Within-Subjects Design

Perhaps testing 15 participants is simply out of the question. Instead of 15, you could get by with only five participants by having each one perform all three modules as shown in the table below. This is called a icithin subjects design. However, you have the same problem of transfer of learning effects to consider.
To mitigate these effects, you must use a technique called counterbalancing, whereby the order of tasks is either randomized or balanced out. By varying the order of the presentation of tasks, you can limit the effects of learning transfer.