# A facial expression-based personalization engine (FPE) for monitoring and modulating real-time effective engagement in cognitive training in older adults at risk for AD/ADRD

> **NIH NIH R61** · STANFORD UNIVERSITY · 2023 · $360,859

## Abstract

ABSTRACT
How to ensure adherence to computerized cognitive training in unsupervised circumstances (e.g., at-home, self-
administered) in older adults at risk for Alzheimer’s disease (AD) or AD related dementia (AD/ADRD) is
understudied. The objective of the R61/R33 is to refine and test a novel facial expression-based personalization
engine (FPE) for monitoring and modulating real-time effective engagement, with an ultimate goal of enhancing
long-term adherence in unsupervised cognitive training in older adults at risk for AD/ADRD. Here, Effective
engagement is defined as the extent to which someone is actively engaged and performing with significant
attention and enjoyment while training, addressing a balance between adherence and cognitive gains/plasticity
from the training. Based on previous work, including ours, we hypothesize that (1) mental fatigue revealed in
facial expressions will reflect a trainee’s degree of effective engagement, which can be modified by modulating
task novelty; (2) our proposed FPE will ensure the effective engagement in cognitive training by monitoring
trainee facial expressions and modulating training in response, promoting the trainee’s long-term adherence to
the training and cognitive plasticity. In R61 (Y1-Y2), we will generate the FPE for monitoring and modulating real-
time engagement in cognitive trainings in older adults at risk for AD/ADRD by refining our established application
programming interface using a Stage I design. In R33 (Y3-Y5), we will conduct a Stage II intervention efficacy
study comparing effective engagement and adherence in unsupervised cognitive training between training
programs with vs. without FPE in older adults at risk for AD/ADRD. We will address milestones proposed in
both stages to (a) ensure the readiness of the proposed FPE for R33 (R61 milestone) and (b) evaluate and
further revision of FPE for future implementation test (R33 milestone). Impact: the proposed FPE may assist in
monitoring and improving effective engagement and adherence in older adults with unsupervised cognitive
training. In the current application, we will test FPE in a cognitive training program called speed of processing
training. However, such FPE may be embedded to any computerized cognitive training in future studies to help
address adherence related issues.

## Key facts

- **NIH application ID:** 10766409
- **Project number:** 1R61AG084471-01
- **Recipient organization:** STANFORD UNIVERSITY
- **Principal Investigator:** Ehsan Adeli
- **Activity code:** R61 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2023
- **Award amount:** $360,859
- **Award type:** 1
- **Project period:** 2023-09-15 → 2025-08-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10766409

## Citation

> US National Institutes of Health, RePORTER application 10766409, A facial expression-based personalization engine (FPE) for monitoring and modulating real-time effective engagement in cognitive training in older adults at risk for AD/ADRD (1R61AG084471-01). Retrieved via AI Analytics 2026-05-25 from https://api.ai-analytics.org/grant/nih/10766409. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*