针对直升机驾驶舱人机交互中认知负荷问题，设计飞行模拟试验，探究视觉和听觉告警对飞行员认 知负荷的影响。基于多元线性回归方法，通过主观评价结果、反应时间、准确率和脑电信号等指标，构建了直 升机驾驶舱人机交互认知负荷评估模型。结果表明文字颜色、语音提示两种告警特征对飞行员信息加工均会产 生影响，使用多元线形回归模型可对不同告警方式下受试对象的认知负荷水平进行评估，构建的评估模型能够 较准确地评估直升机驾驶舱中飞行员的认知负荷，可为优化驾驶舱交互设计、降低飞行员认知负荷、减少人为 差错提供依据。

Aiming at the cognitive load of civil aircraft cockpit interaction, a flight simulation test has been designed to explore the effects of visual and auditory warnings on the cognitive load of pilots. Based on multiple linear regression method, the cognitive load assessment model of aircraft cockpit interaction is constructed by subjective evaluation results, reaction time, accuracy and EEG signals. The results show that the two warning features, text color and voice prompt, have an impact on pilot information processing. The results show that the multiple linear regression model can evaluate the cognitive load of subjects under different warnings. The evaluation model can accurately evaluate the cognitive load of pilots in civil aircraft cockpit. The results can provide a basis for optimizing cockpit interaction design, reducing pilot’s cognitive load and reducing human error.

V223.1

国家自然科学基金委员会- 中国民用航空局民航联合研究基金项目（U2033202，U1333119），国家自然科学基 金项目（52172387），中央高校基本科研业务费（1007-XBC22006）