Affectively Aligned Assistive Technology for Persons with Dementia 1

There are nearly 900 million people aged 60 years and over living worldwide, and this rise in life expectancy is associated with increased prevalence of age-associated diseases such as Alzheimer’s Disease and Related Dementia (ADRD); today 46 million people live with dementia worldwide [1]. Dementia is characterized by the progressive deterioration of both cognitive and functional abilities, leading to a loss of the ability to perform Activities of Daily Living (ADLs) such as domestic activities and medication taking [2]. Assistive technologies have been proposed as a way to help people with dementia perform ADLs independently [3]. An assistive hand washing system, COACH, has previously been developed based on tracking hand locations and using pre-recorded “canned” verbal (audio) and visual (video) cues when the person stops making progress in the task [4]. When tested in a long-term care facility, it was found that in some cases it reduces the need for human assistance by 100% [3]. However, in some cases, the COACH failed to provide appropriate assistance, leading to confusion or agitation in the older adult end-user. This may be due to an affective (emotional) misalignment of the COACH with the specific needs or personality of the individual. While significant effort has been made to design prompts based on the methods and styles of human caregivers [5], a simple onesize-fits-all style of prompting may be limiting. For example, one person may find certain prompts to be too imperious, and would respond better to a more servile approach, while others may prefer a more imperative prompting style. Many studies have found that identity changes dramatically over the course of dementia [6] and that persons with ADRD often have more vague or abstract notions of their self-identity [7] than do non-afflicted persons. Thus, explicit models of affective identity offer an attractive mechanism for developing more appropriate and effective assistive technologies. Our aim is therefore to develop an emotionally Intelligent Cognitive Assistant (ICA) that takes into account the affective identity of the individual. The proposed ICA is based on the social-psychological principles of Affect Control Theory (ACT) [8]. ACT posits that interactions are guided by a psychological need to maximize the consistency between culturally shared fundamental sentiments about identities and behaviours, and the situationally created transient impressions about identities and behaviours in real-time interactions. Fundamental sentiments and transient impressions are interpretations of social objects, such as interactants’ identities and behaviours, represented as vectors in a 3D affective space hypothesised to be a universal organising principle of human socio-emotional experience [9]. The basis vectors of the affective space correspond to feelings of Evaluation/valence, Potency/control, and