Abstract
The time required to begin movement toward a panel of lights and switches of two different complexity levels was investigated in both a between-subjects and within-subjects design. In the between-subjects design, the mean RT on the subjects’ first trial was longer to the more complex task, supporting the memory-drum theory of Henry and Rogers (1960). In the within-subjects design, a significant complexity x order x blocks interaction showed that the subjects receiving the less complex task first responded differently than subjects receiving the more complex task first. Implications for experimental design in investigations of task complexity are discussed.