Construction and usability evaluation of a cognitive rehabilitation assessment and training system platform for MCI elderly people based on human-computer interaction (Preprint)

BACKGROUND Mild cognitive impairment (MCI) is associated with an increased risk of cognitive decline, highlighting the need for accessible and individualized cognitive rehabilitation approaches. Human-computer interaction (HCI) technologies may provide new opportunities for digital cognitive assessment and training in older adults with MCI. However, the development of age-friendly platforms that integrate multimodal interaction and multidomain cognitive training remains challenging. OBJECTIVE This study aimed to develop an HCI-based cognitive rehabilitation assessment and training platform for older adults with MCI and evaluate its preliminary usability and user acceptability. METHODS This study comprised four phases. First, a literature review and evidence synthesis were conducted to identify relevant components and design considerations for digital cognitive assessment and training. Second, two rounds of Delphi expert consultation involving 15 experts were conducted to establish and refine the platform's functional and design indicators. Third, the platform was developed based on HCI theory, cognitive reserve theory, and age-friendly design principles. Finally, a mixed-methods usability evaluation was conducted with 15 older adults with MCI recruited from a nursing home in Suzhou, China. Data were collected through representative task testing, the System Usability Scale (SUS), and semistructured interviews. Quantitative data were analyzed descriptively, while interview findings were summarized to identify perceived benefits and usability barriers. RESULTS The developed platform integrated cognitive assessment, multidomain game-based training targeting memory, executive function and attention, language, visuospatial function, and calculation, and multimodal interaction through touch-based, eye-tracking-based, and gesture-based technologies. It also provided automated feedback on training performance, including task accuracy, duration, and difficulty. During the Delphi consultation, the expert authority coefficient was 0.84, and Kendall's coefficient of concordance indicated statistically significant agreement in both rounds (P<0.01). In task testing, completion rates for accessing the assessment module, completing cognitive assessment, selecting a training task, performing cognitive training, and viewing training results were 100%, 86.67%, 100%, 80%, and 86.67%, respectively. Difficulties were mainly related to understanding task instructions, completing time-limited activities, eye-tracking recognition, and locating navigation controls. The mean SUS score was 74.67 (SD 8.81). Interview findings indicated positive perceptions of the variety of training activities, graded difficulty, and interactive features. However, participants also reported insufficient on-screen instructions, limited response time, visually complex backgrounds, small font sizes in some sections, and occasional eye-tracking inaccuracies. All 15 participants expressed willingness to continue using the platform. CONCLUSIONS The HCI-based platform demonstrated preliminary usability and acceptability among older adults with MCI. Combining multidomain cognitive training with multimodal interaction may provide a feasible approach to digital cognitive rehabilitation. Nevertheless, the identified operational and interaction barriers highlight the need for further age-friendly optimization. Larger studies and longitudinal evaluations are warranted to assess sustained engagement, usability in different settings, and potential clinical benefits.

Authors

Publication Details

Published
2026-11-14
DOI
https://doi.org/10.2196/preprints.116941
Primary Topic
Dementia and Cognitive Impairment Research
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Construction and usability evaluation of a cognitive rehabilitation assessment and training system platform for MCI elderly people based on human-computer interaction (Preprint)

Ting Chu, Jun Luo, Xiaona Bao, Jiayan Mao
Dementia and Cognitive Impairment Research
preprint

Construction and usability evaluation of a cognitive rehabilitation assessment and training system platform for MCI elderly people based on human-computer interaction (Preprint)

Ting Chu, Jun Luo, Xiaona Bao, Jiayan Mao
preprint en

Abstract

BACKGROUND Mild cognitive impairment (MCI) is associated with an increased risk of cognitive decline, highlighting the need for accessible and individualized cognitive rehabilitation approaches. Human-computer interaction (HCI) technologies may provide new opportunities for digital cognitive assessment and training in older adults with MCI. However, the development of age-friendly platforms that integrate multimodal interaction and multidomain cognitive training remains challenging. OBJECTIVE This study aimed to develop an HCI-based cognitive rehabilitation assessment and training platform for older adults with MCI and evaluate its preliminary usability and user acceptability. METHODS This study comprised four phases. First, a literature review and evidence synthesis were conducted to identify relevant components and design considerations for digital cognitive assessment and training. Second, two rounds of Delphi expert consultation involving 15 experts were conducted to establish and refine the platform's functional and design indicators. Third, the platform was developed based on HCI theory, cognitive reserve theory, and age-friendly design principles. Finally, a mixed-methods usability evaluation was conducted with 15 older adults with MCI recruited from a nursing home in Suzhou, China. Data were collected through representative task testing, the System Usability Scale (SUS), and semistructured interviews. Quantitative data were analyzed descriptively, while interview findings were summarized to identify perceived benefits and usability barriers. RESULTS The developed platform integrated cognitive assessment, multidomain game-based training targeting memory, executive function and attention, language, visuospatial function, and calculation, and multimodal interaction through touch-based, eye-tracking-based, and gesture-based technologies. It also provided automated feedback on training performance, including task accuracy, duration, and difficulty. During the Delphi consultation, the expert authority coefficient was 0.84, and Kendall's coefficient of concordance indicated statistically significant agreement in both rounds (P<0.01). In task testing, completion rates for accessing the assessment module, completing cognitive assessment, selecting a training task, performing cognitive training, and viewing training results were 100%, 86.67%, 100%, 80%, and 86.67%, respectively. Difficulties were mainly related to understanding task instructions, completing time-limited activities, eye-tracking recognition, and locating navigation controls. The mean SUS score was 74.67 (SD 8.81). Interview findings indicated positive perceptions of the variety of training activities, graded difficulty, and interactive features. However, participants also reported insufficient on-screen instructions, limited response time, visually complex backgrounds, small font sizes in some sections, and occasional eye-tracking inaccuracies. All 15 participants expressed willingness to continue using the platform. CONCLUSIONS The HCI-based platform demonstrated preliminary usability and acceptability among older adults with MCI. Combining multidomain cognitive training with multimodal interaction may provide a feasible approach to digital cognitive rehabilitation. Nevertheless, the identified operational and interaction barriers highlight the need for further age-friendly optimization. Larger studies and longitudinal evaluations are warranted to assess sustained engagement, usability in different settings, and potential clinical benefits.

Dementia and Cognitive Impairment Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon