The goal of the present experiment is to assess the effects of motor preparation on reaction time (RT) as a function of practice. We used a priming procedure to induce different levels of motor preparation. In reaction to a visual signal, the 16 participants had to respond with a flexion or an extension of the right or left wrist. Each participant performed 6120 successful trials divided into 34 experimental series. The first analysis of RT as a function of the type of cue (valid, neutral, or invalid) suggested that the time needed (a) to inhibit the prepared response and (b) to program the required one, remains stable throughout practice. A second analysis on the amount of information to be processed according to type of cue (0.42, 2.00, and 3.58 bits) indicated that central processing speed did not vary with practice. Indeed, only RT for the 0 bit condition decreased during the first 720 trials, suggesting that only the peripheral processes were influenced in early learning.