The effect of cerebellar cortical degeneration on adaptive plasticity and movement control

Clinical and neuroimaging studies provide converging evidence that the cerebellum plays an important role for sensorimotor adaptation by participating in the adaptive process per se, and/or by evaluating motor performance errors as a prerequisite for adaptation. Recent experimental evidence suggests that error signals pertinent to adaptation are related to sensory prediction rather than to online corrections (Tseng et al. in J Neurophysiol 98(1):54-62, 2007). To further elucidate the role of the cerebellum, the present study uses a multiple regression approach to separate out three independent determinants of adaptive success. Seventeen patients with cerebellar atrophy but without extra-cerebellar lesions, and 17 healthy, sex- and age-matched controls participated. Both subject groups performed center-out pointing movements before, during, and after exposure to 60 degrees rotated visual feedback. From the registered data, we quantified four indicators of adaptive success (adaptive improvement, retention without feedback, intermanual transfer, and de-adaptation under normal feedback), as well as five measures of motor performance (reaction time, peak velocity, movement time, response variability, and ability for online error corrections). The variance of each adaptation indicator was then partitioned into three components, one related to subject group but not to motor performance, a second related to group and motor performance, and a third related to motor performance but not to group. In accordance with previous work, adaptation and motor performance were degraded in patients. The deficit was similar in magnitude for all four adaptation indicators, which suggests that adaptive recalibration rather than strategic control were affected in our patients. No adaptation indicator shared statistically significant variance with group alone; we therefore found no evidence for cerebellar circuitry dedicated to adaptation but not motor performance. Three indicators shared significant variance jointly with group and motor performance; this suggests that the cerebellar contribution to motor performance is related to adaptive success. All four indicators shared significant variance with motor performance alone; this indicates that extracerebellar contributions to motor performance are also related to adaptive success. In conclusion, our data support the view that neural structures inside and outside the cerebellum are processing motor performance-related signals as a prerequisite for adaptation, but provide no evidence for a cerebellar structure related exclusively to adaptation.