Abstract
This study examined why various levels of task autonomy differ in their learning outcomes. We conducted an experimental study in which 119 undergraduate students learned a computer task. During the learning phase, (no versus moderate versus full) autonomy and (cognitively undemanding versus cognitively demanding interruptions) demands were manipulated in a 3 × 2 between-participants design. The results showed that in the no and full autonomy conditions, receiving cognitively demanding interruptions decreased learning outcomes compared to receiving cognitively undemanding interruptions. However, having moderate autonomy resulted in equally positive learning outcomes in both cognitive demands conditions. Thus, having autonomy while learning a new task is essential; however, having too much autonomy may lead to adverse learning outcomes when cognitive demands are high.
Acknowledgements
We thank the Electronic Research Group (especially Hubert Voogd) for programming the Tom and Jerry program. Furthermore, we are grateful to Merel Cornelissen for collecting the data of this experiment.