Algorithmic Spaced Retrieval Enhances Long-Term Memory in Alzheimer Disease: Case-Control Pilot Study.

Area of Science:

• Cognitive Science
• Neuroscience
• Human-Computer Interaction

Background:

• Spaced retrieval is a proven memory enhancement technique, but its application in daily life for Alzheimer disease (AD) patients has been challenging.
• Mild cognitive impairment (MCI) due to AD affects memory, necessitating accessible interventions for sustained cognitive function.

Purpose of the Study:

• To evaluate the efficacy and feasibility of a mobile application integrating spaced retrieval with machine learning for memory improvement in individuals with MCI due to AD.
• To assess the app's ability to optimize memory retention schedules by tracking individual forgetting rates.

Main Methods:

• A pilot study involved 61 participants (including 20 with MCI due to AD) using a mobile app daily for 4 weeks to learn new facts and name-face associations.
• Participants engaged in daily sessions answering 15 questions, with a machine learning algorithm adjusting spaced retrieval intervals.
• Memory recall was assessed via recognition tests after a 1-week break from app usage.

Main Results:

• All participants with MCI due to AD showed significant memory improvements for learned facts (P<.001) after 1 month of app use.
• Memory improvements in the MCI group were comparable to young and healthy older adults, with 95% showing over 10 percentage point gains.
• No significant memory decay was observed for any group after a 1-week hiatus; 80% of MCI participants adhered to the schedule, and 50% desired continued use.

Conclusions:

• Mobile spaced retrieval applications are a feasible and effective tool for supporting memory retention in individuals with early-stage AD.
• The app successfully aids in retaining recently learned facts and name-face associations, showing promise for daily cognitive support.