This study extends previous research examining the locus of the contextual interference effect for tasks with similar relative timing. To learn any task both relative timing and overall duration characteristics of the movement must be mastered. It has been suggested that in general relative timing will be mastered prior to overall duration (Schmidt, 1975; Smith & Penn, 1999; Wright & Shea, 2001). Smith & Penn (1999) found that when only overall duration was varied during practice, this resulted in a contextual interference effect in overall duration error emerging only after a substantial period of practice. In relative timing error, Smith & Penn found a consistent contextual interference effect regardless of the amount of practice. It was suggested that introducing contextual interference in overall duration forces learners to focus attention on the relative timing of the task, which improves performance on that dimension of the task relative to low contextual interference, but impedes overall duration accuracy (Wulf & Lee, 1993). The purpose of this study was to examine whether practice length also moderates the effect of contextual interference among tasks requiring the different relative timings. Participants (n=30) practiced variations of a three segment timing task for 135 trials per day for two days, in either a blocked or serial order. Each practice phase was followed 24 hr. later with a 30 trial retention phase. The three tasks varied in their relative timing goal, but required the same overall duration. Total error (E) was derived for measures of relative timing and overall duration error, across both acquisition and retention phases. The most critical results were in retention, where no interactions involving practice length were found for either relative timing or overall duration. Additionally, there was no contextual interference effect for overall duration regardless of the length of practice, but there was a significant contextual interference effect for relative timing, F(1, 27)=6.66, p < .05, with serial practice resulting in superior performance throughout retention. Apparently, high contextual interference facilitates learning of several different relative timing patterns concurrently, regardless of the amount of practice given. This seems problematic for previous explanations of the locus of the effect (e.g. Wright & Shea, 2001), as it has been suggested that an early, undisturbed focus on a constant relative timing pattern is beneficial for learning. Instead these results suggest that maximizing interference of the learner's focus on this property of the tasks maximized skill retention.Keyword(s): coaching, research