Evaluation of Different Modalities for Self Measuring Impulsivity

Mobile devices are becoming an increasingly integral part of modern life. As the popularity of smartphones, smartwatches, and voice assistants continues to rise, more people are using them to track health and medical statistics. We enable easy and consistent data collection by extending ResearchStack, a framework for developing mHealth applications, to support Android Wear devices. We also evaluate user acceptance of different modalities by collecting feedback from 30 users using a smartphone, a smartwatch, and a voice assistant version of the Balloon Analogue Risk Task (BART), a computerized measure of risk taking behavior. The smartphone was the preferred platform, with users stating ease of use and familiarity with the platform as reasons for their preference. However, almost 17 percent of users listed the smartwatch platform as their favorite, stating that it was also easy to use and quicker than the other versions. The voice assistant application was rated the lowest, but it can be a viable option for others who are not able to use a smartphone, expanding the participation base for data collection. Users also provided feedback on how to improve these new modalities, which will be incorporated into the next version of these applications.