Content uploaded by Amanda Jane Sharkey
Author content
All content in this area was uploaded by Amanda Jane Sharkey
Content may be subject to copyright.
1
Granny and the robots: Ethical issues in robot care for the elderly.
The growing proportion of elderly people in society, together with recent advances in robotics, makes
the use of robots in elder care increasingly likely. We outline developments in the areas of robot
applications for assisting the elderly and their carers, for monitoring their health and safety, and for
providing them with companionship. Despite the possible benefits, we raise and discuss six main
ethical concerns associated with: (i) the potential reduction in the amount of human contact; (ii) an
increase in the feelings of objectification and loss of control; (iii) a loss of privacy; (iv) a loss of
personal liberty; (v) deception and infantilisation; (vi) the circumstances in which elderly people should
be allowed to control robots. We conclude by balancing the care benefits against the ethical costs. If
introduced with foresight and careful guidelines, robots and robotic technology could improve the lives
of the elderly, reducing their dependence, and creating more opportunities for social interaction
Introduction
Should we be concerned about the use of robots for elder care? It is often
suggested that one way of coping with the increasing proportion of elderly people in
our society is to use robotics and related technology. Indeed, there is a growing
industry in developing robots for elder care. However, some ethical concerns are
beginning to be expressed about these developments (e.g. Sparrow and Sparrow,
2006; Sharkey and Sharkey, 2010; Sharkey, 2008; Wallach and Allen, 2009). There
is concern, for instance, that using robots for elder care could result in increased social
isolation, and could involve deception and loss of dignity. At the same time,
researchers have provided some evidence of benefits for the elderly, particularly as a
2
result of interacting with robot pets (Banks et al, 2008; Kanamori et al, 2002; Tamara
et al, 2004).
In an effort to determine the likely effects of robots on the lives of senior
citizens, we outline recent developments, and identify and discuss the ethical issues
they raise. The three main ways in which robots might be used in elder care are: (i) to
assist the elderly, and/or their carers in daily tasks; (ii) to help monitor their behaviour
and health; and (iii) to provide companionship. It will become apparent that in each
of these areas there are some reasons to fear that the introduction of robots could
reduce the quality of life of elderly people. However, it will be argued that such
effects are not inevitable, and robots and robotic technology could, if introduced
appropriately, solve a number of the problems that elderly people face.
In our assessment of robots in elder care, we consider two different bases for
the associated ethical concerns: human rights, and shared human values. We shall
outline these in turn. An emphasis on human rights provides support for the
assumption that the physical and the psychological welfare of the elderly is as
important as the welfare of others. Amongst the fundamental human rights
established in documents such as the Charter of the United Nations, and the Universal
Declaration of Human Rights, those thought of as particularly relevant to the elderly
include the right to a standard of living adequate for health and well-being; to private
and family life; to freedom from torture and inhuman or degrading treatment; and to
freedom from discrimination. If the human rights of the elderly are to be respected as
much as the rights of other members of society, it is important to ensure that robots
introduced into elder care do actually benefit the elderly themselves, and are not just
designed to reduce the care burden on the rest of society. A human rights
3
interpretation of our opening question, “Should we be concerned about the use of
robots for elder care?” implies a consideration of the effect of robot care on the
elderly themselves, as well as on their family and carers.
Human rights are inextricably related to morality, but they are more often
discussed in terms of legislation, than practical ethics. A different way of grounding
our deliberations is to consider the effect of robots in elderly care in terms of enduring
human values, as discussed in the Value Sensitive Design approach (Friedman and
Kahn, 2003; Friedman, Kahn and Borning, 2006).
Friedman and Kahn (2003) identify and discuss 12 human values that are
implicated in technological design: human welfare; ownership and property; privacy;
freedom from bias; universal usability; trust; autonomy; informed consent;
accountability; identity; calmness; and environmental sustainability. In this paper, we
attempt to identify the particular ethical concerns and human values implicated in
robot care for the elderly. The most obviously relevant, and fundamental, human
value that relates to robot care of the elderly is that of human welfare. It seems
reasonable to assume that there is shared human concern about the physical and
psychological welfare of the elderly. If robots were found to have a negative effect on
their wellbeing, it would provide a clear reason to object to their use. Accordingly,
we take concern about the possible impact of robots on the welfare of the elderly as
our starting point.
In considering the ethical issues raised by various applications of robotics in
elder care, our approach differs from that advocated in Value Sensitive Design, and
other approaches that propose solutions and methods for the ethical design of
technology. We are not trying to provide a complete account of how robots can be
4
introduced without associated ethical problems. Our goal is a more preliminary one:
to identify the probable risks, and possible advantages, of different forms of robot use.
Doing so is a necessary first step towards the development of ethically satisfactory
solutions. There is a pressing need to identify the likely effects of robot care before it
becomes commonplace.
One of the main reasons to expect that robots will be used in the care of the
elderly is that the number of elderly people in the population is beginning to overtake
the numbers of young people able to do such caring. Japan has, for some time, had
an eye to using machines to fill this care gap. Because of a low birth rate and long life
expectancy, their population is aging rapidly, with 22% over the age of 65 in 2009,
predicted to rise to 34% by 2035.1 The worry is that there will not be enough young to
look after the old and part of the Japanese government’s health policy is to use robots
for care. In March 2009, a Japanese ministry of trade and industry official, Motoki
Korenaga, told Agence France-Presse that, “Japan wants to become an advanced
country in the area of addressing the aging society with the use of robots."2 He spoke
of plans to prepare safety rules soon for robot nurses that are expected to serve elderly
needs within the next five years.
Other countries may well wish to follow suit. Europe and the United States are
facing similar aging population problems over a slightly longer time scale. In 2009, it
was estimated that 16.2% of the population in the UK was aged 65 or older (CIA
World Factbook). Spain and Italy are the oldest in Europe with 18.1%, and 20.2%
over 65 respectively (CIA World Factbook). These figures are increasing sharply. In
1 http://www.reuters.com/news/globalcoverage/agingjapan
2 http://www.google.com/hostednews/afp/article/ALeqM5juSqhZryHpsVuY6mf93nr92g1qdA
5
the UK, the fastest growing age group is made up of those aged 80 years and over
who in 2009 constitute 4.5% of the population. In the US, 12.8% are over the age of
65, expected to rise to around 20% by 2030. It seems likely that Europe and the US
may want to follow the Japanese lead into robot care.
Whether or not robots are the best way to provide it, the frail elderly are
clearly in need of care. Some lucky individuals manage to remain mentally and
physically active until an advanced age, but there are many who need assistance in
their daily lives and some who need complete care. The mobility problems of older
people can mean that they need help to perform daily tasks in the home, and that they
are not able to go out shopping, or to visit people. Some form of memory loss is
inevitable with increasing age, and dementia is a prevalent problem. As well as
lacking the physical ability to perform tasks, elderly people may need to be reminded
of the need to perform various activities, or reminded of the locations for, and
components of, those activities. Elderly people can be in danger of falling over, or
becoming ill, or confused, or lost. They also need company, love and attention. In
this paper, we provide an account of ways in which robots can be used in the care of
the elderly under three headings: (i) Assistance; (ii) Monitoring, and (iii)
Companionship. Our aim is to raise and discuss the ethical issues involved in each of
these areas, and to consider their likely effects on the elderly themselves.
Assistive robots and elder care
6
A number of robots have been developed to assist with some of the daily tasks
involved in eldercare. Such robots include the Japanese Secom “My Spoon”
automatic feeding robot, and the Sanyo electric bathtub robot that automatically
washes and rinses. Riken have developed the Riba robot (Robot for Interactive Body
Assistance): it has a teddy bear face, and can pick up and carry humans from a bed to
a wheelchair. It can recognise faces and voices and respond to spoken commands. In
the US, Charlie Kemp at Georgia Tech, has developed an assistive robot, EI-E that
can respond to many of the same commands, and perform the same tasks in a house
that a guide dog is capable of. When coloured towels are tied to doors, and drawers,
the robot can be instructed to open them, using a combination of verbal commands,
and a laser pointer to indicate the target location. It can pick up quite a wider variety
of objects from flat surfaces (Nguyen et al, 2008).
There are some interesting new developments arising from robotic technology
that could improve the lives of the elderly. Cyberdene has developed the Hybrid
Assistive Limb suit (HAL) which is currently available to rent in Japan. It detects
nerve signals sent by a person attempting to move, and then is said to automatically
move the muscle as the person expects. It can apparently multiply original strength
by a factor of 2 to 10. There are also walking machines from Honda: ‘Stride
Management Assist’, and their ‘Walking Assist’ device with body weight support
system (New Scientist online, November 2008), although these are some way away
from mass production. Wheelchairs for the elderly and the disabled are becoming
increasingly sophisticated. For example, at MIT, Nicolas Roy and colleagues are
working on an interface to a wheelchair that will enable it to take the user to a
7
designated known location in an indoor environment, on vocal command (Doshi and
Roy, 2008).
These developments appear promising, but we cannot assume that all assistive
robotic technology will be a good thing for the elderly people who encounter it.
There are two main ethical concerns about the use of assistive robot care for the
elderly and its effects on their welfare – first that it might reduce the amount of human
contact that the elderly have, and second that if used insensitively, it could increase
senior citizens’ feeling of objectification and a lack of control over their lives. We
consider each in turn.
One of the problems of aging is that it often results in a loss of social life and
human contact. The worry is that the use of robots in elder care for tasks such as
lifting, carrying, or even cleaning, might result in a reduction in the amount of human
social contact that an elderly person experiences. Sparrow and Sparrow (2006), in
their consideration of robots for the elderly concluded that if robots were used for
tasks such as cleaning floors, this would remove a valuable opportunity for social
interaction between a human cleaner, and the senior citizen, and as such should not be
encouraged. Sharkey and Sharkey (2010) make a related argument about “dull and
dirty” tasks in the context of robot childcare – pointing out that these tasks often
provide the opportunity for social interaction and bonding. It might be convenient to
have an automated spoon feeding a frail elderly person, but this would remove an
opportunity for detailed and caring human interaction.
Depriving senior citizens of social interaction with their fellow humans is an
ethical issue that is not explicitly addressed by human rights legislation. Such a right
is perhaps implied such as in Article 5 of the Universal Declaration of Human Rights,
8
“No one shall be subjected to torture or to cruel, inhuman or degrading treatment or
punishment.”, or Article 9, “No one shall be subjected to arbitrary arrest, detention or
exile.
Concern about the level of social contact experienced by the elderly follows
from concern about their welfare. It might seem obvious that the elderly need contact
with fellow human beings, and that their welfare would suffer in its absence.
However many elderly people still live quite isolated lives, and it seems worth
considering some of the evidence that social interaction, or its absence, affects both
the physical and psychological well-being of the elderly.
For instance, there is research that suggests that an extensive social network
offers protection against some of the effects of aging: being single and living alone
has been shown to be a risk factor for dementia (Fratiglioni et al, 2000). Saczynski et
al (2006) in a study of 2513 Japanese-American men, found that decreased social
engagement from midlife to late life was associated with an increased risk of
dementia. They speculate that social engagement reduces the risk of dementia by
reducing stress, and consequently by reducing the hormones associated with stress
(glucocorticoids and corticosteroids). Social and physical activity may also help
because it “increase[s] a person’s ability to tolerate brain pathology through enhanced
synaptic activity and more efficient brain recovery and repair” (Saczynski et al, 2006).
Wilson et al (2007) studied the relationship between loneliness and the risk of
Alzheimer’s disease in 823 senior citizens in Chicago over a five year period. Those
with a higher loneliness measure were more likely to develop Alzheimer’s disease.
Holtzman et al (2004) found that frequent interaction in larger social networks was
positively related to the maintenance of global cognitive function. They suggest that
9
the effect might be the result of experiencing more novelty and variety, and from
handling more complexity. Wang et al (2002) similarly found evidence that a rich
social network may decrease the risk of developing dementia, and concluded that both
social interaction and intellectual stimulation play an important role in reducing such
risks.
There is further evidence of the beneficial effects of social contact on stress
levels. Stress has been shown to exacerbate the effects of aging (Smith, 2003), and
recent research shows that social contact can reduce the level of stress a person
experiences. Kikusui et al (2006) provide a wide ranging review of the phenomena of
social buffering, whereby highly social mammals show better recovery from distress
when in the company of conspecifics. Social contact has a positive influence on both
psychological and physiological aspects of social animals, and can be shown to
reduce neuroendocrine stress responses in humans, rodents, birds, and non-human
primates. It seems that social contact can have demonstrable biochemical effects on a
person’s body. A recent review (Heinrichs et al, 2009) concludes that the stress
protective effects of social support may be the result of the oxytocin that is released in
response to positive social interactions, and that oxytocin can have the effect of
reducing stress. For instance, Heinrichs et al (2003) found that when men were
subjected to psychosocial stress (by means of the Trier Social Stress Test), their stress
response was reduced by the presence of their best friend, and reduced further if
oxytocin was also (intra-nasally) administered to them.
The research reviewed here shows both the beneficial effects of social contact,
and attests to the increased stress and cognitive decline that can result from a loss of
social interaction. Such evidence adds further weight to the concern that robots might
10
result in the elderly having less human contact. It seems that reduced social
interaction can have a measurable impact on the health and well-being of the elderly,
and reinforces the idea that depriving them of such contact is unethical, and even a
form of cruelty. Of course, what is not yet fully understood, and this is a question to
be returned to in the section of robot companions, is the extent to which reduced or
missing interaction with humans can be compensated for by interaction with robots.
There is another significant ethical concern to be raised in the context of
assistive robotics, which concerns the problem of objectification of the elderly, and
the way in which robots carry out elder care tasks. Who controls the robots? Are they
are actually designed to help the elderly person, or to cut costs and reduce the
workload of their carers? Often the focus is more on improving the lives of the
caregivers, rather than ensuring that robotic assistance is provided in such a way as to
improve the lives of the elderly themselves. Robots designed as replacement nurses
or carers that carry out some of the same tasks of feeding, lifting etc., may make their
charges feel like objects. Such robots could make elderly people feel that they had
even less control over their lives than when they are dependent on human nursing
care. Kitwood (1997) identifies “objectification” as one of the problems faced by
those with dementia, and the same point surely applies to the frail elderly in general;
“Objectification – treating a person as if they were a lump of dead matter: to be
pushed, lifted, pumped or drained, without proper reference to the fact that they are
sentient beings”. If robots were to be used by human carers to lift and move elderly
people insensitively, the well-being of those elderly people would be likely to be
reduced.
11
Loss of human contact and increased objectification are examples of the
negative impact that assistive robots could have on the lives of the elderly. At the
same time, it is still possible that robotic technology could be utilised in such a way as
to improve the lives of the elderly. Robots could be introduced with the aim of
reducing senior citizens’ dependence on the people that look after them. This could
help, because it is unfortunately the case that human carers do not always treat the
elderly with sufficient respect for their human rights, as discussed in the Joint
Committee report on ‘Human rights and elder care’(House of Lords, House of
Commons Joint Committee on Human Rights, 2006-7). The following example is
taken from the report, and illustrates the kinds of problem that can arise when elderly
people are entirely dependent on busy nursing staff. It comes from an account given
by a woman of her experience visiting her elderly husband in hospital:
“I went to visit my husband on the first day and he is a very private person, he
doesn’t like anything to embarrass him and when I went in he was almost in tears
which is not my husband. He said ‘please, please go and get a bottle I am nearly
wetting myself’. I rushed out I got a bottle and I said to him ‘Well why didn’t you just
ring the nurse’, in my innocence. ’I have, for an hour and a half I’ve been asking for a
bottle’. Well when I went out [and] told the nurse she said ‘Oh don’t worry we would
have changed the sheets’. Now his dignity at that stage would have gone out of the
window. There was no dignity.” (Pg 15)
Dignity, as referred to in this quotation, is a difficult concept to define,
although intuitively quite easy to appreciate. It is related to the value of human
welfare, since a person’s psychological welfare is going to be affected by their
physical needs failing to be met in this way. Dignity is also closely related to human
12
rights (Schacter, 1983). However, the point we wish to make here is that such
situations could potentially be ameliorated through the use of robot technology
developed with the aim of increasing the autonomy of the elderly, and decreasing
their dependence on other people.
Robot technology that was under the control of an elderly person could
empower them and increase their independence. For example, a robot wheelchair that
could be summoned by an elderly person and instructed to take them to the toilet
could greatly improve their sense of control over their environment, and reduce the
occurrence of distressing situations such as the one described in the example above. It
might even turn out that, given the choice, some of the frail elderly might prefer
robotic, as opposed to human, assistance for certain intimate tasks such as toileting, or
bathing. The robotic technology could be developed more as a tool, rather than as an
identifiable robot. It could even begin to function as an extension of the elderly
person’s body and mind as discussed by Clark and Chalmers (1998), and become
“invisible in use” as discussed by Takayama (2010). The exoskeleton suit referred to
above provides a good example – a frail elderly person’s mobility and command of
the environment would be considerably enhanced by such a suit. Empowering the
elderly in this way and increasing their autonomy is likely to improve both their
psychological and physical welfare (see Langer and Rodin 1976 for a classic example
of the beneficial effects of greater control on the elderly, albeit not one enabled
through robotics). As well as giving the elderly an increased sense of control and
autonomy, robotic assistive technology could increase the social contact the elderly
person experiences, by making it possible for them to get to and from social meeting
places; again with likely improvements in their psychological welfare.
13
Assistive technology could improve the lives of the elderly, but it also
implicates ethical concerns about autonomy. How much control, or autonomy, should
an elderly person be allowed. To what extent should the answer to such a question
depend on medical assessments of their mental capability? The right balance would
need to be found between empowering an elderly person by making them mobile, and
protecting them from the dangerous situations they might encounter as a result. If a
senior were to request that a robot throws them off the balcony, should the robot carry
out that command? Answering such questions with respect to the elderly is
particularly complex because their cognitive, and physical, abilities cannot be
assumed to remain at a particular level. It is likely, but not inevitable, that with
increasing age a senior might become increasingly confused, and that their
movements, particularly if facilitated by exoskeletons and the like, should perhaps be
restricted in some way. Again, we are not offering a solution here, (although perhaps
some form of driving test would be appropriate), but identifying the areas of concern
for which solutions are needed. There is a delicate balance here to be struck between
improving the psychological wellbeing of the elderly by granting them greater
autonomy and independence, and between protecting their physical welfare and
keeping them from danger.
The issue of accountability should also be mentioned. There is obviously a need
to ensure that robots used by the elderly are safe – and that if, for example, they are
asked to pick a senior up, they do not miss the bed or chair, and drop them on the
floor, or crash them into the wall. Furthermore, in a system in which a robot is
responding to the commands of an elderly person, who or what should be held
responsible and accountable if something goes wrong, resulting in injury or damage?
14
If a senior citizen in an exoskeleton suit were to kick and injure a nurse, should they,
or the suit be blamed?
Robots that monitor and supervise the elderly
A number of robots have been developed for monitoring the health and safety
of elderly people. Pearl, developed at CMU (Pollack et al, 2002), is a “nursebot” that
reminds seniors about routine activities (e.g. eating, drinking, taking medicine and
using the bathroom/toilet), and that can guide them through their environment. RP-7
is a tele-operated robot that is used to facilitate doctor-patient interactions without the
doctor leaving her office. It has been used at the Silverado Senior Living Apsen Park
(Winner, 2007). The uBot5 robot, developed by University of Massachusetts,
Amherst (Deegan et al, 2007) has manipulators that enable it to perform a number of
tasks such as picking up and moving objects around. It has been proposed as a means
of doing house calls to check for signs of a stroke and could be used to monitor an
elderly person for signs of a fall. It can also be used for social telepresence, since it
can be remotely controlled by authorised users over the internet, allowing a virtual
visit and two way conversation, with the visitor’s face appearing on a video screen.
The robot can also be remote controlled to perform tasks in the home. Mitsubishi
Heavy Industries developed the Wakamura robot mainly as a companion (see next
section), but it can also be used for monitoring the elderly, delivering messages and
reminding about medicine.
Gecko Systems International Corp has predicted that its sales of eldercare
personal robots will reach 8.3 Billion dollars by 2014. They have developed the
15
CareBotTM , a personal robot equipped with multiple vital sign sensors, that can follow
an elderly person in their home: home evaluation trials with the elderly began in
November 2009. The CareBot is capable of verbal interaction, the delivery of
medicine, video monitoring, two way interaction, and can provide verbal reminders at
predetermined dates and times of the need for medical checks, or of other
appointments. Gecko Systems claim that the CareBot provides a more cost-effective
version of a “smart home”, because the robot can follow the occupant from room to
room, and reduce the need for sensors and wires throughout the house. They suggest
that the CareBot could be used to allow a human carer some time off, by using it to
monitor the elderly person, and remind them to take their medicine. RoboSoft, a
French robotics and automation manufacturer, recently introduced RoboLAB10, a
home-assistance robot designed to assist in home care of the elderly. Details as yet
are sketchy, but it apparently comes in two versions —one designed to supply
physical assistance to elderly patients who have difficulty moving themselves or
doing physical tasks around the house. The other is designed to assist those with
cognitive difficulties due to Alzheimer’s disease or other disorders.
In Osaka, Japan, Matsushita Electrics runs “Sincere Kourien”, an elderly
retirement home with 106 beds, and used it as a test bed for robot teddy bears that
monitor patients’ response times to spoken questions (Guardian online, 21st February
2002 http://news.bbc.co.uk/1/hi/sci/tech/1829021.stm). The bears record how long
the residents spend performing various tasks, and can relay conclusions to staff, or
alert them to unexpected changes. They form part of the monitoring network that
exists throughout the building, and for instance alerts staff when residents leave their
beds. Their use in the Sincere Kourien home illustrates a likely direction for
16
monitoring robots, as a robotic component is integrated into a network of fixed
sensors more typical of “smart” homes. Smart homes are also under development in
the UK (e.g. Orpwood et al., 2008) but do not yet have robot components.
There are various reasons for expecting an expansion in this area of the
robotics market. Robotic surveillance devices have already been developed for
warfare, for policing and for home security (Sharkey, 2009) and these could easily be
adapted for monitoring the elderly. A robot that traverses the house, and relays
information picked up by its sensors, is something that is well within the current
technological limits.
We can get an idea of the future potential of monitoring robots with the
Remote Presence robots (RP-6 and RP-7) made by the Californian company, Intouch
Health (see http://www.intouchhealth.com/5Parkview-ICU-IHI-JCAHO.pdf ). The
RP-7 Robot is a 5 ft 5 inch robot that uses two way cameras, microphones and
wireless technology including a joystick for control and a high speed internet
connection to stream information. These have been piloted in a number of hospitals in
Europe, Canada and the United States. Doctors can visit patients by remote
controlling the robot to their bedside – the doctor’s face appears on a monitor
representing the robots head. They have even been used as a remote presence at a
range of 5,400 miles by a surgeon in the US to guide an operation in Argentina.
Although such robots are currently prohibitively expensive (the RP-7 costs
around $80,000) prices will fall and cheaper versions will appear. A monitoring robot
could increase the safety of an elderly person in their own home, and make it possible
for medical staff to virtually visit the elderly person and provide health checks. The
17
elder person’s family could also to check that they were taking their medicine, feeding
themselves and so on. A remote controlled robot could reduce the loneliness of an
elderly person if it were used as an interface for virtual visits by family and friends.
On the downside, we see three major concerns about the application of robots
for monitoring. First, it is likely that the use of robots to monitor elderly people would
result in a reduction in human contact and companionship. We have already
discussed how using assistive robots could mean reducing contact with care workers.
If a monitoring robot were used to make virtual visits to elderly parents, fewer real
visits might take place. Although virtual family visits might alleviate loneliness to a
certain extent, as suggested above, they would not be an adequate substitute for a
living relative sharing your physical space and giving you a hug. If you could use a
remote controlled robot to virtually visit your elderly mother, you might well be less
likely to get in the car and go over to see her. Such virtual visits might alleviate a
family’s, and society’s guilt about the elderly: providing reassuring evidence that they
were still alive, and on their feet.
Our second concern is that there is a risk that monitoring could infringe on the
right to privacy. The privacy of people in general should be respected (see Article 12
of the Universal Declaration of Human Rights, “No one shall be subjected to
arbitrary interference with his privacy, family, home or correspondence, nor to
attacks upon his honour and reputation. Everyone has the right to the protection of
the law against such interference or attacks) and there seems little reason to make an
exception for the old. Privacy can be expressed both as a right, but also as a generally
recognised human value; it has certainly been discussed before in terms of computers,
18
the internet and surveillance in general (e.g. Nissenbaum, 1998; Tavani, and Moor,
2001).
An elderly person might not like to find that an operator could remote control a
robot to peer round their apartment before they are dressed, or when they are taking a
bath. They might prefer the robot to have to do the equivalent of knocking on the
door and waiting to be invited in. The issue becomes more complex if an elderly
person’s mental state deteriorates further and they become confused. A person with
Alzheimer’s would probably forget that the robot was monitoring them, and could
perform acts or say things thinking that they are in the privacy of their own home.
Moreover, who should have access to the information, and how long it should be kept
for? With the massive memory hard drives available today, it would be possible to
record the entire remainder of an elderly person’s life, but this is not something that
they would necessarily consent to if they were able to. Their children, or family could
be asked for consent, but should they be allowed to sign away the senior’s right to
privacy?
Finally, there is a worry that loss of freedom could result from the use of robots
to restrict the activities of the elderly. Such loss of freedom is related to the issue of
autonomy raised in the preceding section. If robots are to be used to help elderly
people to live safely in their own homes for longer, there is likely to be a need for
them to function as autonomous supervisors. A simple extension would be to allow
home customisation so that the robot can recognise danger signs, such as the cooker
being left on, or the bath overflowing. A robot could make a first pass at warning a
senior to stop doing or engaging in a potentially dangerous activity. But there are
ethical problems here. If a robot can predict a dangerous situation, it could be
19
programmed to autonomously take steps to prevent it, e.g. turning the cooker off, or
even restraining the elderly person (gently) from carrying out a potentially dangerous
action such as climbing up on a chair to get something from a cupboard. However,
restraining a person to avoid harm could be a slippery slope towards authoritarian
robotics.
A similar issue could arise with smart homes if sensors were used to detect that
the elderly person is attempting to leave their room, for example, and to lock the door,
or call the staff. Future developments are likely to create similar problems. For
example, Lopes et al, (2009) have developed a method that allows tracking of people
in a range of environments and lighting conditions without the use of sensor beacons.
It might be possible in the future for a robot to follow a senior outside and alert carers
of their location or encourage and guide them back into the home. The ethical danger
here is that this could turn into the equivalent of imprisonment in the home without
trial.
It is not difficult to construct consequentialist, or utilitarian scenarios where
some robot action seems advisable. For example, if an elderly person was about to
walk into the road onto heavy oncoming traffic and a robot could stop them, should it
not do so? The problem here is in trusting the robot’s classification and sensing
systems to determine what is a dangerous activity. As an extreme case, imagine an
elder having high calorie foods taken from her because the robot was preventing her
from becoming harmfully obese. How would the law view a carer using a remote
controlled robot to restrain an elder or take objects of potential harm from her, or
him? Are the safety and health gains great enough to justify the resulting restriction of
the individual’s liberty? There are many discussions to be had over the costs and
20
benefits of robot constraint and where to draw the line. As with many of the ethical
concerns discussed in this paper, similar issues have been ethically debated in other
contexts, (cf Wallach and Allen 2010), but it seems important to identify their
particular manifestation in the context of robotics in eldercare before such
applications have become commonplace.
There clearly needs to be a balance between improving the lives of the elderly
by enabling them to live in their own homes for longer whilst protecting their
individual rights, and their physical and psychological welfare. On the positive side,
some of the concerns expressed here could be alleviated by the employment of a
carefully customised system. A good model is provided, outside of robotics, by
Orpwood et al (2008) who report the development of a smart home for a particular
man with dementia. Detailed records of his behaviour were taken, together with
extensive consultation with his daughter, and care workers. Considerable effort was
made to develop systems that increased his safety, but that did not remove control
from him entirely. For instance, the cooker or taps would be turned off automatically
if left on for some time, but he could override this. A system was developed that
helped with his disorientation and confusion at night – when he left the bed at night,
the light in the toilet was turned on, and a voice prompted him to visit the toilet.
Similarly, if he went near the door of the apartment at night, the recorded voice of his
daughter was played, encouraging him to go back to bed. If he ignored this, and left
the apartment, then care staff in the building were alerted. The smart home
development was shown to increase his urinary continence, and to improve his
sleeping pattern.
21
Although the smart home solution did not involve robots, it provides a good
example of the detailed consultation and customisation needed to produce a good
working solution for an individual. It also provides an illustration of the achievement
of an apparently successful balance between protecting this elderly person’s physical
health, whilst still preserving his freedom and control over his life. Similar sensitive
customisation of monitoring robots, using a value sensitive design approach, may well
turn out to be the best way of gathering the advantages afforded by robots, without
reducing the quality of life of the elderly (see Forlizzi et al, 2004 for a related
approach).
Robots as companions for the elderly
It is sometimes suggested that robots could provide companionship for the
elderly. Indeed, an elderly person might be more likely to tolerate, or even enjoy, the
presence of a monitoring robot if it also had some ability to be a companion. For
instance, the Gecko CareBot referred to in the previous section is described as “a new
kind of companion that always stays close to them enabling friends and family to care
from afar”. There is also a growing number of robot toys, some of which have been
have been mooted as possible companions for the elderly. Such toys have the
advantage of being relatively more affordable than some of the more elaborate robots
we have considered so far. We describe four such examples, before turning a cautious
eye on research that claims to show the beneficial effects of robot companions, and
moving to a consideration of the ethical implications of robot companions in general.
Paro, a fur covered robotic seal, was specifically designed for therapeutic uses
with the elderly. Developed by AIST, it responds to petting by moving its tail and
22
opening and closing its eyes. It reacts to sounds and can learn to respond to its name.
It makes seal-like sounds, and is active in the day, preferring to “sleep” at night. It
can detect light and dark by means of a light sensor, and recognise when it is being
held, stroked, or hit, by means of posture and tactile sensors.
Sony’s AIBO robotic dog, developed as an “entertainment robot”, has also been
used in robot companions research. It has a metallic dog-like form, and can walk, or
chase a ball. It has sensors that can detect distance, acceleration, sound, vibration and
pressure. It can express six “emotions” (happiness, anger, fear, sadness, surprise and
dislike), by means of its tail, body movements, and the colour and shape of its eyes.
More recent versions can recognise voice commands, and the robot exhibits slightly
different behaviour depending on the interactions it has experienced.
NeCoRo (OMRON) is a robotic cat covered in synthetic fur. Its behaviour
depends on the history of its interactions and it can ‘learn’ to recognise its name.
Stroking and hugging it elicits positive responses while treating it roughly elicits
angry behaviour. Like Paro, it has internal rhythms that lead it to sleep, or seek to be
cuddled. And finally, My Real Baby (iRobot) is described as an “interactive
emotionally responsive doll”. It can make different expressions, blink, and suck its
thumb or bottle. It changes its expressed emotion depending on how it is treated – so
that when fed and rocked for instance, it behaves as though it were sleepy. When
unfed, it gets hungry and cries. It also makes a number of random sounds and words,
which become more like sentences the longer it is played with.
There are other robots that have also been touted as possible companions for the
elderly (see Wilks, 2010 for interesting discussions of digital companions in general).
Toy robots that could entertain the elderly include: Pleo, ifbot, and Primo Puel. Pleo
23
is a robotic dinosaur with many sensors, which responds with different behaviours
depending on its treatment. The ifbot was developed by Business Design Laboratory
Co. for elderly people and can converse with them by means of a large number of
stored interaction patterns. Primo Puel is an interactive doll that talks, giggles, and
asks for cuddles. It was originally designed as a stand in for a boyfriend for young
single women, but proved unexpectedly popular with elderly women in Japan.
Various studies have looked at the effect on the elderly of interacting with such
robot companions, and claim to have found evidence that the elderly can benefit from
them. The positive effects that are found are often said to be similar to those obtained
from animal assisted therapy in which visits from a pet animal to residents of long
term care facilities are found to reduce loneliness (Banks and Banks 2002; 2005). For
instance, Kanamori et al (2002) showed various improvements in elderly persons who
interacted regularly with a Sony AIBO robotic dog – their loneliness scores were
reduced, and their Quality of Life assessment scores improved. Banks et al (2008)
even found that elderly people in long term care facilities benefited as much from
interacting with an AIBO robotic dog as from interacting with a real dog; both
resulting in reductions in their reported levels of loneliness. Elderly dementia patients
have also shown positive outcomes, including increased communication, as a result of
sessions with an AIBO (Tamura et al, 2004). An interesting possibility that could be
explored in future research is that some of the benefits that result from playing with,
or “nurturing” robot pets are a consequence of increased levels of oxytocin, which as
discussed earlier, appears to reduce stress.
These outcomes sound promising, but there is a need for caution in their
interpretation because they depend on what the interactions are being compared to.
24
The results reported by Kanamori et al (2002) showed improvements in well being
over time between initial and later sessions. Banks, Willoughby and Banks (2008)
showed that beneficial effects were obtained for those interacting with either the real
or the robotic dog, as compared to their control group who received no such
opportunities for interaction. However, such improvements could have been found
because the alternative was so dire and unstimulating. Someone in solitary
confinement might benefit from being given a robot companion, but they would
benefit far more if they were offered a friendly social environment. It is not clear that
the same relative improvements would be found if the comparison were to a control
group that received some other form of intervention. If the comparison group
received some other individual attention, such as a visit by someone who massaged
their neck and shoulders, or who sat holding their hand and chatting, the benefits of
the robot dog might not be so apparent. It is also important to see that any benefits
are maintained over time, because a robot that was interesting to interact with initially
may rapidly lose its appeal. A Reuters article (21st September 2007) reported that
director of a Japanese care home had said that residents “liked the ifbot for about a
month before they lost interest”. The robot had spent most of the past two years
languishing alone in a corner, and the director of the facility commented, “stuffed
animals are more popular”.
As well as raising doubts about the reported benefits of interacting with robot
pets, we can identify several ethical issues involved in the use of robot companions.
First among these, as we have discussed in the other areas of robot care, is that their
introduction could lead to a reduction in the amount of human contact that the elderly
person experiences. At present, robots are far from being real companions. They can
interact with people, and even show simulated emotions, but their conversational
25
ability is still extremely limited. They cannot form adequate replacements for human
love and attention. Unfortunately this does not mean that they will not be used as
such: the Gecko CareBot is already being advertised as “a new kind of companion”.
There is a risk that despite its limitations, the provision of a robot companion could be
used as a justification for leaving the senior citizen on their own for longer. It could
be used to alleviate the family’s guilt about doing so – “Don’t worry about Granny,
she’s got the robot to talk to”.
Then there are the issues of deception and infantilisation of the elderly. Sparrow
(2002) and Sparrow and Sparrow (2006) argue that any beneficial effects of robot pets
or companions are a consequence of deceiving the elderly person into thinking that
the robot pet is something with which they could have a relationship, and are adamant
that this should not be encouraged. Turkle et al (2006) expressed similar disquiet:
“the fact that our parents, grandparents and our children might say ‘I love you’ to a
robot who will say ‘I love you’ in return, does not feel completely comfortable; it
raises questions about the kind of authenticity we require of our technology” (Turkle
et al, 2006, pg 360). Wallach and Allen (2009), in a discussion of the ability of robots
to detect basic human social gestures, and respond with human-like social cues,
suggest that, “From a puritanical perspective, all such techniques are arguably forms
of deception” (Wallach and Allen, pg 44). Sparrow argued that the relationships of
seniors with robot pets, “are predicated on mistaking, at a conscious or unconscious
level, the robot for a real animal. For an individual to benefit significantly from
ownership of a robot pet they must systematically delude themselves regarding the
real nature of their relation with the animal. It requires sentimentality of a morally
deplorable sort. Indulging in such sentimentality violates a (weak) duty that we have
26
to ourselves to apprehend the world accurately. The design and manufacture of these
robots is unethical in so far as it presupposes or encourages this”, (Sparrow, 2002).
Related to deception is the concern that encouraging elderly people to interact
with robot toys has the effect of infantilising them. Similar points have been made in
the context of doll therapy for dementia patients. Positive results have been obtained
where dolls are introduced to elderly clients to act as a focus for reminiscence and
conversation (Cayton, 2006). For example, the “Someone to Care For” doll is made
especially for the elderly, and the manufacturers claim that “These beautiful dolls
offer comfort, care and happiness to senior citizens, especially people living with
Alzheimer’s disease”. However, critics such as Cayton (2006) suggest that doll
therapy is predicated on the idea that those with dementia are going through a second
childhood, and that this notion is dispiriting and encourages a rigid authoritarian,
deficit-based approach to care. Kitwood (1997) describes a malignant way of caring
for those with dementia that leads to disempowerment, disparagement and
infantilisation. The same could be said of the use of robot toys – which in some cases
such as My Real Baby, are dolls with the addition of sensors and movement
capabilities.
Despite these worries, we suggest that it is still possible to envisage ways in
which robot companions, or robot pets in particular, might improve the lives of the
elderly. Their introduction does not necessarily need to lead to a reduction in human
contact. It is possible that robot pets could act as social facilitators, and lead to
increased interactions between their elderly owners and other people. Robot toys can
give the elderly person something to talk about, and to show, and other people
something to talk to them about. For instance, when Wada and Shibata (2006)
27
videoed interactions between a Paro robot seal and a group of elderly care home
residents, they found evidence that the social interactions between the residents
themselves increased, at the same time that physiological indicators showed reduced
stress levels. It seems that Paro even encouraged positive communication, and
resulted in a reduction of the “backbiting” that had previously characterised some of
their interactions. One 75 year old female resident greatly increased her interaction
with fellow residents. A friend of hers commented that she had been taciturn before
Paro’s arrival, but that now she was more cheerful and talkative.
A robot that facilitates conversation may function as an attractor for visitors. It
gives a focus of attention for chat. Children may want to play with the robot and have
fun with granddad’s big toy. Kanamori et al (2002) report the case of an 84-year-old
man with cerebral apoplexy sequelae. He talked much more to his children after the
introduction of an AIBO robot dog. It gave both him and them a focussed object to
talk about. Thus benefits could be obtained from robot pets that are due to the robots
stimulating increased interaction with other people; benefits that have nothing to do
with deceiving the elderly person.
In addition, on further consideration the argument can be made that the issue of
robot companions and deception is not as straightforward as previous claims suggest
(Sparrow 2002; Sparrow and Sparrow 2006; Wallach and Allen, 2009). As discussed
by Sharkey and Sharkey (2006), humans in general are all too ready to
anthropomorphise machines, and other objects, and to imagine that they are capable
of more than is actually the case. Zizek (2002) the psychoanalyst describes the way in
which people can chose to act as though something were not real, “I know very well
that this is just an inanimate object, but none the less I act as if I believe that this is a
28
living being”. There may well be elements of a “willing suspension of disbelief”
(Coleridge 1817) when the elderly interact with robot pets.
It is quite possible that elderly people might enjoy, and benefit, from interacting
with a robot pet without thinking that it is actually sentient. It is likely that their views
about such artefacts are unclear – and that they will be seen neither as being sentient,
nor as objects, but as falling “betwixt and between” known categories as discussed by
Turkle et al (2006). Related questions have been investigated by Peter Kahn and his
associates, albeit in the context of children and adults rather than the elderly. Kahn et
al (2006) looked at preschool children’s reasoning about a Sony AIBO pet in
comparison to a stuffed toy dog. Their results suggest that there was some confusion
and inconsistency in the children’s beliefs about the robot dog. The children’s
answers to questions about the robot dog and the stuffed dog were similar, but their
behaviour was found to differ. They were more likely to mistreat the stuffed dog, and
to animate it, than the robot dog, and they made more attempts to engage in reciprocal
interaction with the AIBO. In a further study, (Melson et al, in press) children’s
interactions with and views of a living and a robot dog were compared. Again,
although their views of the two differed (they saw the live dog as more likely than the
robot to have physical essences, mental states, sociality and moral standing), most of
the children still interacted with the AIBO as if it were a real dog. Kahn et al. (2006)
suggest that a new technological genre of autonomous, adaptive, personified and
embodied artefacts is emerging that the English language is not well-equipped to
handle. They believe that there may be need for a new ontological category beyond
the traditional distinction between animate and inanimate.
29
More psychological research is needed to provide a clearer picture of the beliefs
that the elderly hold about robots they encounter, and of the effect of those beliefs.
Elderly people might derive pleasure from acting as if the robot could understand
them, whilst remaining aware of its mechanical nature. It is also possible that the
elderly might still obtain benefits from interacting with a robot, even if efforts are
made to ensure that they are made aware of its machine-like nature. For example,
Turkle et al (2006) found that children who interacted with the MIT robots Cog and
Kismet continued to see them “not only as “sort of alive” but as capable of being
friends and companions” even when the researchers spent some time showing them
how they worked and emphasising their underlying machinery. There are also likely
to be considerable individual differences here, depending on factors such as for
example, elderly people’s knowledge and experience of technology, and their beliefs
about and attitudes to mechanical devices. Turkle et al (2006) found that seniors
interacting with robots showed considerable variations in their attitudes and behaviour
towards them: for instance, some wanted to know how they worked in a mechanical
sense, whereas others were content to interact with them ‘as they presented
themselves’, with no interest in their underlying mechanism. It is quite likely that
elderly people might vary in their level of interest in interacting with robot pets, and
in the benefits that they obtained from such interactions.
Essentially, our suggestion here is that to claim that robot companions are
unethical because their effectiveness depends on deception (Sparrow, 2002; Sparrow
and Sparrow, 2006), oversimplifies the issue. The whole question of deception, and
the possibility of the willing collusion of the elderly person themselves, is a complex
one, and it is not clear that the benefits of robot companions depend on deceiving
elderly people about their true nature. It is also not necessarily the case that
30
interacting with robot pets means infantilising the elderly. Arguably, interacting with
robots is more socially acceptable than playing with dolls, and robot toys are more
likely to be viewed as “cool gadgets”. This view is supported by Turkle et al (2006
pg 354) who reported that elderly people were prepared to interact with robots:
“Seniors felt social ‘permission’ with the robots presented as a highly valued and
‘grown up’ activity. ” (Turkle et al, 2006 pg 354). It seems that, if used with care,
robot pets or companions could add extra interest to an elderly person’s life, and
might even improve their social life.
Conclusions
Our answer to the question posed at the outset of this paper, “should we be
concerned about the use of robots for elder care?” is ‘yes’. In each of the preceding
three sections, we have identified ethical concerns. However we have also found
reasons to think that robots could make a positive contribution. It is not the use of
robots in elder care per se that should concern us, but the ways in which they are
used.
Several possible positive contributions that robots and robotic technology
could make to the physical and the mental welfare of the elderly were highlighted.
Assistive robots and robotic technology could help to overcome problems of mobility,
and reduce elderly people’s dependence on busy, and sometimes inattentive, care
staff. The use of remote controlled robots to monitor, and virtually visit elderly
people could enable the elderly to live independently for longer. Robots could remind
them what medicines to take, watch out for health problems and safety risks.
Companion robots could facilitate the social lives of elderly people, by giving them an
31
interesting gadget to talk to other people about. Social interaction could also be
facilitated by monitoring robots that enabled virtual visits from friends and family.
At the same time, in the course of this paper we have identified a number of
ethical issues of which developers and users need to be aware. If measures are not
taken to guard against these concerns, there is a danger that robot use in elder care
could lead to a dystopian environment. The quality of life of the elderly should always
be put ahead of convenience to the system of care.
We have identified six major issues that need to be considered before
deploying robot technology fully in eldercare: (i) Opportunities for human social
contact could be reduced, and elderly people could be more neglected by society and
their families than before. Robots could provide an excuse for such neglect, if others
mistakenly chose to believe that the seniors’ physical and emotional needs were being
taken care of by machines; (ii) The insensitive use of robots developed for the
convenience of carers, and the consequent increase in elderly people’s feelings of
objectification, and loss of control. This could occur if, for instance, robots were used
to lift or move people around, without consulting them; (iii) Loss of privacy, and (iv)
The restriction of personal liberty that could result from the use of robots in elder
care. The extent to which robots should be allowed to restrict the behaviours of
humans is a difficult question, with implications beyond the care of the elderly; (v)
The deception and infantilisation of elderly people that might result from encouraging
them to interact with robots as if they were companions, (although as discussed, the
possible willing collusion of the elderly with such deception makes this issue a
complex one); and finally (vi) If robots were placed under the control of elderly
people, there is the issue of responsibility if things were to go wrong. But this opens
32
up other important issues such as the extent to which the wishes of the elderly person
should be followed, and the relationship between the amount of control given to the
elderly person, and their state of mind.
Clearly issues such as privacy, safety, and autonomy are not unique to robotics
for elder care and have been raised and discussed elsewhere - and there is an
increasing interest in the notion of machine ethics in general (Allen et al, 2006). The
contribution of the present paper lies in identifying and bringing together the various
ethical issues for specific consideration in the context of robots in elder care. This is a
pressing issue, for the indications are that robots will be regularly used in this domain.
There has been some previous work in this area which has raised a small subset of
similar ethical concerns, (Sparrow, 2002; Sparrow and Sparrow 2006), and some
which has focused on the development of eldercare robots that have some ability to
ethically reason about what actions they should take (Anderson et al 2006; Anderson
and Anderson, 2008).
Robots and robotic technology require a new perspective on many ethical
issues, because of their embodiment, and their often life-like appearance. Their
physical embodiment means that they can be used to perform tasks in the world to a
greater extent that purely computational devices. For instance, robots, unlike
computers, can follow a person around; as technology develops they are likely to be
able to follow someone wherever they go. Similarly, as developments such as the
Riba robot demonstrate, robots can be used to physically interact with people – for
example, lifting them and moving them around. Their often personable appearance
may lead them to be welcomed in the home and other locations, where for instance a
surveillance camera would not be accepted. Finally, that personable, or animal-like
33
appearance, can encourage and mislead people into thinking that robots are capable of
more social understanding than is actually the case. Their appearance and behaviour
can lead people to think that they could form adequate replacements for human
companionship and interaction. Identifying the various consequences of such
embodiment is an important contribution of this paper.
Identifying the ethical concerns associated with robots in elder care is a
necessary first step in ensuring that their introduction will not result in a reduction in
the welfare of the seniors exposed to them. Having identified these concerns, what
should the next step be? A blanket ban would mean missing out on the many potential
advantages that could be gained from the use of robots in eldercare. A particularly
useful step would be to develop guidelines, and eventually legal restrictions, on their
use.
Since the effect of robots on the lives of the elderly depends on the ways in
which they are deployed, the development of guidelines about their use in care homes,
and in their own homes, could help to guard against their misuse. At present, apart
from fundamental human rights legislation, there is little protection for elderly people
against the potential downsides of robot care. In particular, there are no obvious
restrictions on the amount of time that elderly people could be left in the care of
robots, nor on the amount of human contact that they should experience. Like
children, the very old and infirm can be seen as being in need of special protection.
As discussed elsewhere (Sharkey and Sharkey 2010), children could possibly be
afforded some protection from extensive robot care by the legislation pertaining to
neglect, but there is no such legislation for the elderly.
34
The lack of protection for the human rights of the elderly is often
acknowledged. For instance, a recent report on human rights for the elderly (Older
people and human rights, March 2009) points out that there is no binding international
human rights treaty that specifically protects the rights of older people. Legal
protection for elderly people relies on human rights legislation, but this report argues
that ‘There is a lack of enforceable economic, social and cultural rights in the UK,
with the Human Rights Act focusing largely on civil and political rights. This
omission has major implications for older people, who face a range of economic,
social and cultural issues.’ There are some international treaties that relate to the
elderly. In 1991, the UN produced the “UN principles for older persons”, but
although this represents progress it is a declaration rather than binding international
law. While such progress on the human rights of the elderly is encouraging, as yet
there has been little discussion about the use of robots in eldercare.
Guidelines, and ideally legislation, about the uses of robots in the care of the
elderly could limit the use to which robots were put. For example, guidelines could
be drawn up about the amount of time that a person could be left alone with a robot
without human company. There could be a requirement that any robot physically
interacting with, or moving a person must first ask their permission. Further
development of guidelines to guard against the likely problems of robot care would
need to be based on careful evaluation of the effect of different forms of robot care on
older people.
Ethical concerns about robots in eldercare could also be mitigated by means of
detailed consultation with elderly people, with the aim of ensuring that robots
introduced into elder care do actually result in improvements in the lives of the
35
elderly. It is likely that there will be considerable individual differences in the
attitudes of older people towards robots, (although these might be susceptible to
change if they experienced some beneficial effect). Seniors are likely to differ in the
extent to which they might be prepared to interact with a robot, and in their
preferences. For instance, some older people might prefer the increased autonomy
afforded by robotic technology that allowed them to feed themselves, whilst others
might rather retain the opportunities to interact with human beings over such basic
caring tasks. Sensitive customisation is likely to be needed in order to ensure a
positive effect on the quality of life of the elderly.
Another approach would be to encourage robot developers to take a value
sensitive design approach – building in protection against the ethical concerns raised
here in the robots themselves. For example, a monitoring robot could have to always
request permission before it started any form of monitoring of an elderly person, and
it could be required to explicitly indicate when monitoring was taking place. Perhaps
a companion robot could also keep track of the amount of interaction its elderly
companion has had with other humans; it could take steps to increase the level of
interaction when the level was very low. Maybe it could alert human carers that its
charge was being neglected, or it could display a set of further behaviours designed to
attract more interaction from other humans in the vicinity. But these are only
preliminary suggestions; their further development is beyond the remit of the present
paper.
In this paper, we have highlighted and discussed six ethical concerns
associated with the incorporation of robots into eldercare. These concerns are related
both to human rights, and to shared human values. At the same time, we have also
36
identified benefits that could result from the careful incorporation of robots in elder
care. It is argued that identifying both the risks, and the possible benefits, of elder
care is a necessary precursor to the development of ethical solutions. In addition, we
briefly considered ways of ensuring that the elderly experience the benefits, and not
the downsides of robot care. The methods we suggest are (a) the development of
guidelines, and ideally legislation, about their use, (b) basing these guidelines on
carefully controlled studies and consultation with the elderly, and encouraging the
development of customised solutions, and (c) value sensitive design of the robots in
the first place.
References
Allen, C., Wallach, W. and Smit, I. (2006) Why Machine Ethics? IEEE Intelligent
Systems, 21(4), 12-17.
Anderson, M., Anderson, S.L. and Armen, C., (2006) An Approach to Computing
Ethics, IEEE Intelligent Systems, 56-63.
Banks M.R., Banks W.A., (2002) The effects of animal-assisted therapy on loneliness
in an elderly population in long-term care facilities. J Gerontol A Biol Sci Med Sci
57A:M428–M432.
Banks M.R., Banks W.A.. (2005) The effects of group and individual animal assisted
therapy on loneliness in residents of long-term care facilities. Anthrozoos; 18:396–
408.
37
Banks, M.R., Willoughby, L.M., and Banks, W.,A., (2008) Animal-assisted therapy
and loneliness in nursing homes: use of robotic versus living dogs, J Am Med Dir
Assoc 9, pp. 173–177.
Boas, I. (1998) Learning to be rather than do, Journal of Dementia Care, 6, 6, 13.
Clark, A. and Chalmers, D. J. (1998) The extended mind. Analysis 58:10-23.
Cayton, H. (2006) From childhood to childhood? Autonomy and dependence through
the ages of life. In Julian C. Hughes, Stephen J. Louw, Steven R. Sabat (Eds)
Dementia: mind, meaning, and the person, Oxford, UK: Oxford University Press
277-286
Coleridge, S.T. (1817) Biographia Literaria, chapter 14, page 314
Deegan, P. Grupen, R., Hanson, A., Horrell, E., Ou, S., Riseman, E., Sen, S.,
Thibodeau, B., Williams, A., Xie, D. (2007) Mobile Manipulators for Assisted Living
in Residential Settings, Autonomous Robots, Special Issue on Socially Assistive
Robotics. Volume 24, Number 2.
Doshi, F., and Roy, N. (2008) Spoken language interaction with model uncertainity:
an adaptive human-robot interaction system. Connection Science, 20, 4, 299-318.
38
Forlizzi, J., DiSalvo, C., and Gemperle, F. (2004) Assistive Robotics and an Ecology
of Elders Living Independently in Their Homes. Human-Computer Interaction, 19,
25-59.
Fratiglioni L, Wang H-X, Ericsson K, et al. (2000) Influence of social network on
occurrence of dementia: a community-based longitudinal study. Lancet 355:1315–19.
Heinrichs, M., von Dawan,s B., and Domes, G. (2009) Oxytocin, vasopressin, and
human social behavior, Front. Neuroendocrinol, doi:10.1016/j.yfrne.2009.05.005
Heinrichs, M., Baumgartner,T., Kirschbaum, C.&Ehlert,U. (2003) Social support and
oxytocin interact to suppress cortisol and subjective responses to psychosocial stress.
Biol. Psychiatry 54, 1389–1398.
Holtzman RE, Rebok GW, Saczynski JS, et al. (2004) Social network characteristics
and cognition in middle-aged and older adults. J Gerontol B Psychol Sci Soc
Sci;59:P278–84
House of Lords, House of Commons Joint Committee on Human Rights, (2006-7)
The human rights of elder people in health care. London: The Stationary Office Ltd.
Kahn, P. H., Jr., Friedman, B., Perez-Granados, D., & Freier, N. G. (2006). Robotic
pets in the lives of preschool children. Interaction Studies, 7(3), 405-436.
39
Kanamori, M., Suzuki, M., & Tanaka, M., (2002) Maintenance and improvement of
quality of life among elderly patients using a pet-type robot. Japanese Journal of
Geriatrics, v. 39, p. 214-8.
Kikusui, T., Winslow, J.T., and Mori, Y., (2006) Social buffering: relief from stress
and anxiety. Philos Trans R Soc Lond B Biol Sci December 29, 361 (1476): 2215-
2228.
Kitwood, T. (1997) Dementia Reconsidered: The Person Comes First. Open
University Press.
Lopes, M. M., Koenig, N. P., Chernova, S. H., Jones, C. V., and Jenkins, O. C. 2009.
Mobile human-robot teaming with environmental tolerance. In Proceedings of the 4th
ACM/IEEE international Conference on Human Robot interaction (La Jolla,
California, USA, March 09 - 13, 2009). HRI '09. ACM, New York, NY, 157-164.
Melson, G. F., Kahn, P. H., Jr., Beck, A. M., Friedman, B., Roberts, T., Garrett, E., &
Gill, B. T. (in press b). Robots as dogs? - Children's interactions with the robotic dog
AIBO and a live Australian shepherd. Journal of Applied Developmental Psychology
New Scientist online (2008)
http://www.newscientist.com/blogs/shortsharpscience/2008/11/honda-helps-you-
walk.html)
40
Langer, E. J., and Rodin, J. (1976) The effects of choice and enhanced personal
responsibility for the aged: a field experiment in an institutional setting. Journal of
Personality and Social Psychology, 34,2, 191-8.
Nissenbaum, H. (1998a), “Protecting Privacy in an Information Age: The Problem of
Privacy in Public,” Law and Philosophy, 17: 559-596.
Nguyen, H., Anderson, C., Trevor, A., Jain, A., Xu, Z., and Kemp, C.C. (2008) El-E:
An Assistive Robot that Fetches objects from Flat Surfaces, HRI Workshop on
Robotic Helpers: User Interaction Interfaces and Companions in Assistive and
Therapy Robots.
Older people and human rights: Research and Mapping Report. 2009-11-03 Age
Concern England and British Institute of Human Rights.
Orpwood, R.., Adlam, T., Evans, N., Chadd, J., (2008) Evaluation of an assisted-
living smart home for someone with dementia. Journal of Assistive Technologies, 2,
2, 13-21.
Pollack, M.E., Engberg, S., Matthews, J.T., Thrun, S., Brown, L., Colbry, D., Orosz,
C., Peintner, B., Ramakrishnan, S., Dunbar-Jacob, J., McCarthy, C., Montemerlo, M.,
Pineau, J., and Roy, N. (2002) Pearl: A Mobile Robotic Assistant for the Elderly,
AAAI Workshop on Automation as Eldercare, Aug 2002.
Saczynski1, J.S., Pfeifer, L.A. , Masaki. K., Korf. E.S.C., Laurin, D., White, L. , and
Launer, L.J. (2006) The effect of social engagement on incident dementia: The
Honolulu-Asia Aging Study. Am J Epidemiol. Mar 1;163(5):433-40.
41
Schacter, O. (1983) Human Dignity as a Normative Concept. The American Journal
of International Law, 77, 4, 848-854.
Sharkey, N.E., (2008) The Ethical Frontiers of Robotics, Science, 322. 1800-1801
Sharkey, N.E. (2009) The Robot Arm of the Law grows longer, IEEE Computer,
August, 2009
Sharkey, N., and Sharkey, A., (2006) Artificial Intelligence and Natural Magic,
Artificial Intelligence Review, 25, 9-19
Sharkey, N. and Sharkey, A. (2010) Living with robots: ethical tradeoffs in eldercare,
In Wilks, Y.. Close Engagements with Artificial Companions: Key psychological,
social, ethical and design issues pp 245-256 Amsterdam: John Benjamins
Sharkey, N. and Sharkey, A. (2010) The crying shame of robot nannies: an ethical
appraisal. Interaction Studies, 11, 2, 161-190.
Sparrow, R., (2002). The march of the robot dogs. Ethics and Information
Technology, 4, 305–318.
Sparrow, R., and Sparrow, L. (2006) In the hands of machines? The future of aged
care. Mind and Machine, 16: 141-161.
42
Smith, J., 2003. "Stress and aging: theoretical and empirical challenges for
interdisciplinary research." Neurobiology of Aging, Suppl 1:S77-80; discussion S81-2.
Takayama, L. (2010). On making robots invisible-in-use. Proceedings of the
International Symposium on New Frontiers in Human-Robot Interaction: AISB
2010, Leicester, UK.
Tamura, T., Yonemitsu, S. Itoh, A., Oikawa, D., Kawakami, A., Higashi, Y.,
Fujimooto, T., Nakajima, L. (2004) Is an entertainment robot useful in the care of
elderly people with severe dementia? J Gerontol A Biol Sci Med Sci. Jan;59(1):83-5.
Tavani, H. and J. Moor (2001), “Privacy Protection, Control of Information, and
Privacy-Enhancing Technologies,” Computers and Society, 31(1): 6-11.
Turkle, S., Taggart, W., Kidd, C.D., Dasté, O.,(2006) Relational Artifacts with
Children and Elders: The Complexities of Cybercompanionship. Connection Science,
18, 4, pp 347-362.
Turkle, S., Breazeal, C., Dasté, O., and Scassellati, B., (2006b) “First Encounters with
Kismet and Cog: Children Respond to Relational Artifacts”. In Digital Media:
Transformations in Human Communication, Paul Messaris and Lee Humphreys
(eds.). New York: Peter Lang Publishing
Wada, K. and Shibata, T. (2006) Robot Therapy in a Care House: Its
Sociopsychological and Physiological effects on the Residents. Proceedings of the
43
2006 International Conference on Robotics and Automation, Orlando, Florida, May
2006. 3966-3971.
Wallach, W., and Allen, C.,(2009) Moral Machines: Teaching Robots Right from
Wrong. Oxford University Press, New York.
Wang, H., Karp, A., Winblad, B., and Fratiglioni, L. (2002) Late-Life Engagements in
Social and Leisure Activities is Associated with a Decreased Risk of Dementia: A
Longitudinal Study from the Kungsholmen Project. American Journal of
Epidemiology, 155, 12, 108101087.
Wilks, Y. (2010) Editor, Close Engagements with Artificial Companions: Key
psychological, social, ethical and design issues Amsterdam, John Benjamins.
Wilson, R.S., Krueger, K.R., Arnold, S.,E., Schneider, J.A., Kelly, J.F., Barnes, L.L.,
Tang, Y., Bennett, D.A., (2007) Loneliness and risk of Alzheimer’s Disease. Arch
Gen Psychiatry. 64:234-240.
Winner, S., (2007) Beam me inside, Scotty! Assisted Living Consult
Zizek, S., (2002) The Zizek Reader. London: Blackwell.
