of the ROIs.

Table 3. Volume of functional ROIs (cm³)

Note. Volume change and the volume of the overlapping portions of these ROIs are reported to further illustrate the consistency of each localization method. Reported values are in cubic centimeters, with means and standard errors for the individually defined ROIs in atlas and native space.

Table 4. Differences between the center of mass of individual ROIs in atlas space and the group ROIs

Note. Positive values indicate the group ROI was located to the left of (x) or posterior (y) or dorsal (z) to the individual centers of mass.

Table 5. Summary of results

Note. A plus sign (+) indicates the method that yielded the most advantageous results (the highest retest mean and proportion of overlapping voxels and the smallest distance). Localization methods that did not differ significantly from this method are indicated with a period. Localization methods that were significantly different (P < 0.05) are left blank. Some comparisons could not be performed due to too few observations (overlap MT+ comparison of group observations) or the inability to compare coordinates in atlas space to coordinates in native space (distance to group comparison).