ABSTRACT
This paper is to explore suitable training methods for complicated operation in spaceflight. Three training method named Routine method, Compounding-feedback method and Integrated method were proposed from theory analysis and astronaut training experience. To validate these methods, two-factor experiment was designed. The two factors are training methods and operation complexity. Twenty-eight participants were divided into three method groups to finish three stages experiment. The av~rage operation time, the average error, the subjective workload and the subjective survey of the first stage and the final stage were analyzed. Results show that different training methods impose varied effects on the two complexity levels at the three stages. With the increase oftraining method level (from Routine method to Compounding-feedback method and Integrated method), the average operation time of two complexity levels do not change significantly, neither do the average error of low-complexity level. However, compared with Routine method group, the average error of high-complexity level show a significant decrease in Compounding-feedback method group and Integrated method group (p=0.02, 0.039 for the first stage and the final stage respectively). The subjective workloads at both stages also show a degressive trend. Additionally, the subjective survey
showed Compounding-feedback method and Integrated method are more effective on high-complexity operations than on low-complexity operations. The conclusion is that Compounding-feedback method and Integrated method are suitable to highcomplexity operations.
