Collaborative Research: Measuring Attention, Working Memory, and Visual Perception To Reduce Risk of Injuries in the Construction Industry

人为错误（例如，糟糕的决定或不安全的行动）是造成各行各业高达80%的工作场所事故的主要偶然因素。从某种程度上说，我们有限的信息处理能力是这类错误的主要来源，更好地理解认知过程将产生更有效的方法来预测和减少让员工面临风险的糟糕决策。因此，本研究将完成一系列的眼球追踪实验，以建立一个错误检测框架——人为错误检测框架——计算职业环境中人为错误的可能性，从而使主动对策能够保证工人的安全。随后，为了扩展该框架的价值，该项目将丰富和扩展基于研究的教育材料、推广和参与活动，以向社区和工人传播该框架的意识。为了实现这些目标，这个多学科项目将眼球运动和工人注意力的研究与工作记忆负荷和决策的研究相结合，以发现动态工作环境中的工人如何以及为什么不能发现、理解和/或应对身体风险。使用动态和高风险的建筑环境作为测试平台，拟议的框架将把实验室和现场实验中的眼球运动和认知操作与工人人口统计学联系起来，以确定预测动态工地中导致事故的人为错误的前兆。总之，这个项目将展示综合认知心理学、工程学和先进计算来改善决策和职业安全的价值和有效性。人为错误检测框架将利用实时眼动模式来识别人为错误，从而为综合数据分析技术奠定基础，从而在伤害发生之前自动识别和中断人为决策错误。使用本研究得出的预测模型不仅有助于显著减少事故，而且还将提供一个关键的验证措施，以确认培训计划在提高工人素质方面的有效性。风险分析的技能。此外，由于该项目为研究人员、学生和工人提供了工具和见解，用于加强职业安全和多学科研究，该项目将发展决策、风险和管理领域更广泛的教学格局。由于这项创新的研究挑战了传统的、反动的安全风险管理范式，通过使用他们的认知过程的可测量指标来识别有风险的工人。建议的主动职业安全方法有可能避免跨行业的职业事故，从而可以预见地防止损害数百万美国工人及其家庭福祉的伤害。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Human error (e.g., poor decisions or unsafe actions) are a main casual factor in up to 80% of all workplace accidents across a breadth of industries. To the extent our limited capacity for information processing capacity is a major source of such errors, better understanding of cognitive processes will yield more effective methods for predicting and reducing the poor decisions that put workers at risk. Accordingly, this study will complete a series of eye-tracking experiments to build an error-detection framework - the Human-Error Detection Framework - that computes the likelihood of human error in occupational settings to enable proactive countermeasures to keep workers safe. Subsequently, to extend the value of this framework, this project will enrich and expand research-based educational materials, outreach, and engagement activities to spread awareness about this framework to communities and workers. To achieve these goals, this multidisciplinary project blends research linking eye movements and worker' attention with research focused on working-memory load and decision making in order to discover how and why workers in dynamic work environments fail to detect, comprehend, and/or respond to physical risks. Using the dynamic and high-risk environment of construction as a testbed, the proposed framework will connect eye movements and cognitive manipulations in laboratory and field experiments with worker demographics to identify precursors that predict accident-causing human errors in dynamic worksites. In all, this project will demonstrate the value and effectiveness of synthesizing cognitive psychology, engineering, and advanced computation to improve decision making and occupational safety.The Human-Error Detection Framework will harness real-time eye-movement patterns to identify human errors and thereby lay the foundation for synthesizing technology with data analysis to automatically identify and interrupt human decision-making errors before injuries occur. Using the predictive models resulting from this study will not only contribute to significant accident reduction but will also provide a critical validation measure to confirm the effectiveness of training programs in enhancing workers? risk-analysis skills. Furthermore, since this project provides tools and insights for researchers, students, and workers to use to enhance occupational safety and multidisciplinary research, this project will evolve the broader pedagogical landscape of the decision, risk, and management sector. As this innovative research challenges the conventional, reactionary paradigm of safety-risk management by enabling the identification of at-risk workers using a measurable indicator of their cognitive processes?i.e., their eye movements, the proposed proactive approach to occupational safety has the potential for averting occupational accidents across industries and thereby will foreseeably prevent the injuries that undermine the well-being of millions of American workers and their families.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.