ATSA-ESI: Assistive Technology Supporting Aging with Emotional and Social Intelligence

在过去的12年里，我一直在构建复杂的辅助技术（AT），旨在帮助老年痴呆症患者（例如阿尔茨海默氏症）。日益老龄化的人口给医疗保健系统带来了巨大压力，给非正式照顾者（如配偶/子女在家照顾配偶/父母）带来了不可持续的负担。我开发的技术可以在老龄化过程中提供自主性和独立性，从而将非正式护理人员从这种负担中解脱出来，并降低医疗保健成本。已经取得了很大进展，但要使这些设备有效和可用，还有很长的路要走。本研究旨在建构一个整合工程学、心理学与社会学观点的老年照护辅助科技运作模式。越来越多的社会，情感和情感智能的概念在追求有效的人机交互，包括AT中发挥着关键作用。对于AT的发展，这些概念被认为对痴呆症患者的行为有重大影响。例如，由于阿尔茨海默病干扰记忆，它通常会导致一个人扮演过去的身份，否则治疗这样的人会产生不和谐，导致冲突，压力和抑郁，当然也会导致AT不采用。为了解决这个问题，我开发了人工智能系统，将社会和情感因素整合到包括AT在内的人机交互系统中。我使用概率和决策理论顺序动态模型的基础开发这些系统。** 我的研究的未来五年将集中在群体过程模型的理论和实践发展，社会心理学理论的指导下，具体应用于老化的AT。我将沿着三个关键途径工作：个人群体中情感的分层建模和特定辅助技术的应用，从个人和群体交互的传感器数据中对用户和上下文模型进行机器学习，以及基于社会心理学原理，应用于社交网络的一般AT之外的群体过程模型的开发。我目前支持的研究着眼于具体的社会情感问题在个人互动的水平（例如，一个人从事一项活动），并与用户评价的重点。在这个提案中，我解决了模拟个人群体的问题，超越了需要照顾的人，包括他们的非正式照顾者，他们的医生，专业照顾者，后代，兄弟姐妹，邻居和社区。基于社会心理学理论，我将开发和测试老年人的社交网络模型，以提供更有效和有针对性的护理。这项工作将与我的其他赠款的其他更面向个人的工作相吻合。

For the past twelve years I have been building sophisticated assistive technology (AT) designed to assist elderly persons with dementia (e.g. Alzheimer's). An increasingly elderly population is placing a significant strain on the healthcare system, and an unsustainable burden on informal caregivers (e.g. spouses/children caring for their spouse/parent at home). I develop technologies that can provide autonomy and independence in aging, thereby releasing informal carers from part of this burden, and decreasing healthcare costs.******Much progress has been made, but there is a long way to go before these devices are effective and usable. This proposal is to build a unified operational model of engineering, psychological and sociological perspectives on assistive technology (AT) for elder care. Increasingly, the notions of social, emotional and affective intelligence are playing a key role in the pursuit of effective human-machine interaction, including AT. For AT development, these notions are known to have a major impact on the actions of persons with dementia. For example, as Alzheimer's disease interferes with memory, it can often result in a person enacting a past identity, and treating such a person otherwise creates dissonance leading to conflict, stress, and depression, and certainly to AT non-adoption. To handle this, I develop artificially intelligent systems that integrate social and emotional factors into human-interactive systems including AT. I develop these systems using a foundation of probabilistic and decision theoretic sequential dynamical models. ******The next five years of my research will focus on the theoretical and practical development of group process models, guided by social-psychological theory, with specific application to AT for aging. I will work along three key avenues: hierarchical modelling of sentiment in groups of individuals and applications to specific assistive technologies, machine learning of user and context models from sensor data of individual and group interactions, and development of models of group processes in general beyond AT, based on social-psychological principles, with application to social networks. My currently supported research looks specifically at socio-emotional issues at the level of individual interactions (e.g. a single individual engaged in a single activity), and with a user-evaluative focus. In this proposal, I tackle the problem of modelling groups of individuals, going beyond the person in need of care to include their informal caregiver, their doctor, professional carer, offspring, siblings, neighbours and community. Based on social-psychological theory, I will develop and test models of social networks of elderly persons to provide more effective and tailored care. This work will dovetail with the other, more individual-oriented work funded by my other grants.**