Project Summary Health self-management—gathering, organizing, and acting on health information from a rapidly growing array of online sources of healthcare information—is cognitively challenging for many older adults, especially those with cognitive impairments. The goal for this project is to support the cognitive components of older adults’ health-management activities through development of digital assistant technology tools tailored to three exemplar healthcare management task activities: accessing support services, managing healthcare finances, and using the health-management tools provided by Medicare.gov. This project will leverage the machine- intelligence expertise of our collaborators and our experience in developing and evaluating technologies for supporting the health and wellbeing needs of older adults to harness technology to provide cognitive support to aging adults, including those with Mild Cognitive Impairment (MCI) and lower SES. The project has three aims: 1) identify barriers and facilitators to diverse older adults use of current tools for managing healthcare in the three targeted task activities and provide requirements for technology tools to support these activities; 2) develop highly usable and reliable intelligent digital assistants capable of adapting to a large array of healthcare tasks in each of the three exemplar areas; and 3) evaluate the efficacy of the assistants compared to a usual tool including a digital assistant control (e.g., Google Assistant). The project will be comprised of three phases. Phase 1 will use a multimethod approach across our three study sites to assess the demands and challenges facing diverse older adults in the performance of the three health-management activities. Techniques include structured one-on-one interviews of subject matter experts, cognitive task analysis of existing tools, and process tracing of older adults’ task performance. In Phase 2, we will use the knowledge derived from Phase 1, together with an understanding of older adults’ cognitive capabilities and limitations, and their needs and preferences, to conduct iterative design of the digital assistant tools and evaluate their effectiveness and perceived usability using older adults with and without cognitive impairments and diverse in technology skills. In Phase 3, we will do a comparative assessment of the digital assistant tools for the three health-management activities (4 problems in each domain) by randomizing a cross-site sample of 240 participants, with and without MCI, varying in age, ethnicity/race, SES, and technology experience, to novel tool and control conditions, assessing efficacy and usability of the novel tools. The project will yield important information on how best to design technology aids to provide cognitive support for health decision making.