The reading for this class was “Parallel Prototyping Leads to Better Design Results, More Divergence, and Increased Self-Efficacy,” a work from Stanford researchers Dow et al. Though the idea behind this paper is not too complicated, I found the methods and results to be intriguing. The study concerned itself with the following task: create an attention-grabbing online advertisement. Participants would go through an iterative process, receiving feedback on designs as direction for their future designs. Overall, participants were allowed to create five designs on which they would receive feedback. They would create one final design after this process, serving as the basis for overall assessment of their effort. The main idea in this study is in the way that the designs and corresponding feedback were interspersed. For one group, feedback was provided after every single design (Serial condition). For the other group, multiple prototypes were created before each round of feedback, which focused on the group of prototypes (Parallel condition). The authors found that those in the Parallel condition group outperformed the Serial condition group in the quality of their final design. In addition, the Parallel condition group created more diverse prototypes throughout the process and reported a higher level of self-efficacy in the design process at the conclusion of the study.
I found the results concerning quality to be fascinating, as well as the authors’ discussion on this. They believe the reason the Parallel condition group had higher quality final products than the Serial condition group is the following: by receiving feedback on a group of prototypes, participants were able to make more connections between design principles. The Serial condition group took the critiques as suggestions for modification of their current design, while the Parallel condition group took the critiques more holistically. This is a suspicion that I have held for a while, so it was nice seeing research backing this up. This conclusion is further supported by the results of diversity among prototypes. Those in the Parallel condition group had a much more diverse group of prototypes than those in the Serial condition group. The fact the the Serial condition group had less diversity lends credence to the theory that those participants were more focused on using the critique to iterate on their current design. The authors back up this idea with research from Osborn, who developed a set of rules for the process of brainstorming. By postponing feedback until multiple ideas have been proposed, the focus shifts from refinement of ideas to diversity of ideas. Based on the ideas and results presented in this paper, I’m looking forward to implementing aspects of this parallel approach to creation in my course project.