CHAPTER FIVE · LET THE GAME TESTER DO THE TALKING protocols will provide more information about details of the interaction in a particu- lar context. Conducting a think-aloud study generally requires that a prototype is available, or at least some form of a mockup that allows the participant to get an idea of the flow of interaction. However, this also means that any feedback collected comes quite late in the development process, which in most cases will mean that no longer fundamental changes can be made. Again, this points out that it is important to adopt a multi-method approach, in combination with continuous feedback sessions throughout the development cycle. And even more importantly, that adoption of a certain method or tool for evaluation should be guided by the question(s) one wants to address. 5.8 References Als, B.S., Jensen, J.J., & Skov, M.B. (2005). Comparison of think-aloud and constructive inter- action in usability testing with children. In: Proceedings of the Conference on Interaction Design and Children, Boulder, Colorado. ACM Library, pp. 9­16. Bainbridge, L., & Sanderson, P. (2005). Verbal protocol analysis . In J.R. Wilson, & E.N. Corlett (Eds) , Evaluation of Human Work. London: Taylor and Francis, pp. 159­ 184 .