Switching between tasks is an essential factor involved in everyday life. Early research by Jersild (1927) identified a time cost in switch performance when compared to blocked performance. Studies showed that the time cost of switching persists even when the task is well known and the individual is prepared for that switch (Meiran, 1996). Allport (1994) interpreted the time cost as Task Set Inertia (TSI), a form of proactive interference between conflicting S-R mappings for successive tasks. Then question remains whether the strength of the TSI effect would persevere through practice, or if practice would influence the switch cost, where the time detriment may be ameliorated. The present study involved two key pressing tasks (Task A and Task B). Performances on the two tasks were evaluated in terms of practice, using both reaction time (RT) and movement time (MT) measures. Two pre-switch schedules were adopted, pre-switch one (PS1) and pre-switch nine (PS9). PS1 involved 1 trial of task A followed by 9 trials of task B, with PS9 administering 9 trials of task A prior to 9 trials of task B. Subjects were split into 4 groups, two groups were administered the pre-switch schedules before practice with the other groups having the pre-switch schedules administered after 72 trials of practice. The random switch practice trials were constrained by no repeated selection of 3 trials of task A or B, with the last trials of practice always being task B. Total amount of practice for both tasks was equal across groups. Performance on the 9 trials of Task B was used to examine the effect of switching from Task A. Results demonstrated RT and MT for both PS1 and PS9 late in practice were significantly lower than early in practice (p<0.001). This indicates practice of both tasks may benefit both response planning and execution for a motor task. The amount of pre-switch trials demonstrated PS9 had lower RT and MT both early and late in practice than PS1. An interesting "task verification" phenomenon occurred during the middle of the task B trials in both RT and MT. The findings indicate that executive control processes rather than TSI as suggested by Allport may govern switch costs. Extended pre-switch trials make subjects more alert to the potential switch of the task. Therefore, switch strategies may be ameliorated by random switch practice, and the extended pre-switch helps the executive control on the task switching.Keyword(s): exercise/fitness, performance, play