Design for compassion: designing robots for e-mental healthcare

Abstract

As the demand for mental health treatments rises, technology is called to bridge the gap between a traditional mental healthcare approach and the new needs of our evolving society. In the field of mental healthcare, the goal is to create a more sustainable system in which the adaptability of people is increased and supported by technology. However, when it comes to introducing technology, seen as cold and distant in a field so connected to humanness, there is a fear of losing the therapeutic alliance between therapist and patient. For this reason, we address the need for technology to be designed and developed in a way that increases the importance of benefiting the people in the mental healthcare domain. We believe that this is achieved through a design for values approach, where the value of compassion, a very important one in the field of mental healthcare, is given a central role in the development of new therapeutic technologies. In this paper, from considerations on different kinds of applicable technologies, we will focus on and shortly explore what it means to design a robot, based on the value of compassion. We will address the differences between a human-human interaction and a robot-human interaction in the compassion process, and how this affects the design of a compassionate robot.

Full text

First international workshop on Designerly HRI Knowledge. Held in conjunction with the 29th IEEE International
Conference on Robot and Human Interactive Communication (RO-MAN 2020) – http://hridesign.eu/
Design for compassion:
designing robots for e-mental
healthcare
Benedetta Lusi1, Charlotte M. van Lotringen2, R. Klaassen3, Geke D. S. Ludden4, Matthijs
L. Noordzij5
1 Interaction Design, University of Twente, Enschede, Netherlands, b.lusi@utwente.nl
2 Psychology Health and Technology, University of Twente, Enschede, Netherlands,
c.m.vanlotringen@utwente.nl
3 Human Media Interaction, University of Twente, Enschede, Netherlands,
r.klaassen@utwente.nl
4 Interaction Design, University of Twente, Enschede, Netherlands, g.d.s.ludden@utwente.nl
5 Psychology Health and Technology, University of Twente, Enschede, Netherlands,
m.l.noordzij@utwente.nl
Abstract — As the demand for mental health
treatments rises, technology is called to bridge the
gap between a traditional mental healthcare
approach and the new needs of our evolving
society. In the field of mental healthcare, the goal
is to create a more sustainable system in which the
adaptability of people is increased and supported
by technology. However, when it comes to
introducing technology, seen as cold and distant in
a field so connected to humanness, there is a fear
of losing the therapeutic alliance between therapist
and patient. For this reason, we address the need
for technology to be designed and developed in a
way that increases the importance of benefiting the
people in the mental healthcare domain. We
believe that this is achieved through a design for
values approach, where the value of compassion, a
very important one in the field of mental
healthcare, is given a central role in the
development of new therapeutic technologies. In
this paper, from considerations on different kinds
of applicable technologies, we will focus on and
shortly explore what it means to design a robot,
based on the value of compassion. We will address
the differences between a human-human
interaction and a robot-human interaction in the
compassion process, and how this affects the
design of a compassionate robot.
Keywords —compassion, design for values,
robots, methodology, e-mental healthcare.
1. INTRODUCTION AND MOTIVATION
In the Netherlands, the costs associated with
mental healthcare are rising and waiting lists are
getting longer, with more than one million patients
per year seeking support. In order to keep offering
high quality mental health care in the future, we
need to realize a more sustainable, economically
viable healthcare, focused on prevention and self-
management. For this purpose, ICT has been put
forward as a solution. New paths of treatment by
technology mediated mechanisms were opened up,
and evidence of effectiveness of e-mental
healthcare treatments is rising. However, the
transition to a more sustainable mental healthcare
has not taken place to the extent envisioned. There
are barriers that prevent this transition from
happening: clients show low adherence to the
technology mediated options, and professionals
show hesitation in prescribing e-mental health
options and a lack of digital skills to confidently
explain and adapt them [1]. Furthermore, the
technology mediated treatments seem to be
perceived by mental health professionals as cold
and inferior options that can be ignored in favor of
the warmer, more humane and more effective Face-
to-Face (F2F) options. In fact, current mental health
technologies are often a direct translation of F2F
existing therapies in and/or through a technological
medium. This results in low societal readiness
levels of the current mental health technologies.
Societal Readiness Levels (SRL) are a way of

Figure 1. Compassion dynamics analysis: Human-
Human Interaction
assessing the level of societal adaptation of, for
instance, a product or technology to be integrated
into society (as defined by Innovation Fund
Denmark). If the societal readiness for a technology
is expected to be low, it is necessary to offer
suggestions for a realistic transition towards
societal adaptation. While, currently, we are not
using technology in the way it can best benefit
people in this domain, innovative technologies,
such as wearables, interactive devices and robotic
instances could more closely support people at any
time and place. To accelerate a more widespread
transition to sustainable mental health care, we
propose applying a further development of design
for values [2]. More specifically, we will study how
to design for a crucial value in the mental
healthcare context: compassion. Our ambition is to
support the more traditional approach with
technology-mediated treatments and provide new
knowledge and tools to designers and developers
that will enable them to design technology for
Mental Healthcare with a higher Societal Readiness
Level.
2. TO DESIGN FOR COMPASSION
Compassion differs from related constructs,
such as empathy, because it not only involves
recognizing and imagining the suffering of others,
but also includes an acting component: a
motivation to act to relieve this suffering. From a
psychological point of view this value has proven
to be beneficial in mental healthcare settings
because it supports the therapeutic alliance [3] and
it is central to the process of recovery from
psychopathology [4]. As for interaction design and
human-robot interaction, compassion is one of the
three core components of Amit Ray’s
Compassionate AI [5] for generating awareness
and deep feelings in robots. Even so, in order to
better understand the significance of compassion
we need to first establish what it is.
2.1. Compassion as a process
Compassion can be broadly described as a
response to suffering [6]. However, an even more
relevant definition, with regards to the design of
technology, is found in a recent review that
describes compassion as a five-step process [7]:
1. Recognize the suffering
2. Understand the universality of human
suffering
3. Feeling empathy for the person(s)
suffering and connecting with the distress
4. Tolerate uncomfortable feelings
5. Be motivated to act/alleviate the suffering
What already can be inferred from this
conceptualization is that compassion is both an
attitude or sensitivity to suffering and well-being,
and taking the expedient action to improve the
human condition. This combination of sensing and
acting makes it a more comprehensive concept to
link to technology, which often both senses and
acts, than for example empathy or intimacy, which
mainly have to do with the sensing aspects of
interaction [8]. In Fig.1 a first attempt is shown at
analyzing the dynamics of the compassion process
in a human to human interaction. From this
analysis different questions are derived, such as: is
compassion a P1+P2 process, or does it involve
only P2? Is P1 active and P2 passive? Is it a
voluntary or involuntary process? These and other
questions become even more relevant when being
transferred to technology. In Fig. 2 the same

Figure 2. Compassion dynamics analysis: Human-
Computer Interaction
compassion process is shown, but this time in the
context of Robot-Human interaction. In this
scenario we focus on interactive behaviour, rather
than specific instances that would lead to equally
specific technology solutions [9]. The main effort,
in the case of Fig. 2, is translating behaviour that is
typically human, into possible steps that are
replicable for a machine.
In the workshop, we would like to discuss how we
may adapt the compassion process to a machine.
What technology can we use to make a machine
compassionate technology? Does a machine need
to experience compassion itself or is it sufficient to
design detailed and targeted actions for a variety of
cases? For instance, if we analyze step 2, we went
from understanding the universality of human
suffering to comparing data from users/user groups
for interpretation. This step involves not only
sensing, but also understanding that: ‘other people
are sentient, feeling organisms: a skill that is
needed by the robots to treat other agents
accordingly’ [10], a principle key to the theory of
mind [11]. We asked ourselves what this could
Figure 3. Value System Exploration
mean for a machine, and for the sake of exploration
and theoretical speculation we propose that, for
example, the machine could compare data from a
large user group, in order to “understand” the
characteristics of that group and use it to provide
feedback. Within the context of mental healthcare
this discussion is particularly relevant to
understand near and distant future scenarios, for
instance it is important to consider what will
change for the therapist and for the client with the
adoption of e-health treatments. Thanks to
technology and its ubiquitous nature, the client is
acquiring more and more responsibility: private
medical data is accessible through web-platforms,
self-help modules are available in a number of
programs, web platforms and mobile applications
[12]. At the same time, the role of the therapist is
also transforming: it is no longer the case that
mental healthcare treatments have to be prescribed,
they can be chosen, the tools of the profession are
changing and learning iterations have to be put in
place to keep up with the updates [13]. Finally, in
Fig. 3 we made some considerations on the value
of compassion and its value system, the following
questions emerged: what value system is
compassion in? What is its

relation to the concept of humanity? How will it
change within the technology context?
2.2. Design for values
The focus of the therapeutic alliance of patient
and professional is human connection: alleviating
mental suffering and increasing wellbeing. In the
same way, there is a need to focus technology for
mental healthcare on compassion. As B. Friedman
states: “Value Sensitive Design is theoretically
grounded approach to the design of technology that
accounts for human values in a principled and
comprehensive manner throughout the design
process.” [14]. This design approach allows the
expression and embedding of values in products
and in ways of working. It affirms that a conscious
and explicit thinking about the values in our
designs is socially and morally significant.
However, the reflection on values has to take place
early on in the design process, when it can still
make a difference. We need to study how this
methodology can maximize the benefits of a
ubiquitous technology.
3. CONCLUSION
Designing for a shift in the way technology is
perceived is essential in general and especially in
the context of Mental Healthcare. In this position
paper we proposed a new way of designing
technology and robots for e-mental healthcare
based on pivotal human values. While the
traditional approach to both technology and mental
healthcare has failed to create a new model for a
more sustainable mental healthcare, we believe that
designing for compassion will not only allow to
intervene at the levels that are of interest for this
specific project, but also create new knowledge and
tools for designers and developers to be applied in
other areas. Through value-based and human-
centred design we want to create a better fit
between the context of mental healthcare and its
end-users. The challenge is centralizing the needs
and values of the users and integrating innovative
technologies, social interactive machines into their
ways of working. Finally, our aim is to use the
benefits machines offer in a compassionate and
human-supporting way, employing design for
compassion to engage and stimulate people in using
current e-health solutions, as well as new and
innovative solutions that enable mental health in a
different way.
Acknowledgment — The 'Designing
compassionate technology with high societal
readiness levels for mental healthcare' project has
received funding from NWO Social Sciences and
Humanities, Minddistrict and Dimence.
References
[1] M. A. Feijt, Y. A. de Kort, I. M. Bongers, and W. A.
IJsselsteijn, “Perceived Drivers and Barriers to the
Adoption of eMental Health by Psychologists: The
Construction of the Levels of Adoption of eMental Health
Model.,” J. Med. Internet Res., vol. 20, no. 4, p. e153,
Apr. 2018.
[2] M. Smits, B. Bredie, H. Van Goor, and P. P. C. C.
Verbeek, “Values that Matter: Mediation Theory and
Design for Values,” in proceedings of Academy of
Design Innovation Management Conference, 2019. I. S.
Jacobs and C. P. Bean, “Fine particles, thin films and
exchange anisotropy,” in Magnetism, vol. III, G. T. Rado
and H. Suhl, Eds. New York: Academic, 1963, pp. 271–
350.
[3] P. Cuijpers, M. Reijnders, and M. J. H. Huibers, “The
Role of Common Factors in Psychotherapy Outcomes,”
Annu. Rev. Clin. Psychol., vol. 15, no. 1.
[4] E. T. Bohlmeijer and G. J. Westerhof, “A New Model for
Sustainable Mental Health: Integrating Well-Being into
Psychological Treatment,” in Making an Impact on
Mental Health and Illness, J. Kirby and P. Gilbert, Eds.
London: Routledge.
[5] Ray Amit, Compassionate Artificial Intelligence:
Frameworks and Algorithms, Compassionate AI Lab,
2019.
[6] Christina Feldman & Willem Kuyken (2011) Compassion
in the landscape of suffering, Contemporary Buddhism,
Vol. 12, No. 1, pp. 144, May 2011.
[7] C. Strauss et al., “What is compassion and how can we
measure it? A review of definitions and measures,” Clin.
Psychol. Rev., vol. 47, pp. 21, Jul. 2016.
[8] P. Gilbert, “Compassion-focused therapy,” Cogn. Behav.
Ther., no. April 2009, pp. 140–165, 2012.
[9] Höök, Kristina & Löwgren, Jonas. (2012). Strong
Concepts: Intermediate-Level Knowledge in Interaction
Design Research. ACM Trans. Comput.-Hum. Interact.
[10] Hudson, Shally. (2020). Social Robots Based on Amit
Ray's Theory of Compassionate Artificial Intelligence, p.
6.
[11] Baron-Cohen, Simon. Theory of Mind in normal
development and autism. Prisme. 34. 2001
[12] G. D. S. Ludden, T. J. L. Van Rompay, S. M. Kelders, and
J. E. W. C. Van Gemert-Pijnen, “How to increase reach
and adherence of web-based interventions: A design
research viewpoint,” J. Med. Internet Res., vol. 17, no. 7,
2015.
[13] D. E. Bowen, “The changing role of employees in service
theory and practice: An interdisciplinary view,” Hum.
Resour. Manag. Rev., vol. 26, no. 1, pp. 4–13, 2016.
[14] Friedman B., Kahn P.H., Borning A., Huldtgren A. (2013)
Value Sensitive Design and Information Systems, p. 2. In:
Doorn N., Schuurbiers D., van de Poel I., Gorman M.
(eds) Early engagement and new technologies: Opening
up the laboratory. Philosophy of Engineering and
Technology, vol 16, Springer, Dordrecht.