Search

It is widely recognized that mental rotation is a cognitive process which engages a distributed cortical network including the frontal, premotor and parietal regions. Like other visual-spatial transformations it could require operations on both metric and categorical spatial representations. Previous reports have implicated respectively the right hemisphere being involved in the metric processing and the left hemisphere in the categorical processing. By using a modified version of the , we attempted to establish the cortical regions relevant for the categorical and metric aspects of mental rotation transformations. Two groups of patients were found to be impaired in our study, namely the left prefrontal and the right parietal. In particular, whereas the right parietal group made poor use of categorical information, the left prefrontal patients showed a broader mental rotation impairment with a significant number of metric errors. The results are discussed in terms of the model of about the possible mental transformation impairments following brain lesions.

It is standardly believed that the localisation of cognitive function by means of impairments arising from cortical tumour is not possible as the functional defects that result are mild and unspecific. These assumptions were not supported in an investigation of four processes generally sensitive to right posterior cortical lesions, when patients with parieto-occipital lesions were compared with prefrontal ones. In three of the tests loading on the individual processes – Reaching Accuracy, Star Cancellation, Fragmented Letters and Cube Analysis – parieto-occipital impairments were found in the basic groups analysis and this was so in the right-hemisphere group. More critically, in these tests Lesion Behaviour Mapping showed the critical lesion site for the tests to have relatively little overlap with those of the other tests, indicating that the cognitive effects were not widespread and diffuse. In addition, in three of the tests the critical lesion sites fitted localisations arrived from other procedures. Patients with high-grade tumours performed considerably worse than those with low-grade tumours in only two of the tests (Star Cancellation, Cube Analysis) particularly in the right parieto-occipital group. In three (Reaching Accuracy, Star Cancellation, Cube Analysis) there was a deterioration with the operation specifically in the low-grade tumour patients. It is suggested that a tumour patient series may provide converging evidence for the localisation of a function initially obtained by some other procedure.