Finally, to allow comparison across participants and to eliminate the effect of grossly different RTs, a change score was calculated for each block of practice (RT change score=mean median RT from the second block of random sequence practice on day 1 minus mean median RT from repeating sequence blocks 1–15).