Personalised healthy ageing: AI-based approach to predicting changes in self-perception and the relationship to cognitive decline

世界正在经历人口结构的变化，老龄化的人口比以往任何时候都多。老龄化是卫生领域的一个关键重点领域，因为老龄化过程往往伴随着身体逆境和认知能力[1]。最近的研究表明，自我参照有助于一生中的认知表现，它与特定的神经回路有关[2]。这就提出了一个问题，即自我参照是否可以用来补救认知衰退。为了帮助促进健康，跟踪衰老过程并了解与自我参照的关系可以突出显示何时需要医疗保健系统。为了应对这一挑战，至关重要的是要了解自我参照是否对老年人认知能力下降的恶化起到缓冲作用，老年人是否通过与年轻人相同的过程来维持自我参照提高的成绩，或者是否存在额外的补偿过程。神经心理学和临床检查显示出不同程度的认知衰退，例如记忆力、执行功能和处理速度的下降。在这个项目中，我们将重点放在使用行为和脑电(EEG)技术的第一个方面：理解自我参照在增强与年龄相关的记忆方面的规则。未来的学生将使用我们最近开发的新实验程序与尖端人工智能方法[3，4]相结合的多尺度数据集，(I)测试健康老年人的记忆增强是否源于对自我相关刺激的超快神经反应，以及(Ii)基于这些神经电路训练离线脑电数据以驱动自我参照记忆，然后使用该分类器解码实时脑电数据以预测认知能力下降。后者将具有巨大的潜力，通过利用人工智能来提供健康状况的读数，来支持老龄化人口对医疗保健监测日益增长的需求。该项目本质上是跨学科的，涉及行为神经科学和计算科学之间的强烈整合。学生将在整个博士期间得到阿伯丁大学心理学学院和NCS优秀合作研究团队的支持。该项目包括一个明确的机会，让学生在我们跨学科人工智能和健康老龄化环境的独特优势中茁壮成长。

The world is experiencing a demographic shift with more people ageing than ever before. Ageing is a critical focus area in health because the ageing process is often accompanied with physical adversity and cognitive capacities [1]. Recent research has shown that self-reference facilitates cognitive performance across the lifespan, and it links to specific neural circuits [2]. This raises the issue of whether self-reference can be used to remediate cognitive decline. To help promote health, tracking the ageing process and understanding the relationship to self-reference could highlight when the healthcare system is needed. To address this challenge it is essential to understand whether self-reference acts as a buffer against the deterioration of cognitive decline in older people, whether performance boosted by self-reference is maintained in older people by the same processes as found in young people, or whether there is recruitment of additional, compensatory processes. Neuropsychological and clinical examinations have revealed different levels of cognitive decline, e.g., reductions in memory, executive functions, and processing speed. In this project, we will focus on the first of these using behavioural and Electroencephalography (EEG) techniques: understanding the rules of self-reference in bolstering age-related memory. The prospective student will use the new experimental procedures we recently developed in combination with cutting-edge AI approaches [3,4] for a dataset with multiple-scales, (i) testing if enhanced memory in healthy older adults stems from ultra-fast neural responses to self-related stimuli, and (ii) training offline EEG data based on these neural circuits for driving self-referential memory, and then using this classifier to decode real-time EEG data for the prediction of cognitive decline. The latter will have a vast potential to support the growing need for healthcare monitoring in an ageing population by harnessing AI to give a readout of health status.The project is inherently interdisciplinary and involves strong integration between behavioural neuroscience and computing science. The student will be supported throughout the PhD by the excellent collaborative team of researchers in the schools of Psychology and NCS at the University of Aberdeen. The project includes a clear opportunity for the student to flourish within the unique strengths of our interdisciplinary AI and healthy ageing environment..