Lesion-symptom mapping of the human cerebellum.

High-resolution structural magnetic resonance imaging (MRI) has become a powerful tool in human cerebellar lesion studies. Structural MRI is helpful to analyse the localisation and extent of cerebellar lesions and to determine possible extracerebellar involvement. Functionally meaningful correlations between a cerebellar lesion site and behavioural data can be obtained both in subjects with degenerative as well as focal cerebellar disorders. In this review, examples are presented which demonstrate that MRI-based lesion-symptom mapping is helpful to study the function of cerebellar cortex and cerebellar nuclei. Behavioural measures were used which represent two main areas of cerebellar function, that is, motor coordination and motor learning. One example are correlations with clinical data which are in good accordance with the known functional compartmentalisation of the cerebellum in three sagittal zones: In patients with cerebellar cortical degeneration ataxia of stance and gait was correlated with atrophy of the medial (and intermediate) cerebellum, oculomotor disorders with the medial, dysarthria with the intermediate and limb ataxia with atrophy of the intermediate and lateral cerebellum. Similar findings were obtained in patients with focal lesions. In addition, in patients with acute focal lesions, a somatotopy in the superior cerebellar cortex was found which is in close relationship to animal data and functional MRI data in healthy control subjects. Finally, comparison of data in patients with acute and chronic focal lesions revealed that lesion site appears to be critical for motor recovery. Recovery after lesions to the nuclei of the cerebellum was less complete. Another example which extended knowledge about functional localisation within the cerebellum is classical conditioning of the eyeblink response, a simple form of motor learning. In healthy subjects, learning rate was related to the volume of the cortex of the posterior cerebellar lobe. In patients with focal cerebellar lesions, acquisition of eyeblink conditioning was significantly reduced in lesions including the cortex of the superior posterior lobe, but not the inferior posterior lobe. Disordered timing of conditioned eyeblink responses correlated with lesions of the anterior lobe. Findings are in good agreement with the animal literature. Different parts of the cerebellar cortex may be involved in acquisition and timing of conditioned eyeblink responses in humans. These examples demonstrate that MRI-based lesion-symptom mapping is helpful to study the contribution of functionally relevant cerebellar compartments in motor control and recovery in patients with cerebellar disease. In addition, information about the function of cerebellar cortex and nuclei can be gained.