Assistive technologies to address capabilities of people with dementia: from research to practice

Assistive technologies (AT) became pervasive and virtually present in all our life domains. They can be either an enabler or an obstacle leading to social exclusion. The Fondation Médéric Alzheimer gathered international experts of dementia care, with backgrounds in biomedical, human and social sciences, to analyse how AT can address the capabilities of people with dementia, on the basis of their needs. Discussion covered the unmet needs of people with dementia, the domains of daily life activities where AT can provide help to people with dementia, the enabling and empowering impact of technology to improve their safety and wellbeing, barriers and limits of use, technology assessment, ethical and legal issues. The capability approach (possible freedom) appears particularly relevant in person-centered dementia care and technology development. The focus is not on the solution, rather on what the person can do with it: seeing dementia as disability, with technology as an enabler to promote capabilities of the person, provides a useful framework for both research and practice. This article summarizes how these concepts took momentum in professional practice and public policies in the past fifteen years (2000-2015), discusses current issues in the design, development and economic model of AT for people with dementia, and covers how these technologies are being used and assessed.

Introduction limits of scientific plausibility; (c) reframe complex problems, find a shared language, identify research themes; (d) enable a multi-disciplinary, open, progressive debate, of high scientific level, to create exchange spaces (research networking); and (e) explore promising issues, to produce practical guidelines. Since 2012, the group has met once a year. The thematic area in 2014 was: ''Technology & Alzheimer's Disease''. The present article stems from a one-day, multidisciplinary expert group discussion initiated in December 2014 and updated in March 2017, with 20 participants from the fields of cognitive neurosciences, computer sciences, economics, ergonomics, geriatrics, neurology, occupational health, philosophy, psychiatry, public health, cognitive psychology, social psychology, social policy, sociology, and gerontology. To gather expert insight, several methods were combined. Session started with brainstorming on the needs of a person with AD. Then, 10 short oral presentations of 12 minutes each were used to update group knowledge on selected issues, followed by short questions to clarify a precise point. Technologies were discussed within the WHO framework of Functioning, disability and health. In-depth, structured group discussion with a moderator was organized in four sequences of 45 minutes. Discussion covered: unmet needs of people with dementia and their caregivers; domains of daily life activities where AT can provide help to people with dementia; the enabling and empowering impact of technology to improve their safety and wellbeing; barriers and limits of use; methods for technology assessment and ethical issues.
To gather further insight, expert discussion was extended, using input from scientific and professional literature. In this article, the angle chosen to present ATs is how they can address the capabilities of people with dementia.

The capability approach
Capabilities are defined as the opportunities that individuals have to achieve particular functionings. They are dynamically shaped by interactions between individuals and their environments, including their social relationships (Entwistle & Watt, 2013). People with dementia appear quite disadvantaged in their possibilities: their capability level (possible freedom) is comparable to that of people without cognitive disorders, but having at least five types of impairments in self-care and participation in domestic life Tellez, Krishnakumar, Bungener, & Le Gale`s, 2016). It is also known that tailoring activities to the capabilities of dementia patients, and training families in activity use, result in clinically relevant benefits for patients and caregivers (Gitlin et al., 2008). The capability approach, largely diffused by the economist Amartya Sen (Nobel Prize 1998), appears particularly relevant in person-centered dementia care and AT development. The focus is not on the solution, rather on what the person can do with it: seeing dementia as disability, with technology as an enabler to promote capabilities of the person, provides therefore a useful framework for practice. We will analyze how these concepts took momentum in professional practice and public policies in the past 15 years (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015). We will also discuss current issues in the design, development and economic model of technologies for people with dementia, and how these technologies are used and assessed.

The Human Rights based approach for people with cognitive disability
In 2002, the World Health Organization introduced the biopsychosocial model of disability ( Figure 1): a biological, individual, and social perspective of disability and functioning, where engagement into an activity is influenced not only by the medical condition and possible impairments, but also by contextual factors linked to environment and the person (World Health Organisation, 2002).
The contextual factors explain why it is essential to recognize that each person with dementia is unique: the same AT device may be enabling, useful, useless or irritating depending on the person using it, for whom it is used, and in which conditions. It is also important to consider that incapacities to perform instrumental activities of daily living (IADL), are affected since the stage of mild cognitive impairment (Jekel et al., 2015), and may influence the appropriation and use of technology. The perceived ability to use everyday technology may actually predict overall functional level in people with mild cognitive impairment or mild AD (Ryd, Nyga˚rd, Malinowsky, Ö hman, & Kottorp, 2016). More generally, tool use disorders in neurodegenerative diseases involve impairments in semantic memory and technical reasoning, besides motor capacity (Baumard et al., 2016).

The organizational approach
Industrial issues concerning technology development are discussed under the global approach of the value chain, a set of activities that an organization carries out to deliver a valuable product or service to its customers (Porter, 1985), value being defined as outcomes (ultimately, satisfaction of customer's needs) relative to cost (Porter, 2010).

Unmet needs of people with dementia
An important starting point to understand how AT can help people with dementia is to identify their unmet needs. A Dutch survey conducted on 236 community-dwelling people with dementia and 322 informal carers showed that most unmet needs are experienced in the domains of memory (support for memory problems), access to information about dementia, company, psychological distress, daytime activities, and eyesight/hearing (van der Roest et al., 2009). Reported reasons for unmet needs included a lack of knowledge about the existing service offer, a threshold to using services, and insufficient services offer.

How can technology help?
The Technology Charter for People Living with Dementia in Scotland (NHS Scotland, The Scottish Government, Alzheimer Scotland, Scottish Fire and Rescue Service, Tunstall, Tynetec, 2015) identified three ways in which AT can help people with dementia: enabling and empowering; health and wellbeing; safety and independence (Table 1). For each point, we will discuss examples of AT, and explain how they can achieve the identified needs.

(a) Enabling and empowering
In order to maintain a good quality of life, improve self-esteem and mood, it is essential that people with dementia can maintain some autonomy in ADL as long as possible, and maintain a regular social activity. AT can provide useful tools to help people affected by physical and/or cognitive decline in that sense, while reducing caregiver's burden. For example, using a mobile telephone may be possible at an early stage of AD with specific learning techniques based on procedural memory, however after a three-month rehabilitation scheme (Lekeu, Wojtasik, Van der Linden, & Salmon, 2002). An assistive dressing system has been described, which continuously monitors the dressing state of the user, identifies and prompts correct and incorrect dressing states, and provides corresponding cues to help complete the dressing process with minimal caregiver intervention (Burleson et al., 2015). A growing AT area is that of socially assistive robots, which are perceived as useful solutions by people with dementia and their caregivers (Pino, Boulay, Jouen, & Rigaud, 2015). Despite several remaining usability problems, socially assistive robots represent promising solutions to assist with communication and reduce the impact of sensory loss, to reduce isolation and disconnection from personal and social communities, and to support the ordinary activities of daily life (Kramer, Friedmann, & Bernstein, 2009;Mordoch, Osterreicher, Guse, Roger, & Thompson, 2013;Tanaka et al., 2012). Communication can also be improved by using avatars, i.e., computer programs able to interact with users thanks to user inputs acquired through a set of sensors (cameras, microphones, touch screen, etc.) and a virtual character. Current applications are able to produce human-like communication behaviors by speaking and gesturing, and can be used also to assess people's cognitive status (Nonaka, Sakai, Yasuda, & Nakano, 2012). (b) Health and wellbeing The area of AT to improve safety and wellbeing is expanding, as evidence about the effectiveness and acceptability of AT for therapeutic purposes is increasing. For instance, several studies have been recently carried out on the use of serious gamesvideogames designed for purposes other than entertaining-for the cognitive and physical stimulation of patients with AD and related disorders (Robert et al., 2014). Several serious games targeting cognitive stimulation (e.g., multi-tasking ability) have been shown to improve cognitive functions in healthy elderly people (Anguera et al., 2014) and studies are now extending to pathological populations. For instance, Ben-Sadoun et al. (2016) showed that X-Torp, a serious game designed to combine cognitive and physical stimulation, is well accepted and positively evaluated by both healthy elderly people and patients with MCI and early stages of AD. These studies suggest that serious games may represent a useful motivational tool to create entertaining trainings for patients with dementia-related disorders, to be used to complement standard pharmacological To maximize confidence, capacity, and capability of the individual to self-manage their dementia and any other health conditions they may have.
To provide individualized solutions to concerns about safety, security, and independence.
To increase choice and opportunity.
To support physical and mental health and wellbeing through the use of technology enabled care.
To reduce risk of fire and prevent accidents.
To support what is important to the person with dementia and their families and carers.
To help create or maintain routine.
To support people with dementia to walk freely and safely both indoors and outdoors. To support the ordinary everyday activities of daily life.
To support the search for information, advice, support, and help.
To encourage going out and about and reduce the risk of being lost or the risks associated with being lost. To reduce isolation and disconnection from personal and social communities.
To reduce the risks associated with falls and disorientation.
To facilitate the person with dementia to communicate with a remote partner in care should clarity or reassurance be required. To maximize existing skills and abilities and support the development of new skills.
To support reminiscence and life story work.
To automatically alert a remote partner in care to the real time occurrence of a risk to the person or property. To assist with communication and reduce the impact of sensory loss.
To support medication regimes.
To support memory, reasoning, judgment, and decision making.
To maintain social networks and enhance personal connection through social and digital media. treatments. In 2017, the X-Torp game was granted the status of medical device (Class 1-93/ 42/CE) from the French health authorities. Gamers will be able to access the game via an Internet platform. Virtual reality (VR) is now seen as another promising tool to create motivational trainings (Rizzo & Kim, 2005). VR cognitive trainings designed for people with dementia are generally well accepted, and preferred over classical paper pencil exercises even by dementia participants with scarce previous experience with ICT devices (Manera et al., 2016). Beyond supporting physical and mental health and wellbeing, VR can also be employed to support reminiscence and life story work. In a recent study we have shown that VR is a promising tool to improve reminiscence therapy in the elderly (Benoit et al., 2015): thanks to a high level of immersion in the virtual environment, which generates a feeling of presence, VR can successfully help triggering old memories. An emerging trend, which has recently generated a large public interest worldwide, is the use of VR on a mobile serious game (Sea Hero Quest) to collect a normative data set in healthy people to create a diagnostic test based on spatial navigation skills. This would help understand how the early symptoms of dementia can be differentiated from the consequences of natural ageing, to develop dementia prevention strategies and early interventions using gaming technologies (Morgan, 2016). AT supporting the search for information, advice, support, and help are also emerging, targeting both people with dementia and their formal and informal caregivers. AT in this domain include both serious games designed for caregiver training (Manera et al., 2016); web-based blogs and platforms to facilitate and improve the communication between caregivers and healthcare professionals (Span et al., 2015); websites designed to provide information on dementia-related disorders targeting caregivers, patients with dementia, and professional organizations, such as DEM-DISC, a dynamic interactive social chart for dementia care (Droe¨s et al., 2005); and e-learning platforms for carers, volunteers, and professionals, such as Star course (Hattink et al., 2015). (c) Safety and independence Several AT designed to improve safety and independence are already available on the market, and others are being developed. Social alarms can be defined as ''devices (with or without some intelligence) located in the home, which, when activated, facilitate communication with a responder and the sending of information relevant to the user's well-being'' (Fisk, 2003). These include smart smoke, gas and carbon monoxide detectors, which can be wireless connected and in case of an accident can alert people in the neighborhood through loud alarms, as well as local fire services and caregivers through automatic phone calls and text messages. The connected devices can be checked remotely for proper functioning (e.g., are they plugged in correctly? Is the battery charged? etc.). A survey conducted in England and Scotland confirmed that over 70% of elderly people have some form of social alarms installed in their homes, and that they are considered as generally acceptable (McCreadie & Tinker, 2005). These social alarms can help both to reduce the risk of fire and other common home accidents, and to automatically alert a remote partner in care to the real time occurrence of a risk to the person or property. Similar functions are covered by connected video monitoring systems. Thanks to automated video analyses techniques, video and RBD cameras can be employed to automatically detect accidents such as falls. Interestingly, algorithms are being developed to use cameras to improve the ecological assessment of ADL of older people. In the context of the European FP7 project Dem@Care, algorithms have been developed for the recognition and classification of everyday activities, such as sitting and waking up, preparing a meal, or going to bed (see Figure 2). Results collected on participants with mild cognitive impairment and dementia suggest that these algorithms are sensitive enough to accurately discriminate different ADL activities when participants follow a pre-defined activity scenario (Ko¨nig et al., 2015). They may therefore contribute significantly to home safety by creating personalized solutions enabling older adults to live longer, autonomously in their preferred environment. Customized solutions are also being developed to support people with dementia to walk freely and safely both indoors and outdoors, to encourage them to go out and about, while reducing the risk of being lost and associated risks, and to facilitate communication of the person with dementia with a remote partner in care, should clarity or reassurance be required. For instance, the COGKNOW Day Navigator (Meiland et al., 2012), developed in a FP6 European project, is a holistic solution that assists persons with dementia within their own homes and outside. It increases independence and quality of life, while giving caregivers the relief of not having to monitor the person with dementia all the time. The service can be easily customized to address the individual needs of each person and their caregivers, for memory support, helping with daily activities, maintaining social contacts and increased safety. The main functionalities are: time indication, remotely configurable reminders, music, radio, picture dialing, activity assistance, house alerts for safety, and mobile navigation for going home.

How can technology be an obstacle?
Despite that an AT can simplify daily life and compensate for disability in several ways, it can sometimes be an obstacle and lead to exclusion, when people are unable to afford it, learn how to use it, or forget how to use it (Damant & Knapp, 2015). As AT is so pervasive in our everyday life, it would be important to include assessment of AT use in the clinical assessment instruments. However, despite technology has changed our everyday activities, ADL/IADL assessment instruments remained mostly the same. For this reason, new clinical instruments assessing the ability to use everyday technology are being developed. The Everyday Technology Use Questionnaire (ETUQ; Nyga˚rd, Pantzar, Uppgard, & Kottorp, 2012;Rosenberg, Kottorp, Winblad, & Nyga˚rd, 2009) was designed to assess perceived difficulty in using everyday technology. Similarly, the Management of Everyday Technology Assessment (META;  evaluates the observed ability to manage everyday technology. Studies employing these instruments have shown that people with dementia and MCI have a greater difficulty to employ some everyday technologies, and thus the evaluation of AT actual use may be a sensitive way to detect early changes in cognitive decline (Ryd et al., 2016). Specifically, it is possible to create a hierarchy of action skills needed when handling everyday technology. This knowledge can be applied to design or adapt devices to reduce difficulty of use by people with cognitive impairment. Patomella, Kottorp, Malinowsky, and Nyga˚rd (2011) and Patomella, Kottorp, and Nyga˚rd (2013) have found that a technology is considered more difficult to use when: it is used less than once a week, has more complex design, requires more actions, requires that actions are used more frequently, and that more difficult actions are used more frequently, requires that actions are taken in particular sequences, gives more complex, mostly visual feedback and uses abstract symbols. Interestingly, this perceived difficulty appears independent of the user's age, gender, and of how long the AT had been used. The same authors found that many different aspects influence whether a piece of technology will be appropriated and used or not. These include the interests and attitudes of the person and the social context; the person's ability to use the technology; the level of challenge of the technology; and the context.

Design and development of technologies for people with dementia
Combining ICT and the social environment of older people, to support a widely selfdetermined autonomous life in their own homes, has led to the development of concepts, products and services for their daily life, under the European research program Ambient Assisted Living (AAL; Siegel, Hochgatterer, & Dorner, 2014). From 2008 to 2013, over 130 projects were launched, leveraging 620 million euros. But only a handful of solutions made it to market. This is largely explained, on one hand, by the difficulty to find a viable industrial business model, and on the other hand by the inadequation of these solutions to the real needs of end users. Although these solutions are not specific for people with cognitive disorders, they surround them and will challenge them over time in their everyday life (Rosenberg & Nyga˚rd, 2014). Three distinct research and development approaches can be observed, with distinct objectives and stakeholders involved. First, research on technologies specifically designed for people with dementia, mostly driven by multi-disciplinary academic/ industrial teams and funded by national or international grants, usually results in the production of prototypes to demonstrate practical feasibility and attract further research funding. An example of such upstream research is a validated automatic video monitoring system for the detection of IADL in people with dementia (Ko¨nig et al., 2015). A second approach is to adapt existing technologies to new potential usage by people with dementia, usually funded by technology industrials seeking first sales or commercial target extension, with eventual support from local or regional partners. To gain customer confidence, technologies are usually tested in reference centers before being promoted at a larger scale. For example, programs combining the use of GPS bracelets, databases of people with dementia, and police training have been effective in locating missing people faster (Petonito et al., 2013). A third approach is to adapt the competence of people with dementia to use everyday technology, i.e. electronic, technical and mechanical equipment, newly developed or common and well-known, in their living environment. Key professionals involved are occupational therapists and neuropsychologists. Family and volunteers may provide useful assistance to people with dementia about using the technology. One example is the use of recreational games: people with dementia, engaging with digital gaming technology in Tech Clubs, were able to challenge informal and formal caregivers' perceptions of their abilities (Cutler, Hicks, & Innes, 2016). Other examples include the use of existing electronic pads or music playing devices for reminiscence interventions or recreational purposes; and serious games, which stimulate executive functions and foster engagement in people with dementia, apathetic or not, while providing a pleasurable activity. Success or failure of these different approaches depends whether products are adapted to the capacities of people with dementia, respond to actual needs of the persons and caregivers, are endorsed by Alzheimer's associations and scientific societies, and provide relevant support services with a human assistance at the end. Co-design of technologies with advice from people with dementia and their caregivers is an emerging trend (Span, Hettinga, Vernooij-Dassen, Eefsting, & Smits, 2013). In understanding the limitations experienced by people with dementia, co-design activities allow developers to reduce challenges introduced by technology use, while empowering the persons. A positive impact is the increased caregivers awareness of their remaining abilities and reducing the potential of a person to be seen as a victim (Jury, 2016;Swaffer, 2014). AAL environments can be designed to support informal caregivers in fashioning ''do-it-yourself'' solutions that complement tacitly improvised care strategies and enable them to try, observe, and adapt to solutions over time, decide which activities to entrust to technology support, how a system should provide support and when adaptations are needed (Hwang et al., 2015).

Usages and appropriation of technologies by people with dementia
Within the biopsychosocial model of disability, technology can be viewed as an enabling tool to promote autonomy and citizenship of a person with dementia throughout daily life activities. Technology has multiple purposes, which can be summarized in a simple way, using the standards of the WHO International Classification of Functioning, Disability and Health (World Health Organisation, 2002). Nine domains of daily life activity appear relevant for technology: earning and applying knowledge; general tasks and demands; communication; mobility; self-care; domestic life; interpersonal interactions and relationships; community, social and civic life; recreation and leisure (Table 2). Learning how to use technologies is key, given their multitude and ever-changing designs. How people with dementia strive to learn and maintain their skills and knowledge related to technology deserves further attention. A pilot study, involving people with mild to moderate stage dementia, suggests that they use different ways of learning in their daily life, relying on one's habituated repertoire of actions, on other people or on technology itself, or belonging to a learning context (Rosenberg & Nyga˚rd, 2014).

Ethical, financial, and legal issues
Key ethical issues concern privacy, decision-making about the use of technology by people with dementia and financial access to technologies. In a guidance issued in 2015, the Mental Welfare Commission of Scotland states that ''efforts must always be taken to support someone to make a decision whenever this is possible. This may include taking extra time to explain what is being proposed, involving advocacy, and using communication aids to help promote discussion and understanding.'' Advance planning can help to gather adequate willingful consent from the person with dementia about technology use, before severe cognitive impairments preclude this consent (Yang & Kels, 2016). Concerning financial access from a consumer perspective, cost of technology must include acquisition and rental costs, plus support and maintenance costs. AT services are subject to major variations in pricing in a mixed economy of multiple dementia care providers, including primary care, local authorities, private companies, and local or national resources dedicated to ATs (Gibson et al., 2016). Solvency of consumers is key to facilitate purchase of AT, and can partly be compensated by public policies. Affordability of technology can been quantified using willingness to pay approaches. A recent survey of American caregivers of older people showed that the mean amount they are willing to pay is approximately $50 per month for monitoring technologies and $70 per month for technologies that would both monitor and provide some assistance (Schulz et al., 2015). Younger caregivers, those caring for a person with AD, and caregivers with more positive attitudes toward and experience with technology are willing to pay more. However, 20% of the caregivers are not willing to pay anything. Most caregivers expect some financial help from the government or private insurance to have access to technologies. Ethical issues for equal access to technologies therefore involve some part of collective funding, eligibility of users being based on individual financial and social situation.
Key legal issues in using technology concern primarily contract obligations (who should be deemed responsible in case of device or support failure causing harm to the user?), conformity of technology to human rights and individual data sharing, both in research and practice. The use of mass data for research generated by monitoring behaviors in smart homes and serious games on mobile phones constitute a novel issue. The first large-scale inhome monitoring using pervasive computing technologies to give continuous access to AD progression and intervention efficacy has been installed in Oregon (United States) in 500 homes, collecting data for up to eight years on gait and mobility, sleep and activity patterns, medication adherence, and computer use (Lyons et al., 2015). Patterns of intra-individual variation detected in each of these areas are used to predict outcomes such as low mood, loneliness, and cognitive function. The web game Sea Hero Quest intends to collect a normative data set on healthy people worldwide, to help create a cognitive diagnostic test based on spatial navigation skills. It may boost research on prevention of dementia by identifying early, subtle behavioral, cognitive, and executive changes, in a ''massive online citizen science experiment'' able to test 200 people in 1 minute (Morgan, 2016). Altogether, the development of a legal framework for AT in dementia remains a complex endeavor. It has been set as a long-term challenge (2020-2030) by the Japan Advanced Institute of Science and Technology (Sugihara, Fujinami, Phaal, & Ikawa, 2015).
Valuing technologies for people with dementia: Assessment and economic model From a person-centered approach, a technology has a value if it satisfies the needs of people with dementia and their caregivers, at an individual level. But who assesses how technology and related technology support services satisfy these needs? From an organizational point of view, the value chain of AT for people with dementia can be broadly described using 10 basic processes: research, industrial development, production, marketing, distribution, technology assessment, prescription, purchase (co-payment), device use, data sharing, user assistance, and device maintenance. These processes involve multiple professionals from many different cultures, with no clear identification of who is concerned with individual needs and technology use assessment (Table 4). The overall economic model is not yet stabilized and depends in part of collective funding to facilitate access to AT. In this emerging field, industrials are moving towards norms and quality labels of technology solutions to increase confidence of potential users, that these devices will satisfy their needs. Some Alzheimer's associations and people with dementia have started hands-on demonstration of technology devices that may be of interest. Technology assessment also involves opinion leaders running clinical trials, professional societies, care operators, central purchasers, and public funders. Occupational therapists could give relevant individual advice, as they know about activity limitations of people with dementia, if properly informed, trained, and engaged in technology assessment (Jarvis, Clemson, & Mackenzie, 2016). A predictive model to differentiate between adopters and non-adopters of everyday technology aimed at people with dementia has been described (Chaurasia et al., 2016). Validated tools for assessing the ability to use everyday technology in a standard way, such as ETUQ (Nyga˚rd et al., 2012;Rosenberg et al., 2009) and META ) deserve a larger diffusion in clinical practice .
From a broader perspective, should society as a whole invest in AT for people with dementia? In other terms, does AT for people with dementia bring social innovation, which means a novel solution to a social problem that is more effective, efficient, sustainable, or just than current solutions? If so, the value created would accrue primarily to society rather than to private individuals (Phills, Deiglmeier, & Miller, 2008). The approach of social return on investment is emerging. For example, in England, lifetime costs of dementia, including publicly and privately financed health and social care, average 200,000 sterling pounds per person and per year from diagnosis to end-of-life. A technology, along with accompanying assessment and services, would be cost-effective from a societal perspective if it either reduced unpaid care hours by about 8% or improved caregiver quality of life by 0.06 to 0.08 quality-added life-years (QALYs) per year (Knapp, 2016).

Discussion
Development of AT to address capabilities of people with dementia has been shaped in the past decade by major conceptual shifts about daily-life risk perception; viewing dementia as a (potential) disability, which led to enable and empower people with dementia; a more positive communication about technologies, putting benefits forward; and understanding the appropriation process of AT, building on the individual experience of people with dementia.

Changes in daily-life risk perception
Performing an activity of daily life may expose the person with dementia to a number of hazards, minor or harmful. Technology may reduce the occurrence of such events. Within a   (2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015), the notion of risk has evolved and broadened: new expressions emerged, such as positive risk-taking and risk enablement, to account for potential benefits as well as disadvantages upon risk evaluation, to distinguish risk from danger. As the British Department of Health stated in a landmark risk guidance for people with dementia titled: ''Nothing ventured, nothing gained'', risks must be balanced with opportunities that are likely to follow risk-taking (Department of Health, Manthorpe & Moriarty 2010). However, professionals, carers, and people with dementia consider risk differently.
Understanding different ideas about risk is the first step in developing an effective risk mitigation strategy. Decisions about risk should not assume that all people with dementia are ''at risk'' but should be based upon understanding of individuals and their own understanding of the risks they may face. Much early research and communication about technology in dementia care has covered barriers to be overcome, raising concerns about potential detrimental effects, such as intrusion on privacy, stigma (with reference to tracking devices), reduced human contact, or reduced use of paid caregiving. According to the Nuffield Council of Bioethics, these issues have the potential to affect both a person's autonomy, for example through feeling controlled or devalued, and their well-being, for example through impoverished human relationships. Where the persons with dementia have the capacity to choose for themselves whether to accept or refuse a particular technology, their decision should be respected. Where a person with dementia lacks the capacity to decide for themselves whether to make use of a particular technology, the relative strength of a number of factors should be considered on an individual basis, including the person's own views and concerns, past and present, the actual benefit which is likely to be achieved through the use of technology, and the extent to which the caregiver's interests may be affected (Nuffield Council of Bioethics, 2009).

Dementia as a (potential) disability
In 2010, the European Dementia Ethics Network stated that it was essential to recognize dementia as a disability (or potential disability), as the first step towards ensuring the appropriate design of AT to be used by or for people with dementia, and towards ensuring that people with dementia have access to it when it would be of most benefit to them (European Dementia Ethics Network, Alzheimer Europe, 2010). Persons with disabilities include ''those who have long-term physical, mental, intellectual or sensory impairments which, in interaction with various barriers, may hinder their full and effective participation in society on an equal basis with others'', according to the United Nations Convention on the Rights of Persons with Disabilities, an international legal framework ratified by 163 countries. This Human Rights based approach for people with cognitive disability is a new policy currently being promoted by the advocacy group Dementia Alliance International, chaired by Kate Swaffer, a young Australian woman living with dementia, and Alzheimer's Disease International (Dementia Alliance International, 2016). However, only a small number of countries do legally recognize dementia as being a disability, which would facilitate access for people with dementia to appropriate support, including ATs. In most countries, disability allowance is actually considered as a compensation benefit for people who are unable to work as a result of their incapacity. As dementia usually develops far beyond the legal limit of working age, the disability approach is rarely considered for them. Scotland has been the first country to have included people with dementia in its implementation of this UN Convention and to have appointed in 2002 a Scottish Dementia Working Group of people with dementia to advise the government, with the support of Alzheimer's Scotland.
A positive approach to AT for people with dementia In 2015, the Technology Charter for people living with dementia in Scotland introduced a globally positive approach to technology for people with dementia and their partners, putting benefits forward: enabling and empowering, health and well-being, and safety and independence (NHS Scotland, The Scottish Government, Alzheimer Scotland, Scottish Fire and Rescue Service, Tunstall, Tynetec, 2015). Developed in consultation with people with dementia, it provides a general framework, easy to understand for all stakeholders, setting up values and principles as a guide to take position on ethical dilemmas. It stems from a dual perspective: the rights of the people living with dementia, and the commitment and responsibility from the care partners (Table 3). This approach is supported by notions of positive and shared risk-taking, freedom of choice, acknowledgement of strengths and experience of the person with dementia. The Charter also states that technologies in all their forms and applications are an integral part of life today, not just because they are available, but because people find them useful.

Usage and appropriation of AT by people with dementia
Yet, few professionals are studying usages of technology, and how people with dementia appropriate it for themselves. A recent review of commercially available AT for people with dementia identified 171 products and 331 services, which can be grouped in products used ''by'' people with dementia, ''with'' them or ''on'' them (Gibson et al., 2016).
More and more, people with dementia express themselves on blogs: most of them use nonspecific technologies, such as smartphones or computers, generally with a positive experience. Negative experiences are about remembering passwords, or the purpose of the different alarms. Under stress conditions, people with dementia tend to disregard AT devices sensory cues (Teipel et al., 2016). Caregivers also use non-specific technologies, such as electronic pads, baby monitoring systems, smartphones, light sensors, and pillows equipped with loudspeakers. Furthermore, technologies change all the time and may be perceived as more difficult to use by people with dementia (Malinowsky, Kottorp, Patomella, Rosenberg, & Nyga˚rd, 2015). A major barrier to technology use is therefore the difficulty to obtain counseling and support (Knapp, 2016). In-depth interview with people with mild cognitive impairment shows that persons with mild cognitive impairment may relate to technology in various ways to meet needs of downsized doing, but are reluctant to adopt video-based monitoring technology intended to support valued occupations (Hedman, Lindqvist, & Nyga˚rd, 2016). People with dementia living in institution appear ambivalent about surveillance technology: on one hand, it gives them freedom to venture into new spaces; on another hand, they may be reluctant to use the technology provided, as they do not like to be observed (Niemeijer, Depla, Frederiks, & Hertogh, 2015). Concerning recreational use of technologies, a key outcome is the pleasure of people with dementia as a positive experience: pleasure of being enabled, pleasurable moments with significant others, pleasure of just enjoying, playing, and having fun. The use of robotic animals may be beneficial, when introduced prudently after in-depth discussion with persons with cognitive disorders and families, to anticipate potential ethical dilemmas that may arise about values such as honesty, dignity, well-being, and personal choice (Gzil, 2015).
An emerging issue concerns the use of technology to create simulated, fictitious environments for reminiscence purposes, aimed at reducing anxiety and stress of people with dementia, such as a VR forest (Moyle, Jones, Dwan, & Petrovich, 2017), or a fake vintage train station and compartment, with a large television screen as a window, to create the illusion of a real railroad journey. Among people with dementia, the prevalence of those living with different realities and beliefs varies greatly from around 20% to 70%, in small part depending on the type of dementia a person has (Stokes & Kousoulis, 2017). Dilemmas with carers arise when the two realities meet. Truth or untruths can both cause unnecessary distress, but they can also create joy or necessary negative emotions. Approach must be flexible: person first, intervention second. Each perceived reality shall be explored with the carer, to make sense of the world together (Williamson & Kirley, 2014).
Access to technologies involves information, hands-on testing and learning, rapid and flexible procurement, and affordability. In the perspective of people with dementia and their caregivers, Alzheimer's Scotland has developed an initiative to find, test, and share information about domestic products to help people with dementia (clocks, telephones, mobile applications, etc.) These products are tested by the persons themselves and exposed in the resources centers of the association (Gray, 2016). Learning and using everyday technology are intertwined processes. It is therefore important to support continued use of everyday technology as long as it is valued and relevant to the person with mild cognitive impairment or AD (Rosenberg & Nyga˚rd, 2014). Rapid and flexible procurement of AT is required to provide a timely response to the changing needs of people with cognitive disability at the different stages of severity (European Dementia Ethics Network, Alzheimer Europe, 2010).
Finally, two outstanding examples of usage and appropriation of technologies to empower people with dementia, on an individual and collective level, are worth mentioning. On an individual level, Masahiko Sato (2014), in the first book ever written by a person with dementia in Japan, explains how, diagnosed with early-onset AD in 2005, he was able to live actively and independently during 10 years before moving to a retirement home in 2015, partly thanks to AT. He had worked as a systems engineer for a computer company before his diagnosis. He started using social media and an iPad a few years after he was diagnosed with AD. He made it a habit to turn his computer on and check his Google calendar when he gets up every morning, not to lose track of time. Before he goes out, he sets several timers on his mobile phone around the scheduled appointment. Using a train timetable and route finder software, he makes sure he gets an alert on his phone while on the train so he won't forget where to get off. He has more than 1,300 friends on Facebook, and makes frequent speeches around the country to share his experience. He recommends that other patients start using such tools in the early stages of their illness. On a collective level, Dementia Diaries is a UK-wide project that brings together people's diverse experiences of living with dementia as a series of audio diaries. It serves as a public record and a personal archive that documents the day-to-day lives of people living with dementia, with the aim of prompting a richer dialogue about the varied forms of the condition. Project uses 3D-printed mobile handsets which are customized to be as simple as possible (Woodall, Suur, Kinsella, & Bunyan, 2016).

Conclusion
ATs are part of a global approach of non-drug interventions in dementia care. They deserve further attention in both research and practice. Within a psychosocial model of disability, AT can be viewed as an enabling tool to promote autonomy and citizenship of a person with dementia throughout daily life activities, in a positive way. This approach has started to influence public policies. Knowledge about these technologies must be developed among family and professional caregivers. Design and adaptation of AT must be based on individual needs of people with dementia. Validated tools for assessing the ability to use everyday technology by people with dementia deserve diffusion in professional practice. To be effective and respectful of people with dementia, technological innovation must build upon social innovation, with increased input from human and social sciences.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Fondation Me´de´ric Alzheimer, a non-profit organization recognized as being of public interest. chief editor of the national and international Press Review of the Fondation. This knowledge base, unique in its kind, established more than 10 years ago, allows prospective watchkeeping of the context of AD in its scientific, medical, technological, social, cultural, professional, political, economic, legal aspects, and spotting of innovative initiatives worldwide. He also leads the prospective expert groups of the Fondation Me´de´ric Alzheimer. His current work focuses on the sensory approach in Alzheimer's disease and on technologies for people with dementia and their caregivers, early diagnosis and prevention. Email: kenigsberg@med-alz.org.
Jean-Pierre Aquino, MD, specialized in Geriatrics and Public Health, is head physician of the Porte verte Medical Clinic in Versailles, technical advisor of the Fondation Me´de´ric Alzheimer, he has been president of the French Society of Geriatrics and Gerontology (SFGG). Under these different titles, he participated in the Me´nard Commission's work and in several working parties of the French Alzheimer Plan 2008-2012. As a President of the Committee on Advancing in age, prevention and quality of life, he provided a report to the government titled ''Anticipate for a preserved autonomy: a societal issue'', used to prepare the upcoming Law for adapting the Society to Ageing. Email: aquino@med-alz.org.
Alain Be´rard, MD, specialized in Public Health, economist and jurist, is Deputy Director of the Fondation Me´de´ric Alzheimer where he also manages the Coordination, Prospective and Strategy unit. His research work and publications focus on the needs of people of 65 years old and above, on the different professions working with people with dementia, on the social and asset management behaviors of people with young-onset dementia, and more recently on sensory impairments and Alzheimer's disease. He also performs evaluation of health, social and medico-social services for modeling and dissemination purposes. Email: berard@med-alz.org.
Franc¸ois Bre´mond is research director at Inria and he is leading the STARS team at INRIA Sophia Antipolis. He designs and develops generic systems for dynamic scene interpretation. The targeted class of applications is the automatic interpretation of indoor and outdoor scenes observed with various sensors and in particular with static video cameras. These systems detect and track mobile objects, which can be either humans or vehicles, and recognize their behaviors. He is particularly interested in filling the gap between sensor information (pixel level) and recognized activities (semantic level). In 1997 he obtained his PhD degree at INRIA in video understanding and Franc¸ois Bre´mond pursued his research work as a post doctorate at USC on the interpretation of videos taken from UAV (Unmanned Airborne Vehicle). He has also participated to many European and industrial research projects in activity monitoring. Franc¸ois Bre´mond is author of more than 140 scientific papers published in international journals or conferences in video understanding. He is a co-founder of the Nice University team CoBTeK dedicated to the studies of ICT technologies to better understand behavioral disorders. More information is available at: http://www-sop.inria.fr/members/Francois. Bremond/. Email: francois.bremond@inria.fr. health conditions. He undertook work on the Olympic Build, evaluating their Occupational Health facility and also for the Social Care Institute for Excellence (SCIE) where he managed a research project to develop guidance for care homes that wished to use ICT to improve the lives of people with dementia. After this project, he decided to undertake a funded PhD at Bournemouth University Dementia Institute (BUDI), exploring the impact of social clubs for older men with dementia in rural Dorset, England. These social clubs run activities for the men using commercial gaming technology such as iPads, Nintendo Wii, and Microsoft Kinect. Email: bhicks@bournemouth.ac.uk.
Anthea Innes, PhD in Sociology, is University of Salford's first professor of Dementia. She worked as a senior lecturer in Dementia Studies at the Department of Applied Social Science of the University of Stirling, before being the director of launched the Bournemouth University Dementia Institute. Her work explores the process of changing the culture of dementia care in care settings, of dementia provision in rural Scotland and how to elicit service users' views. She was a member of the groups advising the Prime Minister on rural dementia care and dementia friendly technology, while she also introduced the first worldwide postgraduate online program in Dementia Studies in Stirling. Email: anthea.innes@uws.ac.uk.
Sao-Mai Nguyen is associate professor at Institut Mines-Te´le´com Atlantique-Bretagne-Pays de la Loire (Brest, France). After graduating from Ecole polytechnique in Informatics, she led research in Osaka University in Cognitive Robotics. She studied how robotics could help study cognitive impairments, with a focus on the theory of mind and autism. She then completed her PhD at INRIA on learning strategies for robots based on active learning and imitation learning. With Pierre-Yves Oudeyer, she studied how to combine curiosity-driven exploration and socially guided exploration to learn motor skills into an interactive learner. After two postdocs in Queen Mary University of London and University of Plymouth in computational neuroscience, she tries to bridge robotics, artificial intelligence, and cognitive science. Her interests lie in linking research of machine learning, robotic embodiment, developmental psychology, and neuroscience to make information and communication technology improve the everyday lives of the elderly and the disabled. Email: mai.nguyen@telecom-bretagne.eu.
Louise Nyga˚rd, PhD, is professor of occupational therapy at the Division of Occupational Therapy at Karolinska Institutet (KI) in Stockholm, Sweden. She has conducted research in the field of dementia since 1990, with a particular interest in the experiences of persons with dementia, especially in the early and moderately severe stages. She finished her PhD at KI in 1996. Since then, she has established a research group of about 20 persons, Cognitive ACcessibility and Technology Use when aging in home and Society (CACTUS). She has published over 70 peer-reviewed papers and also reports and chapters in international and Swedish books. In recent years, her particular interest has been in the conditions for people with Mild Cognitive Impairment (MCI) or Alzheimer's disease (AD) as users of technology. Her research group has explored how people with MCI/AD manage the common everyday technology of today's society and how they can be supported by assistive technology. Her group has developed two assessment instruments of ability to use technology. These are being validated and increasingly used in both characteristics. She is also interested in dementia care and assistive technology for dementia. Email: hg.vanderroest@vumc.nl.
Herve´Villet, MD, specialized in Public Health and Master in Health Economics, is assistant manager of the Observatory of Alzheimer's Disease Care Services at the Fondation Me´de´ric Alzheimer. For the past 10 years, the Observatory has been conducting national surveys on the different services, in order to produce regular statistics and maintain a service directory, as well as thematic surveys on modalities of care and professional practices. His current work concerns the legal protection of people with Alzheimer's disease, social activities in gerontology and specific Alzheimer units. A national survey on sensory impairment and dementia will be launched in 2015. Email: villet@med-alz.org.
Marion Villez, PhD in Sociology, is lecturer at the Interdisciplinary Research Laboratory on Education and Social Practice Transformations (LIRTES) of the University of Paris-Est-Cre´teil-Val-de-Marne (France). She worked previously as a Local Initiatives Department Manager at the Fondation Me´de´ric Alzheimer. Her mission consisted in building up innovative issues to promote, support, and evaluate new initiatives dedicated to people with dementia, their caregivers and professionals all around the country. Her PhD thesis focused on design and care practices of people with cognitive disorders in nursing homes. Email: villez@med-alz.org.
Philippe Robert is professor of psychiatry at the Nice School of Medicine, Director of the Nice Memory Centre for Care and Research, Director of the Cognition, Behaviour & Technology Unit (CoBTeK) at the Coˆte-d'Azur University in Nice (France). He is the coordinator of the French National Alzheimer data bank. The CoBTeK unit aims to develop and transfer new technologies, developed for the benefit of clinical medicine and most particularly elderly subjects and others patients suffering from neuropsychiatric / neurodevelopmental disorders. CoBTeK provide an optimal partnership between engineers (INRIA) and clinicians (Memory center, CMRR) corresponding to the ''ICT or computer-to-clinic'' link. Email: robert.p@chu-nice.fr.
Valeria Manera is a postdoctoral fellow at the CoBTeK unit of the Coˆte-d'Azur University in Nice (France). She has a Master degree in Clinical Psychology and a PhD in Cognitive Science. She spent long research periods at the University of Leuven (Belgium) and at the Stanford Psychophysiology Laboratory (California). Her domains of expertise include biological motion perception, emotion recognition and regulation, and dementia prevention. She is researcher in the French team of the In-MINDD project. Email: valeria.manera@unice.fr.