Papers Session 4: Input/Output TEI ’20, February 9–12, 2020, Sydney, NSW, Australia

AromaCue – A Scent Toolkit to Cope with Stress

Using the 4-7-8 Breathing Method
Zilan Lin, Kai Kunze, Atsuro Ueki, Masa Inakage
Keio University
Graduate School of Media Design
Yokohama, Japan
[email protected],[email protected],[email protected],[email protected]

Figure 1: The AromaCue Concept: Training a breathing at home using scent conditioning. After breath training the user can
receive the scent, if she presses a stress ball.

ABSTRACT
In this paper, we present AromaCue, an initial design
for a scent-based toolkit to cope with stressful
situations us ing scent conditioning. The AromaCue
toolkit consists of two parts: a breath training device
using multiple stimuli and a wearable scent-emitting
device (with a stress ball as an activator). Scent can
trigger emotional memories. In the initial design, we
utilized the properties of various scents as retrieval
cues for consciousness breathing. In a compara tive
experiment, eight participants showed a significant
heart rate decrease several minutes after a stressor
(the Stroop test) when a scent cue was present, but
not without it. One week

Permission to make digital or hard copies of all or part of this work

for personal or classroom use is granted without fee provided that
copies are not made or distributed for profit or commercial
advantage and that copies bear this notice and the full citation on the
first page. Copyrights for components of this work owned by others
than ACM must be honored. Abstracting with credit is permitted. To
copy otherwise, or republish, to post on servers or to redistribute to
lists, requires prior specific permission and/or a fee. Request
permissions from [email protected].
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
© 2020 Association for Computing Machinery.
ACM ISBN 978-1-4503-6107-1/20/02. . . $15.00
https://doi.org/10.1145/3374920.3374940
265
 of user study revealed significant improvement on
Depres sion Anxiety and Stress Scale (DASS-21) ACM Reference Format:
scores. Therefore, AromaCue can help users to cope Zilan Lin, Kai Kunze, Atsuro Ueki, Masa Inakage. 2020.
AromaCue – A Scent Toolkit to Cope with Stress Using the
with stress.
4-7-8 Breathing Method. In Fourteenth International
Conference on Tangible, Em bedded, and Embodied
CCS CONCEPTS
Interaction (TEI ’20), February 9–12, 2020, Sydney, NSW,
• Human-centered computing→User interface design; Australia. ACM, Sydney, SYD, Australia, 8 pages.
Interaction techniques; Interaction design theory, https://doi.org/10.1145/3374920.3374940
concepts and paradigms; Ubiquitous and mobile
computing systems and tools.
1 INTRODUCTION
KEYWORDS Can you remember the last time you were in a
Scent conditioning; scent-assisted learning; stressful sit uation? Maybe you were pitching a project
scent-based no tification proposal, in a
Papers Session 4: Input/Output TEI ’20, February 9–12, 2020, Sydney, NSW, Australia TEI ’20, February 9–12, 2020, Sydney,

NSW, Australia Lin et al.

job interview, or taking an oral examination. Stress is a (Stroop test) was introduced, (3) we also conducted a
nat ural biological response to any threat, and it one week user study revealing significant improvement
usually causes emotional and physical symptoms. on Depression Anxiety and Stress Scale (DASS 21)
Working under pressure, however, can also motivate scores for users applying our design toolkit. Finally, we
productivity, while overreaction to stress (such as will discuss best practices and limitations.
shortness of breath, dizziness, and nau sea) can lead
to adverse effects on occupational or academic 2 RELATED WORK
performance. There has been increasing interest in scent
In stress regulation, the term “the relaxation interactions in HCI research [14, 15, 18]. This section
response" coined by Dr. Herbert Benson is an opposite explores the oppor tunities for scent-based designs in
reaction to the fight or flight response. This stress HCI from three aspects: scent as an ambient stimulus,
management technique as sists people in reducing scent as a retrieval cue, and scent with other media.
levels of physical and mental arousal and increasing Our related works reviewed the role of scent played in
well-being [4, 27]. This response shares com mon the design of equipment and supported the
features with meditative exercises (such as yoga, Zen conceptual creation of the AromaCue toolkit.
Buddhism, and mindfulness meditation), helping
people to cope with stress [27]. Four essential Scent as an ambient stimulus
elements are taken from this technique help to elicit In psychology research, classical conditioning,
the relaxation response:“(1) a quiet environment; (2) a discovered by Ivan Pavlov, is a process of learning or
mental device; (3) the adoption of a passive attitude; cultivating new
and (4) a comfortable position" [4, 27]. As with many
other relaxation techniques, practitioners may not
always be able to evoke the relaxation response and
deal with stressful situations. Given this limitation, we
based our design thinking on olfactory conditioning
theory to elicit the relaxation response and
concentrate on the benefits of deep breathing for
treating stress, anxiety, and depression.
The main contributions of this paper are as follows:
(1) We present a scent-based toolkit to train and elicit
rhythmic breathing in stressful situations. The
AromaCue toolkit con tains a breathing training device
mixing visual and olfactory stimuli that aids users to
elicit the relaxation response and a wearable
scent-emitting system that sends aroma cues to
simulate a conditioned response (deep breathing) for
coping with stressful situations. (2) We designed and
implemented a comparative experiment in which eight
participants showed a significant heart rate decrease
using our toolkit several minutes after a stressor
 cue for classi cal conditioning [31]. However, Chu
noticed that olfactory conditioning is rarely used in
humans to control behavior consciously. As a result,
he presented successful experimen tal results on
olfactory conditioning, demonstrating positive
performance in humans [6]. In similar work to ours,
Amores et al. discuss the importance of smell in
changing cognitive behavior [1]. There is additional
work from Amores et al. looking into scent to aid
relaxation in virtual reality [2]. Our wearable
scent-emitting device follows work from Dobbel stein
et al. [10], but their application scenarios are vastly
different, focusing on ambient notifications.

Scent as a retrieval cue

Scent can activate the olfactory sensation and trigger
the limbic system in which emotional responses and
memories reside. Kaye [17, 19] did some of the early
works. He outlines the future uses for scent
technology. The most intriguing application for
olfactory technology is to explore the poten tials for
scent-assisted learning. Kaye indicates that smell can
facilitate learning and memorization, and
"scent-evoked memories are deeper and more
emotional"[17]. In the de sign field, Sissel let al.
designed a smell memory kit[26]. She used smell as a
retrieval cue, helping the individual to build a special
connection to the past and enrich their sensorial
experiences.

Scent with other media

Lai et al. presented a hands-on facility offering
opportunities to design olfactory experiences in real-life
applications [20]. Olfactory design is applied to
multi-sensory experiences ranging from VR/MR
applications to gaming-enhanced im mersion,
cross-modal effects, and cognitive and work per
formance [5, 8, 9, 12, 22, 23, 30]. Several publications
focus on meditation and relaxation. Patibanda et al.
presented a training device for breathing using wind
sounds [24]. It fo cusses on isolating people from their
physical surroundings to cultivate mindful breathing.
Patibanda et al. believe that multi-stimuli training can
help to foster stress-relief skills. They describe scent
conditioning as having potential use in behavioral
change technologies, in breaking existing habits and
form new ones [25].

3 THE AROMACUE TOOLKIT: AN OLFACTORY

CONDITIONING PARADIGM
In this section, we discuss our design approach
behind the AromaCue toolkit towards helping users to
relax and re duce stress. The AromaCue toolkit is
composed of two parts: a breathing training device and
a wearable scent-emitting system. The goal of this
266 design is to activate the relaxation re sponse to slow
habits based on paired stimuli. An olfactory stimulus is down muscles and organs and increase blood
often used to improve memory and can be a reliable
Papers Session 4: Input/Output TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
 AromaCue – A Scent Toolkit to Cope with Stress Using the 4-7-8 Breathing Method TEI ’20, February 9–12, 2020, Sydney, NSW, Australia

flow to the brain. We based our design thinking for this on

psychological perspectives, employing olfactory condition
ing theory. Previous related works explained the advantages
and possibilities of using scent in HCI. Smell is an essential
component in the AromaCue toolkit to facilitate learning
the breathing procedure and bring conscious awareness to
olfactory memory. We also believe users can develop a
habit
of deep breathing through the olfactory conditioning tech
nique.
Olfactory conditioning produces a learned response by as
sociating two stimuli, and it contains three primary phases:
before, during, and after conditioning. In our design, we
present a breathing training device that satisfies the four
essential criteria (shown in figure 2) to elicit the relaxation
response. These features serve as unconditioned stimuli
that
will automatically make users feel relaxed. For the breath
training technique, we choose the 4-7-8 breathing technique
as it has shown an effect on self-regulating anxiety and
stress
in several clinical studies [16]. This method regulates breath
ing and is supposed to reduce stress and anxiety by following
five steps: “(1) empty the lungs of air; (2) breathing in week. (5) Wear the scent-emitting system and press
quietly through the nose for 4 seconds; (3) holding the the stress ball when stressed to emit scent.
breath for a count of 7 seconds; (4) exhaling
vigorously through the mouth, pursing the lips and
making a ‘whoosh’ sound, for 8 seconds; (5) repeat the
cycle up to 4 times." as described by Fletcher et
al.[11]. The scent at this stage triggers no effect until it
is paired with the procedure of breath training. Dur ing
conditioning, users practice the breathing exercise
while being also exposed to a scent. Through repetitive
association between two stimuli, we hope we can elicit
the relaxation response through the scent. To trigger
this learned response, we introduced the scent-emitting
system. It comprises a wear able scent-emitting device
and a stress ball as an activator (see figure 2). When
users feel stressed, they can simply press the stress
ball to emit scent, which will trigger their memory of
breath training, directing their awareness to rhythmic
breathing.

Instruction Steps with AromaCue Toolkit

In this section, we present the processes involved in
using the AromaCue toolkit: (1) Place the breathing
training device in a place where the user feels secure
and comfortable. (2) The user chooses their favorite
essential oil and inserts it in the diffuser inside the
breathing training device. (3) Turn on the breathing
training device and follow the visual instruction of the
4-7-8 breathing technique. (4) Practice the procedure
at least twice a day and continue training for up to one
 Figure 2: The AromaCue Toolkit: Framemwork and Olfac
tory Conditioning Paradigm

4 SYSTEM IMPLEMENTATION
Key design features of the AromaCue toolkit and
prototype development are explained in this section.

The Breathing Training Device

The breathing training device contains two features:
visual display and scent. It employs visual (the
Pepper’s ghost tech nique, which is an illusion
technique used in theaters[7]) and olfactory
stimulations to create a multi-sensory training
environment. The visual aspect uses the law of
reflection to deliver an eye-catching visual experience.
For the inter face design, we visualized the 4-7-8
breathing technique as breathing circles to guide the
training. A cat was chosen as a physical background
for displaying the computer-generated breathing circle
and the screen simulated a sleeping cat, snor ing with
a speech bubble coming from its nose. We hope the
cat enhances relaxation and improves the acceptance
of the breathing training device. An aroma diffuser
placed inside the box is ready to emit scent.
At the prototype development stage, the breathing
train ing device (26.5cm[d] x 20.5cm[w] x 19.5cm[h])
met all the criteria for creating a three-dimensional
illusion (see figure3 ). Placed on the top of the device,
an iPad air 2 (24 cm [h] x 16.9 cm [w] x 0.61cm [d])
played the video clip of 4-7-8 breathing technique. The
transparent and reflective surface (the acrylic panel
with the holographic film: 20cm x 25cm) slanted at 45
degrees in the middle of the device, assisted in
267
Papers Session 4: Input/Output TEI ’20, February 9–12, 2020, Sydney, NSW, Australia

TEI ’20, February 9–12, 2020, Sydney, NSW, Australia Lin et al.

generating a virtual image on the panel. The ultrasonic dis

tance sensor, connected to an Arduino and Raspberry Pi, is
a switch operating the video content and the aroma diffuser.
The Arduino sends remote data to activate the video in the
Raspberry Pi, automatically generating the video clip and
smell when the user is sufficiently close to the device.

Figure 4: Prototype: The Wearable Sent Dispenser (with

stress ball to activate it).

technique. The second part explored whether users could

perform conscious deep breathing in response to olfactory
 Figure 3: Prototype: The Breathing Training Device

The Scent Emitting System

The scent emitting device is to remind people to
perform con scious breathing in a stressful situation
when their physical, mental, and emotional responses
start to react. People have different abilities to handle
stress; therefore, the user sce nario may vary between
each individual. When people deal with stress-related
events, clenching a fist helps to relieve tension, so we
designed a stress ball as a switch to activate the scent
dispenser. By using the device, the scent provokes
olfactory memory and emotion, which assist in
reminding users of the breath training procedure.
The prototype of the scent-emitting device was
constituted in two parts: the stress ball and a scent
emitting dispenser (see Figure 1, Figure 2 and Figure
4). A push-button was built inside stress ball to detect
external pressure from the user and a mini diffuser
was built inside the scent emitting brooch. When the
user held the stress ball tightly enough, it automatically
activated the scent to diffuse.

5 EVALUATION
This research aims to assess the user experience of
the breath training device and the scent emitting
system. The experi ment analyzed the feasibility of the
scent-based toolkit used in a real physical
environment. It was vital to demonstrate whether the
scent kit could reduce stress levels and was ca pable
of being used as a relaxation method. The olfactory
conditioning effect was analyzed in observations and
inter views. The first part of the experiment required
one week of associating a sweet orange scent with the
4-7-8 breathing 268
stimulation. The experiment took place over one week.

Study Design
In this study, we applied five research methodologies
to col lect qualitative and quantitative data in the
experiments: heart rate, heart rate variability, the
depression, anxiety and stress scale (DASS-21), a
Smiley Face Likert scale, and the user experience
questionnaire. In the experiment, we used the Wahoo
heart rate sensor recording the variability (HRV). We
apply the root mean square of successive differ
ences(RMSSD) to the R-R intervals. These data
helped us to measure whether participants are in a
more relaxed state before and after the comparative
experiment and allowed us to identify their emotional
states. DASS-21 is a clinical self assessment for users
to measure their states of stress, anxiety, and
depression. We compared results before and after the
experiment, which helped us to understand if the
AromaCue toolkit can lower the stress level of the user.
Also, the user ex perience questionnaire[21] helped us
to rate the AromaCue toolkit from usability and user
experience aspects.
 participants were not familiar with the 4-7-8 breathing
Procedure technique and had no allergic reaction to the es
Eight participants (7 females with a mean age of 24 sential oil (sweet orange). We also asked about their
and a standard deviation of 6 years) were recruited for medical condition and records. One participant was
the exper iment. All of them had an academic diagnosed with an irregular heartbeat; another had
background (students of a graduate program or anxiety issues. Before the experiment, participants
employees or researchers at a uni versity). The were asked to fill out form DASS-21.
Papers Session 4: Input/Output TEI ’20, February 9–12, 2020, Sydney, NSW, Australia

AromaCue – A Scent Toolkit to Cope with Stress Using the 4-7-8 Breathing Method TEI ’20, February 9–12, 2020, Sydney, NSW, Australia

The experiments tested the users’ experience with the

breath
training device and wearable scent-emitting system.
In a pre-test, we selected the essential oil from six dif
ferent options from a standard set of the ease aroma shop.
Participants rated them according to pleasantness and un
familiarity. The sweet orange scent was rated the best with
high pleasantness and medium unfamiliarity. 90%of the par
ticipants chose the sweet orange scent; therefore, we picked
it for the experiments for all participants. We assume that
the scent can be individualized for future use. The
experi ment arrangements were divided into two and 6 DISCUSSION AND RESULTS
conducted in accordance with, and with the approval In the comparative experiment (scent/without scent),
of, the ethics board of Keio University. the decrease in heart-rate (before minus after the
Stroop test) is significantly higher for the scent
Training. The participants were asked to practice the
condition. In other words, the heart beats per minute
breath ing technique at least twice a day for one week
decreased more for the scent con dition (M= -3.1, SD =
with the training device at home. One training interval
3.7) compared to the control condition (slight increase,
lasted for two to three minutes [29]. To ensure that the
M= 3.28 STD=3.7) (t = 2.62,d f = 7,p =
participants com pleted each training session and
used the correct breathing technique, we set a default
task-reminder on their phone. We provided the
instruction steps with the AromaCue toolkit to them,
and we were notified after they finished.
Comparative Experiment. A comparative experiment
(with or without sweet orange scent) was conducted
one week after training with the breath training device.
Heart rate variability, heart rate, and emotional change
scale [13] were measured before, during, and after the
experiment (around one minute after the Stroop test to
get back to baseline). The Stroop test was applied to
simulate stress. Tulen et al. showed that it could
induce short term stress [28]. We used a Latin square
design, alternating the scent and no-scent con ditions.
Figure 4 depicts the setup. During the comparative
experiment, participants were asked to hold the stress
ball. If they felt stressed, they can squeeze it. In scent
condition, when the user squeezed the stress ball, the
sweet orange scent would automatically emit.
After the comparative experiments, the participants
an swered an exit questionnaire and interview,
targeting their overall experience, and evaluating the
design of the Aro maCue prototype. They also
completed the DASS-21 before the training started
and again at the end of all the exper iments [3]. This
should give an indication of whether the overall
experience is useful for reducing longer-term depres
sion, anxiety, and stress indicators.
 Figure 5: Heart Rate Variability after the Stroop Test With or
Without Smell in msecs with Participant No. on the x-axis.

.010162). The result is significant at p < .05. This can

indicate a more relaxed participant.
The heart rate variability (the higher, the better)
increased during Stroop for six participants who used
sweet orange scent, versus only three in the control
condition (without the smell), as seen in Fig. 5.
However, the change is not statistically significant.
Regarding the qualitative results: the emotional
response was recorded before and after each test.
The Likert scale measures from negatively excited (-2),
to negatively mellow (- 1), to neutral (0), to positively
mellow (1) to positively excited (2). Following the use
of the breathing training device,75% of the participants
reported that their mood had improved.
We also used the Smiley Face Likert scale to detect
emo tional changes. 50% (4) of participants reported
that the Stroop test does not influence their mood.
The same pro portion of participants rated a negative
change after the Stroop test (without scent); However,
the second test indi cated that the sweet orange scent
improved their mood, with all of the participants
reporting positive emotional changes.
To evaluate the severity of stress, depression, and
anxiety over the whole experimental time, we used the
DASS-21. Participants filled it out before starting the
training and at the end of all experiments. The
DASS-21 gave us an indication that our overall design
worked, lowering indicators for stress, depression, and
anxiety.
The results for all three (perceived long-term stress,
de pression and anxiety) are statistically significant
using a paired T-test. All indicators decreased over
the design in tervention (see Fig. 678). We saw a
decrease in depression indicators from before (M=10
points STD=7.4) to after (M= 3.11 STD=1.90237946)
(t = 2.21,d f = 7,p = .002), the same is true for the
anxiety indicators before (M=8 points, STD4.25) and
after (M=2.4, STD = 2.71) (t = 2.88,d f = 7,p = .005),
as well for the stress indicators of the DASS21 Scale
before (M=11.125, STD = 6.428571429) and after
(M=2.799872446, STD=3.823486317) 2.89 (t = 2.89,d
f = 7,p = .005). As ex pected, the intervention had big
269 impact on perceived stress levels.
Papers Session 4: Input/Output TEI ’20, February 9–12, 2020, Sydney, NSW, Australia

TEI ’20, February 9–12, 2020, Sydney, NSW, Australia Lin et al.

analyze pragmatic quality and hedonic quality. The mean

value per item scored greater than 0.8, indicating positive
feedback from participants. The final results of pragmatic
quality (1.656) and hedonic quality (1.438) are the overall
perceptions (1.547) that the participants scored for the
design.
However, due to the small sample size, it is hard to
generalize.
 All participants reported that the AromaCue toolkit is
useful in helping them to adopt a better breathing habit and
be more conscious about deep breathing.

7 INTERVIEW
Interviews were conducted after the experiments. Trained
participants reported their overall view of the AromaCue
Figure 6: Results for the Depression Scale for the individ ual
participants (x-axis) before and after all experiments
(DASS21).

Figure 7: Results for the Anxiety Scale for the individual par
ticipants (x-axis) before and after all experiments (DASS21).

Figure 8: Results for the Stress Scale for the individual par
ticipants (x-axis) before and after all experiments (DASS21).

The user experience questionnaire [21] provides an

overall perception of the design. The questionnaire
helped us to

270
toolkit. Before the experiments, only one participant
was aware of the benefit of breathing exercises, and
none of them had trained before. This experience
helped them to under stand a more simplified
approach to relieving stress, and they believed they
would use it to cope with stress. In terms of the
design, most participants liked the creation of a multi
stimuli training environment; however, some reported
that they needed to pay much more attention than they
expected to keep up with the visual guidance. Half of
 the participants requested the use of haptic output in helped them to stay more focused and relaxed during
the breath training device, which could promote their presentations. This case for use was unexpected
relaxation further. Most partic ipants thought since participants were not asked to use the AromaCue
multi-sensory stimulation could enrich their learning toolkit: we were informed after ward. They reported
experience. After using the scent-emitting system, that they felt stressed before the final presentation,
seven participants reported that they squeezed the and their previous breath training reminded them of a
stress ball and took deep breaths during the Stroop strategy to reduce stress. One participant thought
test (with smell). Even though they did not strictly people always overthink when they know they are
follow the 4-7-8 breathing technique, they took deep about to encounter stress, and she believed most
breaths in response to the sweet orange scent. They people might use the toolkit in that scenario.
felt more relaxed in the Stroop test with scent.
9 LIMITATIONS
8 USE CASE: FINAL MASTER’S PRESENTATION Six Our study had two limitations: small sample size and a
of the participants were due to give final presentations gender imbalance in the participants. The small
of their master’s level research, and they used the sample size of the study may have affected the
learned method before and during the presentation to interpretation of the p-value
calm them selves. They all reported that the training
Papers Session 4: Input/Output TEI ’20, February 9–12, 2020, Sydney, NSW, Australia AromaCue – A Scent Toolkit to Cope with

Stress Using the 4-7-8 Breathing Method TEI ’20, February 9–12, 2020, Sydney, NSW, Australia

and confidence intervals. The P-value of our results performance. In Proceedings of the 2017 CHI conference on
human factors in computing systems. ACM, 28–34.
showed a statistical significance; however, due to the
[2] Judith Amores, Robert Richer, Nan Zhao, Pattie Maes, and
small sample size, the assumption of our hypothesis, Bjoern M Eskofier. 2018. Promoting relaxation using virtual
which is that stress levels would be lowered, maybe reality, olfactory interfaces and wearable EEG. In 2018 IEEE
less credible. Our study also relies heavily on user 15th international conference on wearable and implantable
experience with the AromaCue toolkit. Interviews and body sensor networks (BSN). IEEE, 98–101.
[3] Ilse N Beaufort, Gerdien H De Weert-Van Oene, Victor AJ
the final master presentation of user cases confirmed
Buwalda, J Rob J de Leeuw, and Anna E Goudriaan. 2017.
the possibility of using a learned technique in The Depression, Anxiety and Stress Scale (DASS-21) as a
response to stress. Screener for Depression in Substance Use Disorder Inpatients:
A Pilot Study. European addiction research 23, 5 (2017),
260–268.
10 CONCLUSIONS AND FUTURE WORK
[4] Herbert Benson and Miriam Z Klipper. 1975. The relaxation
AromaCue is a scent kit for cultivating the habit of response. Morrow New York.
deep breathing. It comprises two parts: the breath [5] Stephen Brewster, David McGookin, and Christopher Miller.
training device and the scent-emitting system. The 2006. Ol foto: designing a smell-based interaction. In
Proceedings of the SIGCHI conference on Human Factors in
concept development is based on the olfactory computing systems. ACM, 653–662.
conditioning theory. Though the repetitive pairing of [6] S. Chu. 2007. Olfactory Conditioning of Positive Performance
smell with the 4-7-8 breathing technique using the in Humans. Chemical Senses 33, 1 (Sep 2007), 65âĂŞ71.
breath training device, individuals can adopt a use ful https://doi.org/ 10.1093/chemse/bjm063
skill for coping with work-related stress. The wearable [7] Ashley Crowder, Benjamin Conway, and Troy P Senkiewicz.
2016. Enhanced pepper’s ghost illusion using video panel
scent-emitting system reminds users to breathe deeply displays. US Patent App. 14/939,757.
in a real scenario. One week of user study revealed
significant improvement in DASS-21 scores. The use
of the scent system during final research presentations
provided positive feed back from participants.
Therefore, we believe the AromaCue toolkit can
positively enable users to deal with stress. For the
future design, we will compare a haptic simulation with
a visual aid to distinguish the differences in user
experience.

ACKNOWLEDGMENTS
This work is also partly supported by JST CREST
Grant No. JPMJCR16E1.

REFERENCES
[1] Judith Amores and Pattie Maes. 2017. Essence: Olfactory
interfaces for unconscious influence of mood and cognitive
 Mei-Kei Lai. 2007. Smell me: Engaging with an interactive
olfactory game. In People and Computers XXâĂŤEngage.
Springer, 25–40.
[9] Dmitrijs Dmitrenko, Emanuela Maggioni, and Marianna Obrist.
2018. I smell trouble: using multiple scents to convey
driving-relevant in formation. In Proceedings of the 2018 on
International Conference on Multimodal Interaction. ACM,
234–238.
[10] David Dobbelstein, Steffen Herrdum, and Enrico Rukzio. 2017.
in Scent: A wearable olfactory display as an amplification for
mobile notifications. In Proceedings of the 2017 ACM
International Symposium on Wearable Computers. ACM,
130–137.
[11] Jenna Fletcher. 2019. How to use 4-7-8 breathing for anxiety.
https:
//www.medicalnewstoday.com/articles/324417.php#how-to-do-it [12]
Jérémy Frey, May Grabli, Ronit Slyper, and Jessica R Cauchard.
2018. Breeze: Sharing Biofeedback Through Wearable
Technologies. In Pro ceedings of the 2018 CHI Conference on
Human Factors in Computing Systems. ACM, 645.
[13] Lynne Hall, Colette Hume, and Sarah Tazzyman. 2016. Five
degrees of happiness: Effective smiley face likert scales for
evaluating with children. In Proceedings of the The 15th
International Conference on Interaction Design and Children.
ACM, 311–321.
[14] Keisuke Hasegawa, Liwei Qiu, and Hiroyuki Shinoda. 2018.
Midair ultrasound fragrance rendering. IEEE transactions on
visualization and computer graphics 24, 4 (2018), 1477–1485.
[15] Niels Henze and Kai Kunze. 2017. A Dagstuhl Seminar Looks
beyond Virtual and Augmented Reality. IEEE MultiMedia 24, 2
(2017), 14–17. [16] Ravinder Jerath, Molly W Crawford, Vernon A
Barnes, and Kyler Harden. 2015. Self-regulation of breathing as a
primary treatment for anxiety. Applied psychophysiology and
biofeedback 40, 2 (2015), 107–115.
[17] JJ Kaye. 2000. Smell as media.
http://alumni.media.mit.edu/~jofish/ writing/smellasmedia35.pdf
[18] Joseph Jofish Kaye. 2004. Making Scents: aromatic output for
HCI. interactions 11, 1 (2004), 48–61.
[19] Joseph Nathaniel Kaye. 2001. Symbolic olfactory display. Ph.D.
Disser tation. Massachusetts Institute of Technology.
[20] Mei-Kei Lai and Yan Yan Cao. 2019. Designing Interactive
Olfactory Experience in Real Context and Applications. In
Proceedings of the Thir teenth International Conference on
Tangible, Embedded, and Embodied Interaction. ACM,
703–706.
[21] Bettina Laugwitz, Theo Held, and Martin Schrepp. 2008.
Construction and evaluation of a user experience questionnaire. In
Symposium of the Austrian HCI and Usability Engineering Group.
Springer, 63–76.
[22] Emanuela Maggioni, Robert Cobden, Dmitrijs Dmitrenko, and
Mari anna Obrist. 2018. Smell-O-Message: integration of
olfactory notifi cations into a messaging application to improve
users’ performance. In Proceedings of the 2018 on
International Conference on Multimodal Interaction. ACM,
45–54.
[23] Marianna Obrist, Elia Gatti, Emanuela Maggioni, Chi Thanh Vi,
and Carlos Velasco. 2017. Multisensory experiences in HCI.
IEEE MultiMe dia 24, 2 (2017), 9–13.
[24] S Patibanda. 2017. Understanding the design of breathing
exercise games. (2017).
[25] Charlie Pinder, Jo Vermeulen, Benjamin R Cowan, and Russell
Beale. 2018. Digital behaviour change interventions to break
and form habits. ACM Transactions on Computer-Human
Interaction (TOCHI) 25, 3 (2018), 15.
[26] The revolutionary. [n. d.]. SMELL MEMORY KIT. http://
smellmemorykit.supersense.com/
[27] Bob Stahl and Elisha Goldstein. 2019. A mindfulness-based
271
stress re duction workbook. New Harbinger Publications.
[8] Stephen Boyd Davis, Gordon Davies, Ruba Haddad, and
Papers Session 4: Input/Output TEI ’20, February 9–12, 2020, Sydney, NSW, Australia TEI ’20, February 9–12, 2020, Sydney,
NSW, Australia Lin et al.

[28] JHM Tulen, P Moleman, HG Van Steenis, and F Boomsma.

1989. Char acterization of stress reactions to the Stroop Color
Word Test. Phar macology Biochemistry and Behavior 32, 1
(1989), 9–15.
[29] Liza Varvogli and Christina Darviri. 2011. Stress management
tech niques: evidence-based procedures that reduce stress
and promote health. Health science journal 5, 2 (2011), 74.

272
[30] Yasuyuki Yanagida, Shinjiro Kawato, Haruo Noma, Akira
Tomono, and N Tesutani. 2004. Projection based olfactory
 display with nose tracking. In IEEE Virtual Reality 2004. IEEE,
43–50.
[31] Dinghui Yu, Artem Ponomarev, and Ronald L Davis. 2004.
Altered representation of the spatial code for odors after
olfactory classical con ditioning: memory trace formation by
synaptic recruitment. Neuron 42, 3 (2004), 437–449.