287. The role of adaptive human-computer interaction interface design for the elderly with cognitive impairment in maintaining cognitive function

Abstract Background With the increasing aging of the global population, cognitive impairment in the elderly has become an important public health issue. Elderly cognitive impairment includes pathological states ranging from mild cognitive impairment to Alzheimer's disease. In addition to traditional pharmacological and non-pharmacological interventions, human-computer interaction technology is beginning to be applied in cognitive training and functional maintenance for patients with cognitive impairment in current research on psychotherapy. However, traditional human-computer interaction interfaces are difficult to adapt to the heterogeneity of attention, memory, and executive function among cognitive impairment groups, resulting in low training compliance and poor effectiveness. The research aims to explore the effectiveness of an adaptive human-computer interaction interface that combines the principles of mental rehabilitation in maintaining cognitive function in the elderly population with cognitive impairment, providing an efficient technical intervention strategy for the field of mental rehabilitation. Methods The study selected 60 elderly patients who met the diagnostic criteria for mild cognitive impairment in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders as the research subjects. All patients were randomly divided into an experimental group (n = 30) and a control group (n = 30). The former uses a research designed adaptive human-computer interaction interface for cognitive training, while the latter uses a functionally similar but non-adaptive standard interface for training. The expected duration of the intervention is 12 weeks, with 3 standardized cognitive tasks of 45 minutes each time, conducted 3 times a week. The training content includes executive function training such as working memory, attention transfer, and reaction inhibition. Before and after the intervention, the Montreal Cognitive Assessment (MoCA), Digital Span Task (DST), and Hamilton Depression Scale (HAMD-17) were used for evaluation. 'p<.001' indicates a highly significant statistical difference. Results After 12 weeks of intervention, both groups of patients showed varying degrees of improvement in cognitive function and emotional state. On the main cognitive indicators, the adaptive group's MoCA score increased by an average of 3.25 ± 0.88 points, significantly higher than the control group's 1.12 ± 0.71 points (p<.001). In terms of functional execution, the DST score of the adaptive group increased from a baseline average of 6.10 ± 0.95 points to 7.58 ± 0.92 points. In addition, the HAMD-17 score of the adaptive group decreased by an average of 4.15 ± 1.05 points, while the control group only decreased by 1.50 ± 0.88 points, and the difference between the two groups was statistically significant (p<.001). Discussion The adaptive human-computer interaction interface design for the elderly with cognitive impairment has significantly better therapeutic effects than traditional non-adaptive interfaces in maintaining patient cognitive function and stabilizing emotions. This technological intervention strategy effectively enhances the personalization and effectiveness of cognitive training, and can be integrated into community and home rehabilitation systems, becoming an important component of daily cognitive function maintenance. Future research considers further integrating higher-level biological signals to achieve more refined cognitive state assessment and real-time feedback loops. Funding No. 20210113.