Content uploaded by Marc Hassenzahl
Author content
All content in this area was uploaded by Marc Hassenzahl
Content may be subject to copyright.
An Interaction Vocabulary. Describing
the How Of Interaction
Abstract
New technologies and opportunities in interaction
design also come with new responsibilities. Besides the
product's visual aesthetics, design needs to address the
aesthetics of interaction. We discuss the various
starting points of emerging approaches and then
present an own approach to the How of interaction. We
suggest an interaction vocabulary, i.e., a set of eleven
dimensions of descriptive, non-judgmental, non-
technology bound attributes of interaction. First insights
from applying the vocabulary in design and evaluation
studies and future research are discussed.
Author Keywords
Aesthetics of interaction; interaction design; How of
interaction; interaction qualities; interaction
vocabulary.
ACM Classification Keywords
H.5.m. Information interfaces and presentation (e.g.,
HCI): Miscellaneous.
Introduction
Ever new technology creates a multitude of
opportunities for interaction design. This is
accompanied by the dissolution of "traditional", mostly
technology-driven matches between function and
interaction. While for a good part of the last century
Copyright is held by the author/owner(s).
CHI 2013 Extended Abstracts, April 27–May 2, 2013, Paris, France.
ACM 978-1-4503-1952-2/13/04.
Sarah Diefenbach
Folkwang University of Arts
Universitätsstrasse 12
45141 Essen Germany
sarah.diefenbach@folkwang-uni.de
Eva Lenz
Folkwang University of Arts
Universitätsstrasse 12
45141 Essen Germany
eva.lenz@folkwang-uni.de
Marc Hassenzahl
Folkwang University of Arts
Universitätsstrasse 12
45141 Essen Germany
marc.hassenzahl@folkwang-uni.de
Work-in-Progress: Interaction and Experience Design
CHI 2013: Changing Perspectives, Paris, France
607
"dialing a phone number" actually involved a dial (due
to the technical demands of pulse dialing), this
nowadays may involve pressing buttons, various forms
of touch or freeform gestures or speech input.
Interaction becomes less constrained by the underlying
technical effect, but by available interaction technology
in itself. The latter's possibilities are increasing,
creating an unsurpassed freedom of choice for the
interaction designer. Just as a product's form,
materiality and resulting visual aesthetic, small design
choices can be deciding for a resulting aesthetic of
interaction (e.g., [4]).
In the following, we discuss exemplary approaches to
the aesthetics of interaction in the HCI and Interaction
Design literature. We then suggest an additional
approach based on descriptive, elementary attributes of
interaction, i.e., a first interaction vocabulary. We
present examples of practical application in design,
research and evaluation, and provide an outlook on
future research.
Approaches to the Aesthetic of Interaction
The available frameworks and approaches to the
aesthetics of interaction focus on different levels.
Normative approaches suggest qualities of an especially
"good" or "aesthetic" interaction and according design
principles. For example, Djajadiningrat et al. [3]
promote movement-oriented interaction styles, and
suggest principles such as the richness of user actions,
i.e., a high variety in the repertoire of possible user
actions. Others emphasize various experiential qualities
of interaction, such as its dramaturgical structure [9],
its expressions (e.g., anxiety, thrill, trust) [6], or one's
personal sensations, experiences and reflections
connected to the system in context [12]. Some
underline the performative perspective within socially
situated interaction [1] or the emerging social dialogue
between user and the system itself [10]. All these
approaches take a high level perspective on interaction.
They focus on emotional experiences and certain
addressed psychological needs. In sum, they focus on
the reasons (i.e., the Why), which make interacting
meaningful to people (see Fig. 1 for an overview).
From a design perspective, those high level experiential
qualities are important – they imply crucial experiential
outcomes of product use. However, high level qualities
are mediated and created through the arrangement of
concrete interactions on a lower level. This How of
interaction takes place on the level of "operations" or
motor actions (e.g., turning a knob, pressing a lever).
Psychological action theories point out that motor
actions are highly automatized and often performed
unconsciously (e.g., [5]). This is a barrier to conscious
thinking about interaction in terms of its basic qualities
(e.g., powerful, slow). Consequently, only a limited
number of approaches focus on the aesthetics of
interaction on the How-level. Those who do mostly
focus on a specific technology or domain such as
graphical user interfaces [8], gestural interfaces [14],
or movement based interaction [13]. Others mix
qualities of different levels without making this explicit
or discussing the qualitative difference between the
Why and the How. Lim and colleagues [7], for example,
use physical qualities (e.g., movement range) side by
side with qualities addressing complex experiential
qualities (e.g., expectedness) without much reflection
about this large conceptual difference. In addition,
many approaches certainly address relevant qualities,
but it remains unclear on what basis this particular set
of qualities was selected as relevant. Hence, we aimed
at a deeper and systematic exploration of the qualities
Figure 1. Interaction qualities on
the Why- and How-level.
Work-in-Progress: Interaction and Experience Design
CHI 2013: Changing Perspectives, Paris, France
608
of the How of interaction, the realm of interaction
design. Though resulting in highly complex experiences,
on a lower level, interaction consists of concrete motor
actions with particular qualities. We believe that
designing for high quality, aesthetic interaction requires
a deliberate consideration of interaction qualities
already on the How-level. To support this, we
developed an interaction vocabulary. Having a
vocabulary facilitates the explicit description of
interaction and reveals possible starting points for
shaping the interaction at the same time. In the
remainder of the paper, we describe the vocabulary and
provide examples of its use it in interaction design.
An Interaction Vocabulary
Our vocabulary currently consists of a set of eleven
dimensions (Fig. 2). This set resulted from an iterative
process of refinement. Aiming for a general eclectic
approach, our first step was a systematic derivation of
all possible differentiating factors between interaction
concepts with one same function: A simple function
(control of a light) was used as a trigger for collecting
alternative interaction concepts in an online survey.
The resulting collection of more than hundred ways of
interacting with the light was then used for extracting
differentiating attributes. Given our focus on the How-
level, we searched for elementary qualities, in the
sense of descriptive, non-judgmental, non-technology
bound attributes. Attributes were expressed as opposite
word pairs (e.g., slow – fast, originally in German, see
[2]). Figure 3 presents examples of interactions and
differentiating attributes. In a number of successive
studies and workshops, we further refined the exact
wording and presentation of the vocabulary, and
explored its methodological and practical applicability in
design and research.
Applying the Interaction Vocabulary
Inspiration and personal reflection
The vocabulary's range of application is manifold. We
mainly understand the interaction vocabulary as an
inspirational tool for designers (see also [11]). The
vocabulary serves as a repertoire of interaction
qualities to choose from. It hints at potential
alternatives without being prescriptive. Having clarified
the wished for interaction qualities, the designer may
create concrete ideas for realization in the context of
the given domain and technical constraints.
In a course on interaction design at the Folkwang
University of Arts, for example, students were given the
assignment to design an interaction for opening a box,
containing a particular item (represented by a paper
card, e.g., a precious piece of jewelry, a birthday
present, first aid equipment etc.). However, before
realizing a concrete "opening interaction" through a
paper prototype, they were asked to clarify the
interaction's basic qualities, e.g., should it feel fast or
slow? Mediated or rather direct? Here, the interaction
vocabulary was used to clarify and specify the
intended, most characteristic qualities. Julia Lackas, for
example, designed an interaction for a box containing
an engagement ring. She wanted the interaction to feel
slow, gentle, stepwise and delayed. After carefully
opening the box, it plays the couple's favorite song.
After the song has faded, the top part lifts a bit and
reveals the precious content (Fig. 4). In contrast,
Selina Maleska chose quite different interaction
qualities for her box, containing a child's long awaited
birthday present (a Lego Star Wars' Millennium Falcon).
She wanted the interaction to be fast, powerful, fluent
and instant. Her box is perforated and features a
"lightsaber" that allows for a fast and instant access to
Figure 2. Interaction vocabulary.
Eleven dimensions of descripti ve,
non-judgmental, non-technology
bound at tributes.
Figure 3. Examples of interaction
concepts and extracted dimensions.
Above: Dimmer switch (fluent) vs.
flip switch (stepwise).
Below: Switch on the lamp (direct )
vs. switch on the wall (mediated).
Work-in-Progress: Interaction and Experience Design
CHI 2013: Changing Perspectives, Paris, France
609
the content; the sound of tearing paper makes it even
more powerful (Fig. 5). Thus, an important insight was
that the function alone (opening a box) is not sufficient
to determine the aesthetics of interacting with it.
Depending on the experience to be created, interacting
could feel slow, gentle, stepwise, and delayed or fast,
powerful, fluent and instant. The interaction with the
box containing the engagement ring reflects upon the
couple's special moment, by making it last a little
longer, and giving it a ceremonial and ritualistic
structure. The interaction with the box containing the
Millenium Falcon is oriented on the kid's needs in that
moment: finally getting the long wished-for present,
unwrapped in a swirl of activity, with absolutely no
need for reverent slowness or any further delay.
Communication
Besides being a source of inspiration, the vocabulary
supports the explicit communication about interaction
qualities. We experienced this as very helpful in student
courses, participatory design projects as well as
interdisciplinary research projects. Note, however, that
the interaction vocabulary only supports a more
conscious and purposeful interaction design. The
responsibility for determining appropriate interaction
qualities remains with the designer.
External feedback and evaluation
The vocabulary is further a tool for critical reflection
and evaluation of interaction concepts at different
points in time of the design process. For this purpose,
the vocabulary is presented as a semantic differential
questionnaire (e.g., 1=slow, 7=fast, see Fig. 6). Users
are confronted with the interaction concept and then
rate their impression of the interaction along the eleven
dimensions. The resulting interaction profile captures
the most characteristic qualities of the interaction, as
perceived by users. If applied to several interactions,
the profiles reveal felt differences between the
interactions. Figure 7 shows an example: the felt
differences between using software with the computer
mouse versus performing the same task via touch
display (from a research project with our industrial
partner a3 systems). As the comparison of interaction
profiles shows, the most (and statistically) significant
differences are that interacting with the mouse feels
more precise whereas the interaction via touch display
provides a higher degree of spatial proximity (Fig. 7).
While certainly not surprising, these results show that
users are sensitive to the differences in how an
interaction feels. Instead of talking about "mouse
versus touch", interaction designer should focus more
on the qualities of interaction resulting from different
technologies.
Profiles obtained from external raters can be further
compared to the intended profile specified by the
designer (see [11] for an example). If both profiles
follow the same form, the How of interaction is
perceived by others as intended. In contrast,
discrepancies on single dimensions reveal qualities not
being realized as intended yet, and, thus, showing
potential for improvement.
Connecting the How and the Why
The example of the gift boxes revealed that
"appropriate" interactions can be quite different. This is
because design rationales were derived from different
intended "opening experiences". Designing for aesthetic
interaction thus requires considering both in parallel,
interaction and experience, the How and the Why, how
the interaction feels and whether felt qualities facilitate
Figure 4. Interaction for a box for
an engagement ring: slow, ge ntle,
stepwise, de layed.
Figure 5. Interaction for a box for
a child’s long awaited present: fast,
powerful, fluent, instant.
Figure 6. The interaction
vocabulary as a semantic
differential questionnaire.
Work-in-Progress: Interaction and Experience Design
CHI 2013: Changing Perspectives, Paris, France
610
or provide a barrier to the intended experience. To our
mind, an aesthetic interaction stems from the
alignment between these two levels. The awareness of
potential variations within the How of interaction is thus
only a first step. The next question is how to combine
these ingredients to create a particular experience.
Hence, after having established a way to explicate how
an interaction feels, our future work will attend to the
relationship between interaction qualities on the How-
level and experiential qualities on the Why-level.
As a first exploration, we already conducted a workshop
where participants presented interactive products from
their daily life. We asked them to reflect upon how it
felt to operate the product and why this way of
interaction was satisfying for them (or not). This
resulted in a list of matches of interaction and
experiential qualities. An initial analysis revealed some
typical patterns between interaction qualities and
psychological needs addressed in the related
experiential reports. For example, fluent interaction
generally supported feelings of autonomy. As one
participant explained: "I can do it exactly the way I
want it. I have continuous influence and don't need to
stick to predefined grades" (here: interacting with
Grundig Mini Chopper Multicut). In contrast, another
participant reflected on the feelings of security and
routine he got from stepwise interaction, split into
smaller consecutive stages: "Every step has got
meaning. It's like a ritual" (here: interacting with
Bialetti espresso machine). If it should turn out that
there is a relatively stable set of such relations between
interaction and experiential qualities, this set could be
integrated in a tool for interaction design as well, not
only highlighting possible variations in interaction but
also providing a link to resulting experiential
consequences. In a first prototype this was realized by
presenting the interaction vocabulary as a ring binder,
consisting of single segments for each of the
dimensions (Fig. 8). Depending on which of the two
poles are dominant, these may be turned in one or the
other direction, which then reveals associated
experiential qualities as well. However, further research
is needed to explore which kind of tools would be most
useful for designers.
Conclusion
The interaction vocabulary is a set of qualities inherent
to any interaction. It can be used as inspirational tool,
outlining the potential ways to design a particular
interaction. In addition it provides a means to talk more
systematically about how an interaction feels – within
design teams but also with potential users. Given that
the How of interaction is highly automated and
unconscious, providing a vocabulary for communication
is essential. While most people are able to describe in
detail all sorts of positive experience on an emotional
and meaning-related level (the Why), describing how
an interaction feels is more difficult. The interaction
vocabulary is a first step towards a technology
independent perspective on interaction. Especially
practitioners revealed to us that design decisions on the
level of interaction are often made incidentally. "Design
rationales" don't stem from visions of how the
interaction should feel or which kind of experience
should be supported, but are driven by technology and
established solutions. Through its range of qualities, the
interaction vocabulary further establishes the notion of
an aesthetic of interaction beyond efficiency and
beyond an obsession with the currently most
fashionable interaction technologies.
Figure 7. Felt differences in
interaction qualities between
computer mouse and touch display.
! Mouse
! Touch display
Work-in-Progress: Interaction and Experience Design
CHI 2013: Changing Perspectives, Paris, France
611
Acknowledgements
This work was supported by the German Federal
Ministry of Education and Research (BMBF), project
proTACT (Grant: 01 IS12010F).
References
[1] Dalsgaard, P. and Hansen, L.K. Performing
perception—staging aesthetics of interaction. ACM
Transactions on Computer-Human Interaction 15, 3
(2008), 1–33.
[2] Diefenbach, S., Hassenzahl, M., Kloeckner, K.,
Nass, C., and Maier, A. Ein Interaktionsvokabular:
Dimensionen zur Beschreibung der Ästhetik von
Interaktion. In H. Brau, S. Diefenbach, K. Göring, M.
Peissner and K. Petrovic, eds., Usability Professionals
2010. German Chapter der Usability Professionals’
Association e.V, 2010, 27–32.
[3] Djajadiningrat, T., Matthews, B., and Stienstra, M.
Easy doesn’t do it: skill and expression in tangible
aesthetics. Personal and Ubiquitous Computing 11, 8
(2007), 657–676.
[4] Hummels, C. and Overbeeke, K. Special Issue
Editorial: Aesthetics of Interaction Where Do We
Stand Now? What is Our View on the Aesthetics of.
International Journal of Design 4, 2 (2010), 1–2.
[5] Kaptelinin, V. and Nardi, B.A. Acting with
Technology: Activity Theory and Interaction Design.
MIT Press, 2006.
[6] Landin, H. Anxiety and Trust And Other Expressions
of Interaction. Chalmers University of Technology,
2009.
[7] Lim, Y., Lee, S.-S., and Lee, K. Interactivity
attributes: a new way of thinking and describing
interactivity. Proceedings of the 27th international
conference on Human factors in computing systems,
(2009), 105–108.
[8] Lim, Y., Stolterman, E., Jung, H., and Donaldson, J.
Interaction gestalt and the design of aesthetic
interactions. Proceedings of the 2007 conference on
Designing pleasurable products and interfaces DPPI 07,
(2007), 239-254.
[9] Löwgren, J. Towards an articulation of interaction
aesthetics. New Review of Hypermedia and Multimedia
15, 2 (2009), 129–146.
[10] Marti, P. Perceiving while being perceived.
International Journal of Design 4, 2 (2010), 27–38.
[11] Nass, C., Klöckner, K., Diefenbach, S., and
Hassenzahl, M. DESIGNi – A Workbench for Supporting
Interaction Design. Proceedings of the 6th Nordic
Conference on Human-Computer Interaction: Extending
Boundaries, (2010), 747–750.
[12] Petersen, M., Iversen, O., Gall Krogh, P., and
Ludvigsen, M. Aesthetic Interaction: a pragmatist’s
aesthetics of interactive systems. Proceedings of the
5th conference on Designing interactive systems:
processes, practices, methods, and techniques, (2004),
269–276.
[13] Ross, P.R. and Wensveen, S.A.G. Designing
aesthetics of behavior in interaction: Using aesthetic
experience as a mechanism for design. International
Journal of Design 4, 2 (2010), 3–13.
[14] Saffer, D. Designing Gestural Interfaces. O’Reilly,
Sebastopol, 2009.
Figure 8. Interaction vocabulary as
a ring binder.
Work-in-Progress: Interaction and Experience Design
CHI 2013: Changing Perspectives, Paris, France
612
