PROJECT SUMMARY The essence of the Alzheimer’s disease (AD) phenotype is a decline in memory. Well before the onset of obvious dementia, there are changes in memory that patients and family members notice which diminish quality of life for those impacted. Therapeutic interventions that target the presymptomatic and early symptomatic stages of AD often choose a cognitive endpoint to demonstrate efficacy, but conventional memory assessments often fail to capture these subtle changes that occur early in the disease. One reason is that conventional memory assessments do not reflect how memory is relied upon in the everyday lives of participants. For example, it is very distressing for patients to experience memory lapses like forgetting first names, losing personal items, or having difficulty remembering the plot of a book or film. Conventional memory testing assesses recall at short periods, typically 30-minutes or less after learning the information, whereas in the everyday lives of patients, critical information must be recalled over much longer periods like hours, days, or even weeks to maintain quality of life. Several clinical studies have demonstrated that extending recall of newly learned information past the usual 30-minute or less delay period to much longer intervals dramatically increases the rate of decay, a concept known as accelerated long-term forgetting (ALF). We have shown that participants who carry a mutation for autosomal dominant Alzheimer’s disease (ADAD) but who are presymptomatic, perform similarly to non-carriers on conventional memory testing. Critically, when we tested mutation carriers on their long-term recall at 7 days, they had lost much more information than non-carriers, suggesting that extending the recall period may reveal important differences in memory consolidation that manifest well before the onset of noticeable dementia symptoms. An obvious challenge for measuring ALF in clinical populations is burden. Conventional testing would require multiple visits from study participants and require multiple interactions with study personnel, which is practically and financially unfeasible. The increasing ubiquity of smartphones provides an opportunity to assess cognition in populations at risk for AD, while allowing for longer-term follow-up without excessive burden or financial impact. In this study, we will develop novel ALF measures for smartphone-based administration and validate their effectiveness in presymptomatic autosomal dominant AD. Our development process will include user experience studies and clinical trial readiness audits, and the resulting application source code will be made freely available. The ultimate goal is to develop a highly sensitive, accessible, and clinically meaningful cognitive endpoint for use in international AD clinical trials, including adding ALF measures to the Dominantly Inherited Alzheimer Network-Trials Unit (DIAN-TU).