The relationships between response latencies and accuracy on the matching familiar figures test (MFFT) and two gross motor tasks (batting or catching a ball) were studied in twenty-nine 9-year-old boys. Children were classified into four groups using a double dichotomy of response latencies and errors on the MFFT: reflective, impulsive, fast-accurate, and slow-inaccurate on intertask comparison. The components (errors and time) used to classify the children show stability for errors but not latencies on cognitive versus motor intertask comparison. The comparison between motor tasks shows the stability for latencies and accuracy, a nonlinear relationship between latency and accuracy for the ball-hitting but not the ball-catching task, and reflective boys to be the most efficient on the task requirement. These results lend support to the hypothesis that strategies are more a consequence of a ''competence'' than a ''conceptual tempo'' factor.