The second control group did not perform any task but waited for the equivalent duration of the car racing tasks between the two MRI scans. Because the active control group played a different track in each trial (different lengths and scenery), evaluating

their improvement during the task necessitates normalization. This normalization was crucial because the tracks were randomized ATM Kinase Inhibitor purchase between active control subjects. The normalization procedure, performed for each subject, included normalizing the lap time to the track length and dividing by the performance in the first trial. The same procedure was applied to the learning group. MRI was performed at the Tel Aviv Sourasky Medical Center with a 3T (GE, Milwaukee, WI, USA) MRI system. All subjects underwent two series of scans approximately 2 hr apart. Between the two sessions a task was administered to the learning group and the first control group; the second control group

did not perform any task. The MRI protocol of the first series of scans included conventional anatomy sequences, and DTI was acquired with an eight-channel head coil. In the second series only DTI scans were administered. T1-weighted images were acquired with a 3D spoiled gradient-recalled echo (SPGR) sequence with the following parameters: up to 155 axial slices (whole-brain coverage), TR/TE = 9/3 ms, resolution 1 × 1 × 1 mm3, scan time 4 min. In addition to the T1 scan, T2-weighted images (TR/TE = 6,500/85) and FLAIR images (TR/TE/TI = 9,000/140/2,100) were acquired. The entire anatomical data set was used for radiological screening. Double-refocused, find more of spin-echo diffusion-weighted,

echo-planar imaging sequences were performed with up to 70 axial slices (to cover the whole brain), and resolution of 2.1 × 2.1 × 2.1 mm3 was reconstructed to 1.58 × 1.58 × 2.1 mm3 (field of view was 202 mm2, and acquisition matrix dimension was 96 × 96 reconstructed to 128 × 128). Diffusion parameters were Δ/δ = 33/26 ms; b value of 1,000 s/mm2 was acquired with 19 gradient directions, and an additional image was obtained with no diffusion weighting (b0 image). The double-refocused sequence was used in order to minimize eddy currents and susceptibility artifacts. The DTI scan was repeated three times to increase signal-to-noise ratio. For details on the DTI analysis routine, please refer to section 1.2 (Image analysis) of the Supplemental Experimental Procedures. VBA is a whole-brain technique that allows regionally specific differences in quantitative MRI indices (such as FA or MD) to be computed on a voxel-by-voxel basis. The statistical VBA design included three groups (learning and controls) and two scan times (with repeated measures on the second factor). On this design we applied the following procedures. (1) A paired t test on the learning group only (comparing the pre and post-learning scans). To avoid partial volume bias in the statistical analysis, we applied a non-cerebrospinal fluid (CSF) mask.