Use of sensor-based feedback technology in reducing home energy consumption

Traditionally the energy crisis has been addressed through advances in (a) efficient technologies allowing users to carry out the same function with less energy, (b) offsets from non-renewable energy sources through renewable sources, and (c) pricing and regulatory measures. However, these means make incremental changes in face of critical urgency. There is greater potential for large scale energy usage reductions through behavior change. In the past few years there has been heightened interest in using sensor data to reduce homeowners' energy consumption. The assumption is that providing the data is enough of an incentive to significantly alter their consumption habits. Our ethnography research challenges this assumption. We conducted ethnography interviews in dozens of homes of typical and extreme energy consumers. Our research revealed that overall, homeowners do not understand energy. Particularly, they have no sense of how much electricity use is average or how their actions contribute to their consumption. Based on our ethnography research and prototyping studies, we found that even if users are provided with sensor-based feedback, if the feedback is not presented through an interface that takes the users' complex relationship with energy into account, the data will fail to significantly change their behavior. In order to make data feedback effective, we are investigating potential areas of collaboration between (a) users and the energy infrastructure, and (b) multiple users in a social context. Such collaborations will be made possible by an entirely new level of dialogue enabled by interactive and social interfaces that display sensor-based feedback data. As energy measurement systems and the software that processes these measurements become more accurate, users will be able to use such interfaces to interact with and learn from the data on a higher level. We are currently designing a study to test prototypes of online interfaces that encourage users to interact with their energy data. These interfaces will enable us to determine how effective it is to frame sensor data in a way that promotes the user's understanding of energy and their collaboration with other users. We are conducting our research under the federal ARPA-E grant to Stanford University for reducing energy consumption through interactive technology and a human centered design process. Our research into collaboration between users and the energy infrastructure represents an important new approach to addressing how behavioral interventions can ameliorate the energy crisis.