There exists a gap in knowledge between scientists and the larger non-scientist public. Therefore, much of the information provided to the public regarding research that should influence their decisions is often misunderstood. In order to eliminate, or at the very least, minimize this gap, there is a need to educate non-scientists about research methods and experimental design. In order to address this need, we have created a digital game, Mad Science, that allows non-scientists to create and participate in experiments to better understand research methods. The current study analyses the results of a paper prototyping session, where non-scientists were asked to create experiments using the tools and scaffolding provided in the game. Participants were able to create playable scenarios and testable experiments. However, our results suggest a need for further AI support and scaffolding to address common areas of confusion and to facilitate the experimental design process.