Figure 1.
Examples of stimuli for the Array tests.
For faces (upper left), houses (upper right), phones (bottom left) and words (bottom right). A target was present in the four presented examples.
Figure 2.
Examples of stimuli for the Cambridge Memory Test with houses.
Panel A shows study views of a target house. Panel B displays a test item from the introduction. The central image is identical to the leftmost study view in Panel A. Panel C shows an item from the novel image section (the rightmost image is the target). Panel D displays a test item from the novel images with noise section (the leftmost image is the target).
Figure 3.
Examples of stimuli for the Cambridge Memory Test with phones.
See legend of Figure 2.
Figure 4.
Vascular topography of strokes, and overlap of the 31 lesions.
Vascular topography of strokes (left panel A), and overlap of the 31 lesions in MNI space (slices are TC z = −33, −15, −5, 4, 16, 24; right panel B).
Figure 5.
Results of the Array tests for faces, houses, phones and words.
In controls (n = 41), left stroke patients (n = 15), right stroke patients (n = 13) and bilateral stroke patients (n = 3). Error bars represent +/−1 S.E.M.
Figure 6.
Results of Cambridge Memory tests with faces, houses and phones.
In controls (n = 41), left stroke patients (n = 15), right stroke patients (n = 13) and bilateral stroke patients (n = 3). Error bars represent +/−1 S.E.M.
Figure 7.
In controls (n = 41), left stroke patients (n = 15), right stroke patients (n = 13) and bilateral stroke patients (n = 3). Error bars represent +/−1 S.E.M.
Figure 8.
Voxel-based lesion-symptom mapping (VLSM, right panel), comparing the performance of patients with and without a lesion of each voxel, with voxelwise Z statistics corrected for multiple comparisons over the whole brain (threshold: FDR p<0.05). Significant regions are contiguous to the peaks of the corresponding functional areas as identified in normal subjects (red crosshair). Addition of normalized lesions (left panel) for patients with a deficit for words (n = 3; top panel), faces (n = 9; middle panel), and houses (n = 10; bottom panel), minus the lesions of patients with no deficit in the considered category. The regions of maximum overlap are essentially identical to those identified using VLSM.