Content uploaded by Julian Fietkau
Author content
All content in this area was uploaded by Julian Fietkau on Mar 09, 2021
Content may be subject to copyright.
Julian Fietkau
Mandy Balthasar
Volume 6
Schriften zur
soziotechnischen Integration
Professur für Mensch-Computer-Interaktion,
Universität der Bundeswehr München
ISSN 2194-0274 (Print)
ISSN 2194-0282 (Online)
www.soziotech.org
Compatibility of Support and Autonomy in
Personalized HCI
Bibliographic data according to the German National Library
The German National Library is recording this publication in the German National
Bibliography. Detailed bibliographic data is available online: https://portal.dnb.de
Series: Schriften zur soziotechnischen Integration
Published by:
Michael Koch & Florian Ott
Volume 6: Compatibility of Support and Autonomy in Personalized HCI
Contributed by:
Julian Fietkau & Mandy Balthasar
Reference / Citation
Fietkau, Julian & Balthasar, Mandy (2020): „Compatibility of
Support and Autonomy in Personalized HCI”. In: Koch, Mi-
chael; Ott, Florian (Hrsg.): Schriften zur soziotechnischen In-
tegration, Volume 6. Munich, Germany: Forschungsgruppe
Kooperationssysteme, Universität der Bundeswehr München.
https://schriften.soziotech.org/band6
https://doi.org/10.18726/2020_8
ISSN 2194-0274 (Print)
ISSN 2194-0282 (Online)
DOI 10.18726/2020_8
1st edition, November 2020
Cover design: Eva Stuke, Layout: Florian Ott
Professur für Mensch-Computer-Interaktion, Universität der Bundeswehr München
Werner-Heisenberg-Weg, 39, 85577 Neubiberg, Germany
E-Mail: schriften@soziotech.org
Website: https://schriften.soziotech.org
The content of this series is licensed to the public under the Creative Commons Attribution Non
Commercial No Derivatives 4.0 International license. This means that you are permitted to copy
and distribute this content, provided that the work is correctly attributed to the authors, the con-
tent is not used commercially, and the text is not edited, shortened or otherwise changed. For
further details see: https://creativecommons.org/licenses /by-nc-nd/4.0/
Schriften zur soziotechnischen Integration, Volume 6
i
Compatibility of Support and Autonomy in
Personalized HCI
Julian Fietkau, Universität der Bundeswehr München
Mandy Balthasar, Universität der Bundeswehr München
Abstract
Developers and designers of interactive digital systems are faced with many chal-
lenges, some less visible than others. One of these more subtle balancing acts is be-
tween personalized user support and unrestricted user autonomy. It raises the ques-
tion how a supportive design that takes a user’s needs and preferences into account
can be implemented within a technical system while at the same time allowing the
user to make decisions freely without restrictions. How can autonomy by design be
incorporated into the process? In this paper we describe the breadth of choice heu-
ristic, which can show where the support of the user by the system stops and a re-
striction of the user’s autonomy begins. To this end, we reference existing literature
on similar issues, develop our own conclusions and apply them to our work in a pro-
ject on IT for elderly people.
Keywords
Interaction design, user autonomy, value sensitive design, persuasive design,
privacy, ethics
Table of Contents
ii
Table of Contents
Abstract .......................................................................................................................................... i
Keywords ....................................................................................................................................... i
Table of Contents ....................................................................................................................... ii
1 INTRODUCTION ........................................................................................ 1
2 RELATED WORK ....................................................................................... 2
3 PROJECT BACKGROUND......................................................................... 3
4 THE BREADTH OF CHOICE HEURISTIC ............................................. 4
5 DISCUSSION ................................................................................................ 6
6 ACKNOWLEDGMENTS ............................................................................. 8
REFERENCES ....................................................................................................... 9
LIST OF AUTHORS .......................................................................................... 11
Introduction
1
1
1 Introduction
Steven Krug’s Don’t make me think (Krug 2014) has been a valuable guiding princi-
ple for the design of easy-to-operate systems, but its catchy title does not provide a
full picture for the responsible design of human-computer interaction (HCI). Clearly,
interactive systems cannot take all critical thought out of the user’s hands. The blurry
line between support and paternalism, between paving the way and prescribing the
way, must be clarified.
When we speak of autonomy, this does not mean self-legislation as described by Kant
(Kant 2012). Rather, autonomy refers to a process that opens up all possibilities for
an individual to interact or not with his physical and digital world, and to position
themselves within this exchange (Liggieri & Müller 2019). Thus autonomy can also be
seen here as authenticity (Misselhorn 2018).
The daily experience of people is increasingly influenced by digital technology. Their
data is collected, linked, analyzed and used in many ways, often unnoticed. In addi-
tion, the Privacy-by-Default Principle is frequently not applied, leading to the phe-
nomenon of the Privacy Paradox (Joinson et al. 2010, Balthasar & Gerl 2019).
In this paper, especially in Section 2, we gather theoretical knowledge about user au-
tonomy in the context of HCI. In Section 3 we share our approach taken in the Urban-
Life+ project to preserve the autonomy of users of a personalizable information sys-
tem. Subsequently, in Section 4 we present the breadth of choice heuristic, which is
our core contribution to the understanding of the role of user autonomy in the design
of digital systems. Finally, in Section 5 we subject our heuristic to a brief critical dis-
cussion.
Related Work
2
2
2 Related Work
One of the first scholars to think critically about the role of user autonomy in the con-
text of HCI was J. C. R. Licklider (Licklider 1960) in his work on human-computer
symbiosis, in which he anticipates a future reality of fast and dynamic information
flows between the human mind and computers. He concludes that a functional sym-
biosis of man and computer can only be achieved if man remains autonomous in his
ability to think and make decisions even as computing power increases.
In recent years, a section of HCI that deals with playful interactions has paid special
attention to the Self-Determination Theory (SDT) to explain motivation. The SDT as
described by Ryan and Deci (Ryan & Deci 2000) considers three core aspects of in-
trinsic motivation: besides competence and relation, the autonomy of the user is of
central importance. This autonomy should enable users to make their own decisions
at their own pace. Thus SDT, like Licklider, builds on the aspect of user autonomy.
According to a literature analysis by Tyack and Mekler (Tyack & Mekler 2020), the
use of SDT in HCI system design is mostly very abstract and therefore often superfi-
cial.
Two other related concepts are User Openness (Davis et al. 1989), which occurs in
relation to the appropriation of software solutions, and End-User Programming
(Lieberman et al. 2006), which originates from Participatory Design.
A number of promising procedures such as Ethics by Design (Spiekermann & Winkler
2020), Value Sensitive Design (Shilton 2018) or the (now often legally prescribed)
Data Protection by Design (Danezis et al. 2015) can be examined for the support of
an ethical design of technology. In addition, participatory ethics evaluation proce-
dures such as MEESTAR (Weber 2015) can provide concrete guidance. General brain-
storming and analysis tools such as the Ethics Canvas (Lewis et al. 2018) are also
available and will certainly be developed further.
Project Background
3
3
3 Project Background
Without stakeholder commitment, the aspect of user autonomy in HCI remains purely
theoretical and the effort an intellectual exercise. In order to avoid this, we took the
opportunity to incorporate the idea of autonomy by design into a larger project. This
project, named UrbanLife+, is a five-year endeavor in which a number of German-
speaking universities and companies participate. It has developed concepts and pro-
totypes that influence the behavior of seniors through an interactive system by en-
couraging them to participate in more outdoor activities (Fietkau & Stojko 2020).
The resulting system acts autonomously by collecting and aggregating data about the
local context without explicit instructions from the user. For this reason, the topic of
implicit interaction was thoroughly discussed within the design framework. The re-
sult is a system that already reacts to users entering its sensorially detectable area
(knowingly or unknowingly). Each system is designed to provide specific value, but
the human being has their unique dignity (Frankl et al. 2019), which is granted strong
legal protection in German Basic Law (Grundgesetz Article 1). Wherever the two (the
value of the system and the dignity of the human) may come into conflict, design de-
cisions need to be made with great care.
Since the system can take action autonomously, questions arise about possible viola-
tions of the human being’s rights, particularly regarding personal autonomy as well
as data sovereignty. In order to preserve this autonomy within the system, architec-
tural and design decisions were only made after deliberating ethical considerations.
This also involved the differentiation between two scenarios: on the one hand, a mo-
tivating support of the seniors to explore their environment and on the other hand,
the manipulation of people to leave existing routines and comfort zones.
To ensure that the user remains in control, the system described above offers the op-
tion to disable user recognition on an individual basis. Information provided to users
is usually personalized, but if the user prefers to interact anonymously, a simple but-
ton press makes this possible. This is enabled by conducting user recognition via Blue-
tooth in combination with a mobile phone app that provides a toggle switch. When
personalization is deactivated, the user’s identity remains hidden from the system as
well as from other nearby users. At the same time, however, the user’s own prefer-
ences previously defined in the system, such as data protection or the consideration
of accessibility in the selection of information, remain untouched. A conscious inte-
gration of autonomy prevents its loss to automation.
The Breadth of Choice Heuristic
4
4
4 The Breadth of Choice Heuristic
As described in Section 2, existing guidelines for ethical interaction design do not yet
make it easy for designers to assess the impact of concrete design decisions on user
autonomy. The provision of rich personalized functionality requires detailed infor-
mation about the user and their needs and preferences. How can we determine dur-
ing the development of a system whether its design respects the user’s autonomy?
In answering this question, it is important to note that the line between truly helpful
support and unwelcome paternalism cannot easily be drawn for the general case
since it depends strongly on the person in question.
The recognition of this line will vary from moment to moment and from situation to
situation, so the framework of this project allows users to draw this line themselves
by making their own decisions.
During the design process of our UrbanLife+ systems, we have developed a heuristic
to anticipate the impact of a design choice on user autonomy. The question is: Does
this interaction broaden the user’s spectrum of available actions, or does it nar-
row it? We observe that interaction designs which undermine user autonomy will
usually have a specific activity at their core that they want the user to perform, such
as watching a video advertisement. They will then attempt to ensure that this activity
is simpler and more convenient than any possible alternative, e.g. by making the but-
ton to close the advertisement as small and difficult to see as possible. The user’s abil-
ity to decide what their technology should do is intentionally subverted, manipulating
the user and undermining their autonomy.
In contrast, system designs that respect the user’s autonomy usually offer additional
options for action or provide equally transparent conditions for all options. While the
system may provide assessment information on less advisable options or even warn
the user about potential dangers, ultimately its goal is to empower the user to make
rational decisions based on all available information.
The logical argument for the breadth of choice heuristic is as follows:
1. If the system respects the user’s autonomy,
2. then the system leaves all important decisions to the user. If that is the case,
3. then the system extends the user’s range of possible actions.
And in the opposite case:
The Breadth of Choice Heuristic
5
4
1. If the system does not respect the user’s autonomy,
2. then the system makes important decisions on behalf of the user. If that is the
case,
3. then the system limits the user’s range of possible actions.
This breadth of choice heuristic is easier to evaluate and to discuss in the context of
a design process compared to the abstract criterion of respect for the user’s auton-
omy. When design decisions are made, it asks: Does the decision to be made aim to
increase or decrease the user’s range of available options?
Discussion
5
6
5 Discussion
As the size and complexity of a project increases, there is a tendency towards a loss of
autonomy for individuals within it. The responsibility for the resulting system be-
comes blurred, as partial results pass through many different hands and thought pro-
cesses. The awareness of having a certain moral responsibility for the ideas, functions
and their implementation is not automatic. Once an idea has developed into a first
prototype, and that prototype has made contact with the system’s context, it is no
longer possible to “reset” the context back to the earlier state, no matter how small
the step taken (Liggieri & Müller 2019). This is all the more reason to reflect on pos-
sible consequences during the design process and not just at the end.
Describing the work of astronauts in space capsules, Anders (Anders 1994) paints a
picture of humans serving the technical system instead of the other way around: as-
tronauts do not “fly” their capsule in the same sense that a pilot “flies” a plane, instead
they are flown by the capsule. This allegory describes the handing over of the two
human responsibilities of action and creation, leaving humans only with the opera-
tion of the system.
In our project we have attempted to avoid reducing the users’ decision-making to a
mere operation of the system. The goal is to offer choices which can be decided based
on knowledge gained from experience and self-reflection – a design of HCI which
makes Descartes’ anima rationalis (Liggieri & Müller 2019) possible – to be a free
soul.
Nonetheless, the phenomenon of decision paralysis, in which someone confronted
with one or several choices may experience anxiety and avoid or postpone the deci-
sion-making process even to their personal disadvantage, must not be underesti-
mated (Anderson 2003). The answer cannot be to leave every single decision up to
the user. As designers and engineers, we still need to make judgments about which
decisions are important enough to involve the user’s mental resources.
A codification of the goal autonomy within an ethical system development process is
not yet available and its preconditions must first be clarified. If, however, a large num-
ber of relevant actors are involved, as in many interdisciplinary projects like ours, this
process has the potential to produce a binding result.
There is no simple, straightforward answer for the problem of support and autonomy
in HCI, just like there can be no universally optimal chess move (Frankl et al. 2019),
but one possible next step could be to derive more concrete design guidelines from
the presented heuristic. By bringing this conversation into the wider HCI community,
Discussion
7
5
we are hoping not only that our heuristic can be useful to practitioners, but also that
it can serve as a building block in future theories on ethical system design that re-
spects user autonomy.
Acknowledgments
6
8
6 Acknowledgments
This article was originally submitted and accepted for the NordiCHI 2020 workshop
“Strengthening human autonomy in the era of autonomous technology” organized by
Tone Bratteteig, Diana Saplacan, Rebekka Soma, and Johanne Svanes Oskarsen. The
workshop submissions were not published. At the time of this publication, the work-
shop participants are considering future work to build upon core aspects of the work-
shop discussion and contributions, including this article.
As the first author’s contributions to this paper were developed in the scope of the
UrbanLife+ project, this work has been partially supported by the Federal Ministry
of Education and Research, Germany, under grant 16SV7443. We thank all project
partners for their commitment.
References
9
References
ANDERS, GÜNTHER (1994): Der Blick vom Mond: Reflexionen über Weltraumflüge. In:
Beck’sche Reihe 1056, Orig.-Ausg. Auflage. München: Beck.
ANDERSON, CHRISTOPHER J (2003): The Psychology of Doing Nothing: Forms of Decision
Avoidance Result from Reason and Emotion. In: Psychological Bulletin, S. 139–
167.
BALTHASAR, MANDY & GERL, ARMIN (2019): Privacy in the toolbox of freedom. In: 12th
CMI Conference on Cybersecurity and Privacy (CMI). Copenhagen, Denmark, S. 1–
4.
DANEZIS, GEORGE; DOMINGO-FERRER, JOSEP; HANSEN, MARIT; HOEPMAN, JAAP-HENK;
METAYER, DANIEL LE; TIRTEA, RODICA & SCHIFFNER, STEFAN (2015): Privacy and Data
Protection by Design – from policy to engineering.
DAVIS, FRED D; BAGOZZI, RICHARD P & WARSHAW, PAUL R (1989): User Acceptance of
Computer Technology: A Comparison of Two Theoretical Models. Management
Science, 8/1989 (35), S. 982–1003.
FIETKAU, JULIAN & STOJKO, LAURA (2020): A system design to support outside activities
of older adults using smart urban objects. In: Proc. Europ. Conf. on Computer-
Supported Cooperative Work 2020. EUSSET.
FRANKL, VIKTOR E; BAUER, JOACHIM & VESELY, FRANZ (2019): Über den Sinn des Lebens,
1st. Auflage. Weinheim, Germany: Beltz.
JOINSON, ADAM; REIPS, ULF-DIETRICH; BUCHANAN, TOM & PAINE SCHOFIELD, CARINA (2010):
Privacy, Trust, and Self-Disclosure Online. Human-Computer Interaction, 1/2010
(25). Taylor & Francis, S. 1–24.
KANT, IMMANUEL (2012): Grundlegung zur Metaphysik der Sitten. In: Reclams
Universal-Bibliothek 4507. Ditzingen, Germany: Reclam.
KRUG, STEVE (2014): Don’t Make Me Think, Revisited: A Common Sense Approach to Web
Usability. In: Voices That Matter, 3rd. Auflage. San Francisco, CA, USA: New
Riders.
References
10
LEWIS, DAVE; PANDIT, HARSHVARDHAN J; DEVINNEY, HANNAH & REIJERS, WESSEL (2018):
Towards an Open Data Vocabulary for Canvas Driven Innovation Ethics. In:
WATERMAN, K KRASNOW; MCGUINNESS, DEBORAH L & HENDLER, JAMES A (Hrsg.):
Proceedings of the Workshop on Semantic Web for Social Good 2018, 2182.
Monterey, CA, USA.
LICKLIDER, JOSEPH CARL ROBNETT (1960): Man-Computer Symbiosis. IRE Transactions
on Human Factors in Electronics, 1/1960 (HFE-1), S. 4–11.
LIEBERMAN, HENRY; PATERNÒ, FABIO & WULF, VOLKER (2006): End-User Development. In:
WirtschaftsinformatikHuman-Computer Interaction 9, 1st. Auflage. Dordrecht,
Netherlands: Springer.
LIGGIERI, KEVIN & MÜLLER, OLIVER (2019): Mensch-Maschine-Interaktion: Handbuch zu
Geschichte – Kultur – Ethik, 1st. Auflage. Stuttgart, Germany: J.B. Metzler.
MISSELHORN, CATRIN (2018): Grundfragen der Maschinenethik. In: Reclams Universal-
Bibliothek 19583, 2nd. Auflage. Ditzingen, Germany: Reclam.
RYAN, RICHARD M & DECI, EDWARD L (2000): Self-Determination Theory and the
Facilitation of Intrinsic Motivation, Social Development, and Well-Being.
American Psychologist, 1/2000 (55), S. 68–78.
SHILTON, KATIE (2018): Values and Ethics in Human-Computer Interaction.
Foundations and Trends in Human--Computer Interaction, 2/2018 (12), S. 107–
171.
SPIEKERMANN, SARAH & WINKLER, TILL (2020): Value-based Engineering for Ethics by
Design. SSRN Electronic Journal, 2020.
TYACK, APRIL & MEKLER, ELISA (2020): Self-Determination Theory in HCI Games
Research: Current Uses and Open Questions. In: Proceedings of the 2020 CHI
Conference on Human Factors in Computing Systems, CHI ’20. New York, NY,
USA: Association for Computing Machinery.
WEBER, KARSTEN (2015): MEESTAR : Ein Modell zur ethischen Evaluierung sozio-
technischer Arrangements in der Pflege- und Gesundheitsversorgung.
Technisierung des Alltags – Beitrag für ein gutes Leben?, May/2015, S. 247–262.
List of Authors
11
List of Authors
Julian Fietkau M.Sc.
https://fietkau.science/
julian.fietkau@unibw.de
Julian Fietkau studied computer science and human-computer
interaction at University of Hamburg, then worked for a year as
a research assistant at Bauhaus University Weimar until he
joined Universität der Bundeswehr München in 2016. Since
then, he has been working as a PhD student with Prof. Dr. Michael Koch in the field
of human-computer interaction, especially with regard to the research project Ur-
banLife+.
Mandy Balthasar M.Sc.
https://balthasar.digital/
mandy.balthasar@unibw.de
Mandy Balthasar is a research assistant at the Institute of Soft-
ware Engineering at Universität der Bundeswehr München and
a PhD student at University of Passau. Her research focuses on
understanding and designing mechanisms for collective deci-
sion making and collective behavior, both in artificial and living collectives. She also
teaches at various universities in German-speaking countries and for the Springer Na-
ture Campus. For several years, she has served as head of the Usable Security & Pri-
vacy Working Group of the German UPA and is involved in numerous national and
international technology associations, such as ACM, GI, ISACA, UPA, and others.
XII
Developers and designers of interactive digital sys-
tems are faced with many challenges, some less
visible than others. One of these more subtle bal-
ancing acts is between personalized user support
and unrestricted user autonomy. It raises the
question how a supportive design that takes a
user’s needs and preferences into account can be
implemented within a technical system while at
the same time allowing the user to make decisions
freely without restrictions. How can autonomy by
design be incorporated into the process? In this
paper we describe the breadth of choice heuristic,
which can show where the support of the user by
the system stops and a restriction of the user’s au-
tonomy begins. To this end, we reference existing
literature on similar issues, develop our own con-
clusions and apply them to our work in a project
on IT for elderly people.
Compatibility of Support and Autonomy in
Personalized HCI
Schnittstellen Middleware
Activity Streams
Schema Authentifizierung Dienst
Mashup Filterung
Daten Protokolle Caching
Social Software
Aggregation API
ubiquitär Information
multi-user Interaktionszonen
Interaktive Großbildschirme
Freudvolle Nutzung Usability
Visualisierung Benutzerakzeptanz
Sozialer Kontext
Tablets & Smartphones
pervasive multi-touch
Informationsstrahler
Awareness Simplicity
Social Guidelines Enterprise 2.0
Aneignung Einführung
Anforderungsanalyse Motivation
Nutzungsoffenheit
Erfolgsmessung Social Business
Partizipation
Wissensmanagement Community
Social Networking
