Recall performance of the Rey-Osterrieth Complex Figure was examined in patients with partial complex seizures originating from either the right or left temporal lobe and who underwent subsequent unilateral temporal lobectomy. A scoring system was developed to assess the types of errors frequently observed in the recall of patients with right temporal lobe epilepsy (TLE), but absent in left TLE patients. The scoring system was initially developed on a single group of patients, and then “cross-validated” on an independent sample. Performance analysis of the cross-validation group revealed a significant difference in the frequency of right hemisphere errors. In contrast, no significant difference using standard quantitative scoring was present. By applying the new scoring critiera alone, a rater blind to seizure onset correctly predicted seizure laterality in 15/18 of the cross-validation patients. These results suggest that evaluation of qualitative errors may be a valuable adjunct to standard scoring criteria, thereby extending the range of applications for this test.