How efficient are central mechanisms for the learning and retention of coincident timing actions?

We compared the adaptive strategy and retention capacity of a deafferented subject and control subjects when intercepting, with a sliding-throw, an apparent movement coming at various speeds. Subjects were submitted to five practice sessions (30 trials per session) and to a retention test. The throwing kinematics was analysed, and spatial and temporal performance errors were measured. With practice, the deafferented subject showed modifications in movement initiation strategies and throwing patterns. With a slow apparent movement, the deafferented subject's initial behavior was characterized by short movement initiation and movement times. With practice, she showed an important increase in movement time in session 5, allowing longer visual control and leading to better temporal and spatial accuracy than that shown in session 1. In the retention session, the deafferented patient showed a late movement initiation strategy, similar to that of the control subjects. This increased movement initiation time was accompanied by an improved temporal accuracy compared to the deafferented subject's early results. However, spatial accuracy improvement was labile and could not be maintained over the retention interval. At the fast speed, all temporal components of the response, namely, movement initiation time (MIT), movement time (MT), and disk travel time (DTT), were similar for the deafferented and control subjects. Overall, the deafferented subject reduced her temporal error through practice, though without attaining the control subjects' accuracy. However, with a fast-moving stimulus, she showed a deteriorated spatial accuracy, even doubling her spatial errors at retention. In brief, the deafferented subject achieved proper temporal (perceptivo-cognitive) lasting control of her interceptive action, whereas spatial (sensorimotor) regulation raised mnemonic problems.