国内外大量工业事故原因分析表明，不安全行为是导致事故发生的主要原因之一。事故与人的行为有着密切联系，而人的行为又受其心理状态的支配。通过案例分析发现，94%的事故诱因与人的不良心理有关。矿工长期在井下高危及受限空间中工作，极易产生不良心理，导致工作倦怠，致使不安全行为频发。因此，如何通过改变矿工的不良心理状态，避免其工作倦怠，控制不安全行为的发生，从而保障煤矿安全生产已成为煤矿亟待解决的问题之一。安全心理资本作为一种独特的个体心理资源，对促使个体实施安全行为、提升安全绩效具有重要作用。因此本研究基于心理资本理论、资源保存理论、社会交换理论和组织支持理论，采用问卷分析和实验测量两种方法，深入研究矿工安全心理资本与不安全行为间的影响作用机制，以期通过对这一问题的研究为国内关于矿工不安全行为的管控提供新的视角和启示，对提高我国矿工安全心理资本水平、减少不安全行为、提升煤矿安全管理水平具有重要意义。 基于相关理论推演，构建了矿工安全心理资本对不安全行为影响的多层次理论模型。基于资源保存理论、社会交换理论和组织支持理论，探讨了安全心理资本与不安全行为变量之间的中介效应和跨层次调节效应，构建了“矿工安全心理资本-工作倦怠-不安全行为”以及组织支持在以上关系中发挥跨层次调节效应的理论模型。在模型的基础上提出了本研究的3类假设，包括矿工安全心理资本与不安全行为的主效应关系假设，工作倦怠在矿工安全心理资本和不安全行为之间的中介作用关系假设，组织支持在矿工安全心理资本和工作倦怠之间的跨层次调节作用关系假设。 根据我国矿工的实际情况，确定了矿工安全心理资本的结构维度，编制了中国情境下适用于矿工的安全心理资本量表。通过文献分析、现场访谈与专家反馈等方法得到了矿工安全心理资本的六维结构（包括安全自我效能、希望、韧性、乐观、自我调节和冷静）及测量题项，利用题项筛选、信度分析和探索性因子分析对初始问卷进行小规模预测试，得到共28个题项，形成了矿工安全心理资本正式量表，对正式量表进行大样本测试，采用AMOS17.0进行验证性因子分析，检验得出矿工安全心理资本的六维结构具有有效性与稳定性。确定了相关变量工作倦怠、组织支持和不安全行为的测量量表，并进行了信效度检验，在各变量维度确定的基础上，进一步对假设进行细化，共提出28个假设。 对基于理论模型提出的假设进行了检验，通过检验结果的讨论提出了矿工不安全行为管控对策。利用SPSS21.0，采用层次回归方法，检验了矿工安全心理资本对不安全行为的主效应。结果表明，矿工安全心理资本对不安全行为具有显著的负向影响，且矿工安全心理资本除乐观外的5个维度分别对不安全行为的2个维度具有显著的负向影响关系，其中，自我调节对安全不服从行为影响最大，安全自我效能对安全不参与行为影响最大。利用SPSS21.0，采用层次回归方法，检验了工作倦怠在矿工安全心理资本和不安全行为之间的中介效应。结果表明，矿工安全心理资本及各维度对工作倦怠及各维度具有显著的负向影响，工作倦怠及各维度对不安全行为各维度具有显著的正向影响，工作倦怠在安全心理资本和不安全行为的关系中起到部分中介作用，安全心理资本通过工作倦怠影响矿工不安全行为。利用HLM6.0，检验了组织支持在矿工安全心理资本与工作倦怠之间的跨层次调节效应。结果表明，矿工安全心理资本与组织支持之间的交互对工作倦怠会产生显著的负向影响，组织支持的3个维度在安全心理资本与工作倦怠的关系中起到调节作用，即组织支持的程度越高，安全心理资本对工作倦怠的负向影响关系就越强，其中，利益关心这一维度的调节效应最强。煤矿可以通过提升矿工安全心理资本、降低工作倦怠水平和提供组织支持三个方面提高安全绩效，控制矿工不安全行为的发生。 采用ERP脑电实验研究方法，进一步从脑神经机制角度探讨了矿工安全心理资本对不安全行为的影响关系，基于实验结果提出了矿工不安全行为管控对策。通过E-Prime2.0编制爱荷华博弈任务，邀请了16名矿工及具有井下工作经验的研究生参与实验任务，采用NeurOne脑电信号采集系统对实验过程中的脑电数据进行记录，利用EEGLAB和SPSS对数据进行分析。行为数据分析结果表明，低安全心理资本个体比高安全心理资本个体更偏好风险行为。脑电数据分析结果表明，低安全心理资本个体和高安全心理资本个体在风险行为决策时诱发的P300有差异，且低安全心理资本比高安全心理资本个体诱发的P300波幅更大；低安全心理资本个体和高安全心理资本个体在风险行为决策时诱发的FRN有差异，且低安全心理资本比高安全心理资本个体诱发的FRN波幅更大。高安全心理资本的矿工倾向于风险规避，表现出风险规避的行为特点，不会为了即时利益而选择冒险行为，不易发生不安全行为；低安全心理资本的矿工倾向于风险偏好，表现出风险寻求的行为特点，会为了即时利益而选择冒险行为，容易发生不安全行为，因此煤矿可以通过提高矿工风险认知能力从而降低不安全行为的发生。

From the analysis of large numbers of industrial accidents in domestic and overseas, the results shows that unsafe behavior is one of the main causes of accidents. And there’s a closely linked between accident and human behavior and human behavior is subject to its psychological state of domination. Through the author analysis of case studies, they found that nearly 94% of the accident incentives are linked by the bad psychology of the people. Quality of miners are usually worked in a underground limited spaces and high-risks atmosphere for a long-term so that easy to produced a bad psychology, job burnout and resulting in frequent unsafe behavior. Thus, how to changing the adverse mental state, avoiding their job burnout, controlling the occurrence of unsafe behavior for miners has become an urgent problems. As a unique individual psychological resource, security psychological capital plays an important role in promoting individual's safety behavior and improving safety performance. Therefore, this paper is based on psychological capital theory, resource conservation theory, social exchange theory and organizational support theory; using two methods of questionnaire analysis and experimental measurement deeply research the mechanism of safety psychological capital and unsafe behavior of miners, the research on this issue in order to provide a new perspective for the control of miners' unsafe behavior, it is of great significance to improve the level of miners’safety psychological capital, reduce the unsafe behavior and improve the safety management level of coal mine in China. Based on the related theoretical deduction,a multi-level theoretical model of psychological capital on unsafe behavior of the impact of miners has been builds. social exchange theory and organizational support theory, this paper discusses the mediating effect and cross-level regulation effect between safety psychological capital and unsafe behavior variables, and "Miner Safety Psychological Capital-Job Burnout-Unsafe Behavior" is constructed. The organization supports the theoretical model of cross-level regulatory effects in the above relationship. On the basis of this model, this paper puts forward hypotheses, so that it including the main effect between safety psychology capital and unsafe behavior; the mediating effect of job burnout between safety psychological capital and unsafe behavior; the cross-level regulating effective between the safety psychological capital in the miners and job burnout. Based on the reality of miners in China, the structural dimension of miners’ safety psych-ological capital is hereby identified and scale of psychological safety capital of miners in Chi-nese context has been developed. Through literature analysis, interviews and expert feedback methods to obtain the dimension of miners' safety psychological capital, including safety self-efficacy, hope, optimism, toughness, self-regulating and calmness. Exploiting the subjects item selection, reliability analysis and exploratory factor analysis, and 28 items were obtained,and formed a miners safety psychological capital formal scale, the formal scale for large sample test, utilization AMOS17.0 for confirmatory factor analysis, it is concluded that the 6-dimensional structure of safety psychological capital of miners has validity and stability. The measurement scales of job burnout, organizational supports and unsafe behavior were identified, and the reliability and validity were tested. Based on the dimensions of each variable, puts forward 28 hypotheses. The hypotheses based on the theoretical model has been tested and control measures for dangerous behaviors of miners has been presented based on discussion of the test results.Utili-zing SPSS 21.0, forth-putting the method of hierarchical regression to test the main effect of safety psychological capital on unsafe behavior of miners, examining the mediator effect of job burnout on safety psychological capital and unsafe behavior of miners. The results show that the safety psychological capital has a significant negative impact on the unsafe behavior, the five dimensions except optimism of the safety psychological capital of the miners have a significant negative impact on the two dimensions of unsafe behavior, respectively, from all,the self-regulation of security disobedience safety self-efficacy had the greatest influence on the safety non-participation behavior. Using SPSS 21.0, the meditational effect of job burnout on safety psychological capital and unsafe behavior of miners was tested by using hierarchical regression method. The results show that miners' psychological capital and their dimensions have significant negative effects on job burnout and various dimensions. Job burnout and dimensions have significant positive effects on various dimensions of unsafe behavior and it has significant negative effects on safety psychological capital and unsafe behavior of the relationship between the intermediary roles, the psychological safety of capital through the job burnout affect the miners’ unsafe behavior. Using HLM6.0, this paper showed the tests of the cross-level adjustment effect of organizational support on safety psychological capital and job burnout of miners. The results show that the interaction between safety psychological capital and organizational support has a significant negative impact on job burnout. The three dimensions of organizational support play a positive role in the relationship between psychological capital and job burnout, i.e., the degree of organizational support The higher the psychological capital of the negative impact on the relationship between job burnout stronger, of which interest in this dimension of the regulation effect of the strongest. By means of ERP experimental method and, further, from the perspective of the neural mechanism of brain, the relationship between safety psychological capital and dangerous beh-avior of miners has been explored. Besides, control measures for dangerous behaviors of min-ers has been presented based on discussion of test results. Using of NeurOne EEG signal acquisition system record the EEG data and statistical tools (EEGLAB and SPSS) for analysis the data. From the behavioral data analysis shows that low-security psychological capital subjects were more likely to prefer risk than high-security psychological capital subjects. there were differences in P300 between low and high-security psychological capital individuals in risk behavior decision-making, between those two, the low safety psychological capital was greater than the P300 volatility induced by high security psychological capital individual.Thus, there was a significant difference in the FRN induced by risk factors between capital individuals and individuals with high security psychological capital. Those with low security psychological capital were greater than those with high security psychological capital. The results show that the characters of Miners with high security psychological capital tend to risk aversion and show-risk aversion behavior and low security psychological capital tend to risk appetite as well as risk-seeking of the behavioral characteristics for the immediate interests of the choice of risk-taking behavior to prone to unsafe behavior.