Preliminary Exploration of AI Technology in Identifying Facial Emotion Features in Older Individuals with Normal Cognitive Function

This study aims to explore whether artificial intelligence (AI) technology can accurately identify changes in facial expression features in individuals with normal cognitive function while watching positive or negative emotional stimuli videos, as well as during short-term memory recall and calculation tasksThis study aims to explore whether artificial intelligence (AI) technology can accurately identify changes in facial expression features in individuals with normal cognitive function while watching positive or negative emotional stimuli videos, as well as during short-term memory recall and calculation tasks.

Participants will sit in front of a computer and follow these steps:

1. Record a 5-second neutral facial expression as a baseline.
2. Watch a positive emotional stimulus video.
3. Perform short-term memory recall and calculation tasks.
4. Watch a negative emotional stimulus video (each video not exceeding 60 seconds).
    Throughout the process, participants' facial expressions will be recorded and analyzed using AI technology.

Preliminary results show that among 18 participants with normal cognitive function, the AI system correctively identifies 83% participants’ positive facial expressions while watching the positive emotional stimulus video, and 77% negative facial expressions while watching the negative stimulus video. While, the identification rate for the emotional responses during short-term memory recall and calculation tasks were approximately 50% positive and 50% negative. Future studies will continue to recruit participants with mild cognitive impairment and dementia to further explore their differences in positive and negative facial expressions.

Preliminary findings indicate that AI technology can effectively distinguish between positive and negative facial expression features in individuals with normal cognitive function while watching short video.

This study provides initial empirical support for future clinical applications, which may contribute to early screening and identifying cognitive declines, slow disease progressions, improve quality of life for patients, and reduce the burden on caregivers.