[Transformation of kinematic characteristics of a precise movement after change in a spatial task]

Brain mechanisms of motor programming were studied with the use of the model of learning precise horizontal elbow flexion. To exclude visual control and make learning to be based, predominantly, on proprioception, experiments were carried out in darkness. The target position was not demonstrated beforehand. Subject (S) had to find an adequate angle of flexion during training with a short light-diode flash which marked the moment of target reaching. Ss were asked to perform a precise horizontal elbow flexion as fast as possible. Movement amplitude, velocity and acceleration were on-line recorded. Ten Ss were divided in two groups. The first group was initially trained to make the precise movement with the preset amplitude of 70 degrees and the second group had to perform similar movement with the amplitude of 55 degrees. Each S was trained to the moment of acquisition of a stable skill (within the 5% error of preset flexion amplitude). After that the target position was unex pectedly changed (from 70 for 55 degree or visa verse). This work was a continuation of our earlier search for a mathematical hypothesis most correctly explaining the central mechanism of motor learning. The dynamics of kinematic characteristics of learning in our experiments fitted well to A. Barto and J. Houk's "Cerebellar Model of Timing and Prediction". A comparison of a computer simulation of this model to the learning characteristics allowed us to make some refinements of the model very important for data analysis possible under conditions of noninvasive investigations. The analysis of acceleration dynamics not considered by the authors of the model made it possible to identify this index with the "pulse phase" similar to the period of LTD of Purkinje cells (the key mechanism of the model). We took such an interpretation as principal in our analysis of experimental data. We analyzed integrals of positive and negative acceleration which made it possible to gain a deeper insight into the physiological mechanism of a replacement of one central command by the other as a consequence of change in spatial task conditions.