Experimental Research On Human Error By Mental Workload

The security issues are always an inevitable topic of the process of human development. And we need in-depth study of the reasons that lead to security incidents to improve the level of safety management, so we could control the source of security incidents. A variety of reasons that lead to security incidents, but human error plays a major role in the causes of the accidents. However, sorts of reasons may cause the human error, and people do not quite understand the mechanism what reasons induce the human error, and prediction and prevention measures of security incidents which cause by human error even rare. Therefore, it is necessary to strengthen the study about human error.Previous studies have shown that mental workload may lead to human error, there are also evidences support the reasonableness of this conclusion. In this paper, we design contrast experiments which contain a single-task and dual-task, combined with event-related potentials, to investigate the case with or without additional mental workload imposed, the mechanism that mental workload causes human error, and the difference EEG performance and behavior between two tasks.The analysis suggests that excessive mental workload make the performance of the job dropped sharply and remained at a very low level. EEG data analysis revealed the possible causes of the failures:the same task difficulty need more mental resources, as well as the reaction state of readiness has been extent. Based on the analytic results of behavioral data and EEG data, we can identify the False Alarm according to P300amplitude and reaction time, which recommended the implementation of remedies could be taken in order to prevent the occurrence of security incidents.Finally, discrimination to some doubts in this study has been made, summarizes the conclusions of this paper on the contribution to the safety management and the significance in practical application. The deficiencies of this study and the experimental design were presented in the last of the paper; also put forward some ideas for further research.