This study was designed to investigate the effects of two instructional approaches (whole-task versus part-task) and two levels of learner prior knowledge (lower versus higher) on learner acquisition and transfer of a complex cognitive skill. Participants were 51 undergraduate pre-service teachers. In the part-task condition, a complex skill (preparing a grade book using Excel) was decomposed into a series of smaller tasks, each of which was demonstrated and practiced separately. In the whole-task condition, which was based on the 4C/ID-model (van Merriënboer 1997), learners were exposed to the entire complex skill from the beginning of the instruction and were required to practice performing a series of whole tasks throughout the unit. Results indicated that the whole-task group performed significantly better than the part-task group on a skill acquisition test and a transfer test. Possible reasons for these findings and suggestions for future research are discussed.