See Fig. 2 for a lesion overlap map for our eleven cases (the extent and location of each patient’s lesion was defined and visualized using the MRIcro software package Rorden and Brett, 2000; lesions were plotted on 12 axial slices of the T1-weighted template MRI scan from the Montreal Neurological Institute – MNI). All our patients showed neglect
on clinical paper-and-pencil measures including the Mesulam cancellation check details test, a 5-item line bisection task, figure copying and drawing from memory. Diagnosis of left visual neglect involved the fulfillment of at least two of the following criteria: the presence of a minimum 30% omissions on the left side of the page for the cancellation test; a minimum rightward deviation of 12% or more in the line bisection task; omission of left sided elements in the figure copying task; omission of left sided elements in the drawing from memory task. Five out of eleven patients (EY, AK, BH, PH, MM and LG) also presented with complete left homonymous hemianopia as tested on confrontation. See Table 1 for a summary of individual patient details and scores on some paper-and-pencil tasks. Three of these patients ABT 199 (AK, EY and CO) had already
taken part in our previous study (Sarri et al., 2006), but were retested here for the chimeric expression lateral preference task, after a minimum interval of at least one month between testing sessions, to allow within-session comparison with the other tasks. All patients participated in the emotional expressions and the greyscale gradients lateral preference tasks. However, only six patients (EH, AM, PH, EY, LG and MK) were able to participate in the chimeric/non-chimeric face discrimination task. All other patients were excluded from this task as they were found to perform at ceiling-level in this prior to Ergoloid prism adaptation. Please note that in the present study, each patient served as his/her own control (i.e., before versus after prism therapy). For the chimeric face tasks, 20 pairs of chimeric face tasks were used, adapted from Mattingley et al.
(1993). These chimeric face tasks were generated from 10 pictures of 10 different people with a neutral expression, plus 10 pictures of those same people smiling. The photographed faces were divided along the vertical midline, and left and right halves from different photographs of the same person were then juxtaposed in such a way that a smiling half face was on the left and a neutral half face on the right; or vice versa in mirror-image displays. Each chimeric face task subtended approximately 6° × 8°. Chimeric face stimuli were then arranged in vertical pairs, one above the other, so that each pair contained two chimeras of the same person, one neutral in the left half and smiling in the right half, and the other the reverse of this, with vertical position counterbalanced. Thus, the two stimuli arranged vertically were left/right mirror images of each other; see Fig. 3A for examples.