Developing a Multidimensional Approach to Assess Impairment in Skilled Environments

与认知障碍相关的风险在一生中可能是严重的，并与许多因素有关，如疾病、衰老、伤害和物质使用。与专业判断和标准化实践一起，神经心理学和心理测量学测试一直是评估这种性质的风险的典型方法。在工作场所和高技能环境中尤其如此。随着认知神经科学和大数据分析领域最近令人兴奋的进展，现在有可能重新审视评估风险的实用方法。我们现在有能力开发计算算法来模拟人类行为背后的大规模网络，这是传统评估技术不可能实现的。此外，这种机器学习方法可以成为一种强大的工具，基于来自神经科学、认知、生理学、行为、技能表现和病史等领域的人类数据的相对组合来预测特定的真实世界行为。Driveable是一家加拿大公司，从事评估现实世界损害风险的业务，为一系列行业合作伙伴提供认知科学解决方案。他们感兴趣的是开发一套工具，涵盖人类所有的认知和表现，高度预测许多工作场所场景中的真实世界行为。他们的目标是制定可靠的解决方案，以评估适用于运输、执法、大麻使用和其他安全敏感环境的损害风险。这项提议的主要目标是继续促进我在艾伯塔大学的研究项目与Driveable之间建立的关系。长期目标是使用认知神经科学和心理生理学的研究方法以及机器学习来理解人类在熟练的现实世界环境中表现的“蓝图”，以便为Driveable开发一套工具，使其继续成长为评估现实世界损伤风险的行业领先者。

The risks associated with cognitive impairment can be severe across the lifespan and are related to many factors such as illness, aging, injury, and substance use. Along with professional discretion and standardized practice, neuropsychological and psychometric testing have been typical methods for assessing risk of this nature. This is particularly the case in the workplace and in highly skilled environments. With recent and exciting advances in cognitive neuroscience, and big data analytics it is now possible to re-vision the practical approach to assessing risk. We now have the capability to develop computational algorithms to simulate large-scale networks underlying human behavior in ways not possible from traditional assessment techniques. Moreover, this machine learning approach can be a powerful tool for predicting specific real-world behavior based on a relative combination of human data from domains such as neuroscience, cognition, physiology, behavior, skilled performance, and medical history. DriveABLE is a Canadian company in the business of evaluating real-world impairment risk with cognitive science solutions for a range of industry partners. They are interested in developing a set of tools that cover a full range of human cognition and performance that are highly predictive of real-world behavior in many workplace scenarios. Their goals are to develop reliable solutions to assess impairment risk as it applies to transportation, law enforcement, cannabis use, and other safety-sensitive environments. The primary goal of this proposal is to continue to foster an established relationship between my research program at the University of Alberta, and DriveABLE. The long-term goal is to use cognitive neuroscience and psychophysiology research methods along with machine learning to understand the "blueprint" of human performance in skilled real-world environments such that a set of tools will be developed for DriveABLE to continue their growth as an industry leader in evaluating real-world impairment risk.