SWAN: Designing a Companion Spoon for Mindful Eating

Full text

SWAN:
Designing a Companion Spoon for Mindful Eating
Jung-Ying (Lois) Yi
RMIT University
124 latrobe st
Melbourne, 3000 Australia
lois.jyi@gmail.com
Deepti Aggarwal
RMIT University
124 latrobe st
Melbourne 3000 Australia
deepti.r.aggarwal@gmail.com
Rohit Ashok Khot
RMIT University
124 latrobe st
Melbourne, 3000 Australia
rohitashok.khot@rmit.edu.au
ABSTRACT
In this pictorial, we unfold and reect on the design process
behind the creation of a research product - SWAN. SWAN
is an augmented spoon that encourages people to pay more
aention to their food and urges them to eat mindfully.
With SWAN, our aim is to address the increasing tensions
between the lucrative appeal of screen-based media and
ideologies of mindful eating. We present a descriptive
account on how we brought SWAN into being. In aending
to key design decisions across our design process, we
unveil ideas and challenges in creating a domestic research
product to support everyday mindful eating.
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 prot or commercial advantage and that copies
bear this notice and the full citation on the rst page. Copyrights for com-
ponents of this work owned by others than the author(s) must be honored.
Abstracting with credit is permied. To copy otherwise, or republish, to
post on servers or to redistribute to lists, requires prior specic permission
and/or a fee. Request permissions from Permissions@acm.org.
TEI ‘20, February 9–12, 2020, Sydney, NSW, Australia
© 2020 Copyright is held by the owner/author(s). Publication rights
licensed to ACM.
ACM 978-1-4503-6107-1/20/02…$15.00
hps://doi.org/10.1145/3374920.3375009
Authors Keywords
Mindful Eating; Human-Food Interaction; Playful Cutlery
// Figure 1: SWAN is a playful provocation on how to be mindful when dining in front of the screen.
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
743

t = 2-5 mins
Stage 1: Reect
t = 5-8 mins
Stage 2: Drop
t >= 8 mins
Stage 3: Refuse
SWAN (n)
/swɒn/
Spoon With an Added Nudge
//Figure 2-a: e rst move is Redirect, a motion that draws
diners’ aention towards food through rapid short movements.
is motion is triggered if the diner gazes continuously on
screen for 2 to 5 minutes.
//Figure 2-b: e second move is Drop, a motion that drops the
food from the spoon if the diner is occupied with the screen for
a duration of 5 to 8 minutes.
//Figure 2-c: e nal move is Refuse, where the spoon leans
back completely and refuses to pick up the food again. e
diner then has to manually reset the SWAN by pressing a
buon, which is located on the beak of the spoon.
SWAN is an augmented spoon that encourages people
to pay more aention to their food and urges them to eat
mindfully. SWAN has two parts: (1) A soware interface that
tracks the diner’s gaze on screen during mealtime, and (2) A
spoon that behaves dierently based on the duration of the
diner’s gaze on screen, and makes three moves to bring their
aention back to the food. e three moves are explained in
the gure below.
e inspiration behind SWAN came from Rebaudengo’s
[27] work on ‘addicted products’ that questions the model of
ownership of the product and treats products as companions
with their own behavior, motives, and agency. For example,
Brad the toaster is an example of one such product. Brad
is a part of real ctional experimental service by Simone
Rebaudengo and Haque Design Research, where toasters
could not be owned or bought, but only requested and hosted.
ese toasters are connected to the internet, and to other
toasters like them. ey loved being used, otherwise they
demand aention by playing pranks, throwing tantrums,
and expressing their sadness loudly on Twier. Eventually,
they move onto more caring host if kept unused for long.
t = amount of time spent gazing
screen during mealtime.
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
744

Eating while watching TV.
Similar to Brad, SWAN cannot be owned or bought, rather he
needs to be hosted. But like Brad, he loves being pampered,
especially while dining. He loves being used and being looked
at. Or else, he will throw tantrums like dropping food and
refusing to pick the food from the plate. SWAN is not actively
connected to the Internet or to other buddies like him. But if
he is kept unnoticed for long and for multiple dining sessions,
SWAN eventually looks for another more mindful host by
announcing his availability on the internet.
SWAN is a research product [24] that aims to support a
research inquiry on supporting mindful eating amidst screen-
based distractions. Instead of denying the presence of media
during eating, it aims to recongure and repurpose their use
towards instilling mindfulness in eating. In recent years, the
interest in designing research products has grown within the
HCI and TEI community [4,18,24], yet the documentation of
the complex processes involved in designing such products
has been relatively sparse [6,23]. In this pictorial, we respond
to this call by oering a descriptive account of the design
and the making of the SWAN. We use photographs as well
as illustrations to illustrate key stages of the design process,
enumerating the challenges and undertaken strategies to
resolve them. With this work, our aim is to contribute to a
mindful relationship with food.
Eating while laptop in front. Eating while scrolling on phone.
Screen-based media like television, computers, social
media, video games and Internet-based streaming
services (e.g. Netix) is a huge part of our day-to-
day life. ey occupy most of our time [31] and
oen take precedence over the rudimentary tasks
such as eating. As such, fewer of us can have our
meal without having our eyes glued to some form of
screen-based entertainment. However, such behavior
is prone to cause detrimental eects on physical and
social wellbeing in the long run. Several studies point
out that eating while watching television [2,11] and
other forms of screen-based media [16,26] is bad for
our digestive health, as it interrupts the physiological
signals of satiety and hunger. As a result, people nd
it dicult to know when they are hungry or full and
in consequence, they oen overeat, which over time
manifests into much bigger problems such as obesity
and heart diseases [11].
A solution to this problem lies in enabling a beer
understanding of how one eats and in encouraging
people to eat mindfully. Mindful eating, according to
Fung and colleagues “generally refers to the application
of mindfulness techniques to eating, which involves
nonjudgmental awareness of internal and external cues
inuencing the desire to eat, food choice, quantity of
consumption, and the manner in which food is consumed”
[12]. Years of research have shown that being aware
of physiological signals of hunger and satiety, eating
slowly, avoiding distractions while eating and paying
aention to the food are useful mechanisms to
regulate healthy eating behaviors [17,30,32]. However,
practicing mindful eating is signicantly challenging
amidst all screen-based distractions that can allure
one’s mind.
Prior research has tried to moderate the mealtime use
of screen-based media through parenting and self-
control, but these approaches have met with limited
success [5,14,20]. Others have tried to educate people
about mindful eating through in-person sessions [17]
and with smartphone based food journaling apps
[7]; However, these solutions are tedious to maintain
and they do not integrate well with everyday eating
practice. Increasingly, it appears that individuals are
not going to stop using screen-based media while they
eat [19].
RESEARCH CONTEXT: WHY EATING MINDFULLY IS HARD?
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
745

THE THINKING BEHIND SWAN
e nal design of SWAN was the result of many
explorations, developments and printing trials that
spanned over 8 months. We held regular group discussions
amongst us as well as with experts of dierent academic
backgrounds from a local research community to help
us in rening our design choices and to gather diverse
perspectives on the relationship between food and screen
time. Here we discuss some of the key learnings in the
process and we start by discussing the key literature that
inuenced our design choice.
Existing research highlights that food-related decisions are
made instantaneously and are aected by the social and
emotional context and surroundings [28]. By playfully
changing the way in which people understand their
eating habits and its consequence, there is an opportunity
to bring change in those momentary decisions. Previous
studies around habit formation [8] further indicate that
continuous and real-time exposure to triggering cues of
eating behaviors can help to gradually eliminate unhealthy
behavior through ‘dishabituation’. e challenge however,
lies in making sure that these nudges are subtle and do
not appear as patronising. To address this, we leaned on
to the principles of ludic design [13] and explored the use
of curiosity, surprise and companionship to oer new and
richer perspectives on individual’s relationship with eating
and food.
We looked into dierent modalities to nudge individuals
towards mindful eating. We started by exploring the use of
visual and auditory modality but found that their use would
interrupt the pleasures of viewing screen-based media.
Gustatory and olfactory modalities were not considered
for technical reasons as well as their potential to interrupt
the dining pleasures. Finally, we opted for haptic modality,
as haptic feedback is more subtle in comparison to other
modalities [29], making it more appropriate for dining
seing.
We next considered the key actors of a mealtime experience
that include food, the diner, cutlery and screens and
discussed their potentially use as a medium to nudge. Since
our aim was to bring diner’s aention to their food, we
chose to use cutlery. All other mediums except food would
have taken the diner’s aention away from the food. e
use of cutlery also ed within the idea of playful dining
companionship.
e idea of using technological artifacts as companion to
dining however, is not new. Existing work has looked into
the use of video conferencing [3,33], smartphones [9,25],
stued toys [10], robotics [15,21] and interactive tables
[22] to support and nurture commensal dining experiences.
However, instead of bringing new technologies to the dining
table, our intention was to recongure and repurpose an
existing dining seing and use it to nurture mindfulness
towards eating and food. As a result, we leaned on the idea
of using a cutlery as a companion with a specic goal of
mindful nudging. Since a spoon is the most commonly
used cutlery in dining, the SWAN was built as a spoon.
Furthermore, given the mass appeal of immersive media, it
is not far that technology like Virtual Reality (VR) will also
penetrate into our mealtime and people will start eating
while they are immersed in a VR environment. In summary,
innovative solutions are needed that acknowledge the
pervasive presence of screen-based media rather than denying
it when supporting mindfulness in everyday dining. SWAN
is a playful provocation in this direction. It redirects their
aention on the act of eating from time to time, prompting
mindfulness in their eating behavior.
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
746

EXPLORING THE MECHANISM AND MANNERISM OF SWAN
While designing a dining companion in the form of a
spoon, the key challenge was to identify the right balance
between contextual appropriateness and provocation.
Auger [1] noted that achieving this balance is important
because, “If a design proposal is too familiar, it is easily
assimilated into the normative progression of products and
would pass unnoticed, but also that going too far can lead
to ‘revulsion or outright shock’.” Drawing on this, within
the dining context, the design of SWAN needed to have
this balance. e form and the functionality of SWAN
should not look and feel too familiar that its presence
would go unnoticed nor it should have been outrageously
provocative that it would completely disrupt the existing
arrangement and engagement. We, therefore, leaned on
the idea of treating SWAN as a dining partner. As a dining
partner, SWAN demands aention and does not like to be
ignored. For instance, the cutlery gets increasingly upset
if enough aention is not paid to it. e cutlery then
shows its emotions by dropping the food or by refusing
to pick it up.
We explored dierent mechanisms (see the sketches
below) to demonstrate spoon’s actions. Initial ideas were
to rotate or ip the spoon (gure 3.b) or split the spoon in
half so that it drops the food (gure 3.c). However, these
options were discarded aer initial trials. For instance, the
idea of spliing the spoon was discarded because during
the trials we found that food particles can get stuck within
the split region and could aect its normal operation.
On the other hand, rotating the spoon although was a
plausible idea but was limiting in other ways. For example,
when we tried rotating the spoon using a servo motor but
the operation was a lile noisy. We rejected this option
because any loud vibrating or auditory component in the
design could have interfered with the natural course of
the meal. We also tried using stepper motors to rotate
the spoon as they are less noisy in comparison to servos.
However, the size and weight of the motor made the form
bulky; It also created the issue of overheating the spoon,
hence we had to discard the idea of rotating a spoon.
We nally decided on the tilting mechanism (gure 3.a)
that uses a servo motor to tilt the spoon in downward
direction. Although we did encounter the issue of motor
noise in the beginning, it was not as sharp and noticeable
and we were able to dampen it further using voltage
regulators. Using the tilt mechanism, we introduced three
modes of operations for the spoon: redirect, drop and
refuse. ese modes are dened by the duration of screen
gaze of the diner. ese modes were added in order to
balance the provocation with contextual appropriateness
and for giving users the opportunity to rectify the behavior
before it is too late. So depending on the gaze duration,
SWAN can decide whether to drop the food (mode 2), or
simply hint the diner (mode 1) to pay aention to it (and
the food).
// Figure 3.a) Tilting: In this mechanism, a servo motor is used to
tilt the spoon in downward direction.
Figure 3.b) Rotating: In this mechanism, a servo or stepper motor
was used to rotate or ip the spoon.
Figure 3.c) Splitting: In this mechanism, the spoon splits into two
halves which can be dynamically controlled through a push-pull
solenoid or a servo motor.
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
747

FROM DESIGN TO IMPLEMENTATION
Aer deciding the tilting mechanism, the next crucial step
was the selection of the motor. e selection was inuenced
by the size, required power and motors’ ability to pull (food)
weight without causing jier or audible noise. We considered
three motors: 1) 1.5g Goteck Linear Servo, 2) Small Reduction
Stepper Motor - 5VDC 32-Step 1/16 Gearing, and 3) Micro
Servo 9g.
In the rst trial, we used 1.5g Goteck Linear Servo for enabling
the tilt mechanism. We placed the servo to the boom of the
lower casing of the spoon handle, with a metal string hooked
on its wing and a ring on its spoon head. We found that with
such a seing the downward tilt was successful but the force
was not enough to drag the spoon head back to its original
position. As a remedy, we added another Linear Servo inside
the spoon handle. However, we found that two Linear
Servos did not work well in tandem and suered from a
lagging issue. Furthermore, operating with two linear servos
produces more noise, which we wanted to avoid. It led us to
look for a dierent motor. In the next trial, we tried replacing
the Linear Servo with a Stepper motor, but the use of Stepper
motor made the overall structure bulky and inconvenient to
operate. We wanted SWAN spoon comfortable to grip, hence
aimed at keeping its handle length and width optimum for
easy dining. Hence in the nal trial, we switched to using
a small 9g Servo motor. To further assist in the smooth
operation, we added a U shaped holder for the spoon that
made it easier for the servo to hold spoon’s weight and then
to tilt the spoon in both directions (upwards and downwards).
// Figure 4: We explored the suitability
of servos and stepper motors to enable
the silent tilting mechanism while
bearing weight.
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
748

CHOICE OF THE SPOON MATERIAL AND 3D MODELLING
While designing the spoon, we aimed at keeping its form, i.e., shape,
size and weight, as similar as possible to the standard spoons to
ensure its ecological and social validity. We therefore, took special
considerations in selecting the material for designing the SWAN
spoon. We considered light-weight and so-textured material to
support easy dining with SWAN spoon. Our choice of material was
also informed by our access to the resources. For example, we had
access to a desktop 3D printer (Ultimaker 3), which works with
PLA and ABS plastic.
However, we had to be careful about using plastic because SWAN
being a cutlery, was deemed to come in direct contact with food.
Hence, choosing a food safe material was also crucial for us. On the
other hand, the design should also hold the electronics and baery
tightly, making sure that it does not get in contact with water. To
fulll both the aims, we looked at developing SWAN spoon in two
parts: rst is the spoon head that touches the food and the other
is the spoon handle that holds the electronics. Having the spoon
head as a separate unit oered us other benets as well, e.g., it
allowed using dierent size spoons as well as reconguration of
SWAN into other cutleries like a fork or knife.
Spoon head should be replaceable to support reusability and
easy cleaning, whereas spoon handle should be light-weight and
of optimum size to support easy dining. For the spoon head, we
found o-the-shelf cutlery kit in a Japanese store, DAISO that was
made up of BPA free plastic material. We decided to use this kit as
it allowed quick prototyping for spoon head. On the other hand,
there was no ready to use solution for spoon handle, as its shape
and size varied with the use of dierent electronic units. Hence, we
3D printed the spoon handle by using a biodegradable PLA plastic
material to get the desired shape and size.
We also designed a hook structure to join the upper and lower
casing of the spoon handle. A square platform was made on the
inner surface of both upper and lower casing for locking the servo
motor at its position and to prevent it from moving accidentally.
To clip the spoon head into spoon handle, we designed a U-shaped
holder for the spoon head that can be screwed to the servo motor
present in the spoon handle. is overall assembly made SWAN
handle slightly wider and bulkier in comparison to a traditional
spoon but in the initial trials, it was found to be comfortable to
hold and grip.
e spoon head is aached
to the servo using the U
shaped holder.
e upper and lower
casing can be separated
to allow any maintenance
if required.
Square platform is
designed to secure servo
motor in place.
U-shaped holder is xed to
servo using a screw.
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
749

Gaze is captured using an ESP-EYE
camera board.
Gaze is captured using a front camera
on laptop or external webcam.
Identifying diner’s aention towards food was the crucial
aspect of the SWAN design. We initially planned to use the
ESP-EYE camera to detect diner’s gaze on food, because of
its tiny size and reasonably good accuracy in detecting facial
motion. Initially, our plan was to mount the camera on the
spoon itself. However, the initial trials with three users (1
Male, 2 Female) revealed that the camera was very slow (2
frames per second) in detecting the facial motions. We could
not nd a beer alternative to this camera that is tiny enough
to aach on spoon; Hence we switched our focus from using
spoon to using the screen, although doing this meant that
the use of SWAN is limited to screen-based dining context,
which we were okay with. We designed a web application
in Javascript using machine learning and Beyond Reality
Face (BRF v4) library, that tracks user’s gaze on screen by
using the webcam of a laptop. We used the duration of the
diner’s screen gaze as an indicator of their aention to the
food. Although it is a crude way of understanding aention,
as we can also focus on other parts of our surroundings while
eating, it helped us in simplifying the gaze detection.
Since there is no dened guidelines on how much aention
one should give to screen and to the food in such a context,
we manually congured three modes of screen gaze and
tailored SWAN use in relation to them. e rst mode redirect
has a screen gaze duration of 2-5 minutes, we thought this
duration is ideal as a rst warning for the diner. e SWAN
in response only makes small movements but he does not
drop the food. e second mode drop has the duration of
5-8 minutes, and this mode is treated as a second warning,
which will instruct SWAN to drop the food. e nal mode
refuse occurs when diner gazes continuously at the screen for
longer than 8 minutes, resulting in third and nal warning
where SWAN tilts all the way back and picking food from it
requires a manual reset.
GAZE DETECTION
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
750

TRIGGERING SWAN IN RESPONSE
TO THE GAZE
Aer nalising the gaze detection, the next
challenge was to initiate the SWAN motions in
response to the gaze. It was important to nd the
initial state of the spoon before triggering its three
modes, i.e., Redirect, Drop, and Refuse. For instance,
a motion should not be triggered when the spoon
is resting in the bowl or when SWAN is closer
to the diner’s mouth. e motion should only be
triggered when the SWAN is in between these two
states, that is, it has picked up the food but the food
has not yet reached the mouth.
In order to identify the position of SWAN with
respect to the dining table, we initially thought of
using a weight sensor that would identify whether
SWAN has picked up the food or not. However
adding a weight sensor was making the design
bulkier and weight alone was not sucient to
indicate SWAN’s current state, as the spoon can
also have food while resting in the bowl. Next, we
considered the use of barometer to measure the
elevation of the SWAN with respect to its resting
position in the plate or on the dining table. e
initial trials with the barometer however, revealed
its dependencies on the height of the table and
diner’s body postures like leaning forward.
Finally, we decided to use an accelerometer. We
did lab testing with dierent people to identify the
coordinates of X, Y and Z axis when the spoon is
resting and when it is closer to the mouth. Even
though, the accelerometer was not the perfect
solution to the problem, it gave reasonably good
results in the lab trials, hence we incorporated it in
the nal design.
Accelerometer values were captured to identify SWAN’s positioning:
resting, eating or moving.
e use of accelerometer was tested along with
other electronic components
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
751

POWER MANAGEMENT
e 9g Servo motor used in the spoon handle needed power
to work. Also, to allow the SWAN spoon to be used in dining
seing, it was essential to make the spoon wire free and use
baeries instead. Initially, we used a Lithium-Ion 3.7V baery
to power the motor, however, the motor started to give a
buzzing sound. is issue was not present while charging the
motor using the USB port. We tried using ltering capacitors
to suppress the motor noise, however, the impact was too
minor to address the problem. We then tried a 5V voltage
regulator to enhance the power voltage and the noise problem
was solved.
As we saw, the low power voltage can aect the performance
of the servo motor. Hence, it was essential to inform the user
when the baery level is low to signal that it needs to be
charged. Although initially we did not plan to oer a visual
feedback, we had to include LED indicator on the spoon
handle to indicate its current baery status.
We determined the baery level from the spare ADC pin
available on the ESP 32 Arduino board. e ADC pin has a
built-in voltage divider through which the voltage can be
calculated using a mathematical formula. e calculated
voltage value is closer to the readings of the voltmeter.
Adding the LED on the spoon in fact, guided its name as
SWAN. For instance, the LED looked like a bird’s eye. So,
we changed the design of the handle to make it look like a
bird, and then named it SWAN. e LED light not just signals
the need to recharge the baery but metaphorically, it also
expresses the emotions of SWAN (the bird).
Pololu 5V
Step-Up Voltage
Regulator
U1V10F5
e LED colour had changed to red to indicate the
voltage is under 3.55V. It is close to the volume read
by the multimeter.
e meanings of the LED colours
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
752

FINAL ASSEMBLY
Aer nalising the design and the interaction
mechanisms, we were tasked with bringing
SWAN into a reality. Aer going through
dierent electronic options and testing them
together, we seled on the following electronic
parts: Adafruit HUZZAH32 ESP32 Feather
microcontroller, Polymer Lithium-Ion 3.7V
370mAh, a 5V servo motor, an Accelerometer, a
NeoPixel LED, a vertical tactic buon and a 5V
voltage regulator. Our aim was to accommodate
all these electronics in the 3D spoon handle. To
make sure that these parts t well in the spoon
handle, we replicated all the electronics in a
digital format in their original size. It enabled
us to build the 3D model of SWAN with clear
dimensions and precise mechanical structure
to secure the electronics inside. Aer the 3D
model was printed, we integrated the soldered
circuit board of the spoon design into printed
3D models and tested for any printing errors.
e rst print test showed that the opening
of the casing did not align with the micro
USB charging port, and the size was too large
as well. e sleeping mode buon was also
slightly misaligned. e potential reason for
these errors was the mismatch in the digital
model and congured physical model with
soldered peroard. For example, aer all
electronics were assembled, soldered and
inserted inside the SWAN, it’s alignment was
slightly inaccurate. We rectied it in the next
printing trial by rearranging the electronics
parts. e nal design was then sent to a
commercial 3D printing service as it allowed us
to print multiple copies with beer precision.
During the rst calibration, the opening on
the casing was not aligned with the micro USB
charging port, and the size was too large. Also,
the sleeping mode buon wasn’t ed in well.
e second and the last calibration
resulted well aer modifying the 3D les
and rearrange the electronics layout to
make sure the electronics can align well
with the printed model.
e Results Of e First Calibration e Results Of e Second Calibration
Peroard
Pololu 5V
voltage
regulator
Accelerometer
3.7V
370mAh
baery
RGB LED
ESP32
9g micro
servo motor
Verticle
buon
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
753

CONCLUDING REMARKS
Modern society is inundated with dietary-related
diseases and disorders, which could largely be
aributed to the possible disconnection between
individuals and their awareness of eating habits.
Helping people to eat mindfully can nurture a
strong, healthy connection to their relationship
with food. SWAN is a step towards this direction.
SWAN is a provocative technology-driven solution
to mindfulness that acknowledges the pervasive
presence of screen-based media and repurposes the
surroundings to guide the diner’s aention towards
food. We do not position SWAN within the realm of
persuasive behavior change technology rather we
consider SWAN as a playful subversion.
Longitudinal studies of SWAN however, might unveil
interesting ideas and insights towards nurturing
a change in behavior through a companion spoon.
By presenting a descriptive account on how we
brought SWAN into being, we highlighted the key
design parameters that designers should consider
to make a smart cutlery for dining seing. Had the
spoon been big, heavy, noisy and wired the design
would not satisfy as a cutlery tool. Similarly, had the
spoon oered visual or auditory feedback, it would
interrupt with the screen-media content. Designing
for mealtimes in the presence of screen-media has its
own specicity, that SWAN thrives to embrace for its
ecological validity.
ACKNOWLEDGMENT
We thank Prof. Florian Mueller, Prof. Angelina Russo and
Exertion Games Lab team for their input on this work.
We also acknowledge the support from the Australian
Research Council DECRA Award DE190101151.
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
754

[1] James Auger. 2013. Speculative design: craing the
speculation. Digital Creativity 24, 1: 11–35. hps://
doi.org/10.1080/14626268.2013.767276
[2] Amanda Avery, Catherine Anderson, and Fiona
McCullough. 2017. Associations between chil-
dren’s diet quality and watching television during
meal or snack consumption: A systematic review.
Maternal & child nutrition 13, 4. hps://doi.
org/10.1111/mcn.12428
[3] Pollie Barden, Rob Comber, David Green, Daniel
Jackson, Cassim Ladha, Tom Bartindale, Nick
Bryan-Kinns, Tony Stockman, and Patrick Oliv-
ier. 2012. Telematic Dinner Party: Designing for
Togetherness rough Play and Performance. In
Proceedings of the Designing Interactive Sys-
tems Conference (DIS ’12), 38–47. hps://doi.
org/10.1145/2317956.2317964
[4] Andy Boucher. 2016. e Form Design of the
Datacatcher: A Research Prototype. In Proceed-
ings of the 2016 ACM Conference on Designing
Interactive Systems (DIS ’16), 595–606. hps://doi.
org/10.1145/2901790.2901907
[5] Ying-Yu Chen, Ziyue Li, Daniela Rosner, and
Alexis Hiniker. 2019. Understanding Parents’
Perspectives on Mealtime Technology. Proc. ACM
Interact. Mob. Wearable Ubiquitous Technol. 3, 1:
5:1–5:19. hps://doi.org/10.1145/3314392
[6] Peter Dalsgaard and Kim Halskov. 2012. Reective
Design Documentation. In Proceedings of the De-
signing Interactive Systems Conference (DIS ’12),
428–437. hps://doi.org/10.1145/2317956.2318020
[7] Daniel A. Epstein, Felicia Cordeiro, James Fogarty,
Gary Hsieh, and Sean A. Munson. 2016. Crumbs:
Lightweight Daily Food Challenges to Promote
Engagement and Mindfulness. In Proceedings of
the 2016 CHI Conference on Human Factors in
Computing Systems (CHI ’16), 5632–5644. hps://
doi.org/10.1145/2858036.2858044
[8] Leonard H. Epstein, Jennifer L. Temple, James
N. Roemmich, and Mark E. Bouton. 2009. Habit-
uation as a determinant of human food intake.
Psychological review 116, 2: 384–407. hps://doi.
org/10.1037/a0015074
[9] Hasan Shahid Ferdous, Bernd Ploderer, Hilary
Davis, Frank Vetere, Kenton O’Hara, Geremy
Farr-Wharton, and Rob Comber. 2016. TableTalk:
Integrating Personal Devices and Content for
Commensal Experiences at the Family Dinner Ta-
ble. In Proceedings of the 2016 ACM International
Joint Conference on Pervasive and Ubiquitous
Computing (UbiComp ’16), 132–143. hps://doi.
org/10.1145/2971648.2971715
[10] Carmen Fishwick. 2014. Table for one? Restau-
rant oers giant stued animals for company.
e Guardian. Retrieved September 9, 2018
from hp://www.theguardian.com/world/2014/
may/06/table-for-one-restaurant-giant-stued-ani-
mals-loneliness-japan
[11] Heather M. Francis, Richard J. Stevenson, Megan
J. Oaten, Mehmet K. Mahmut, and Martin R. Yeo-
mans. 2017. e Immediate and Delayed Eects of
TV: Impacts of Gender and Processed-Food Intake
History. Frontiers in psychology 8: 1616. hps://
doi.org/10.3389/fpsyg.2017.01616
[12] Teresa T. Fung, Michael W. Long, Pamela Hung,
and Lilian W. Y. Cheung. 2016. An Expanded
Model for Mindful Eating for Health Promotion
and Sustainability: Issues and Challenges for
Dietetics Practice. Journal of the Academy of Nu-
trition and Dietetics 116, 7: 1081–1086. hps://doi.
org/10.1016/j.jand.2016.03.013
[13] William W. Gaver, John Bowers, Andrew Bouch-
er, Hans Gellerson, Sarah Pennington, Albrecht
Schmidt, Anthony Steed, Nicholas Villars, and
Brendan Walker. 2004. e Dri Table: Design-
ing for Ludic Engagement. In CHI ’04 Extend-
ed Abstracts on Human Factors in Computing
Systems (CHI EA ’04), 885–900. hps://doi.
org/10.1145/985921.985947
[14] Alexis Hiniker, Sarita Y. Schoenebeck, and Julie
A. Kientz. 2016. Not at the Dinner Table: Parents’
and Children’s Perspectives on Family Technology
Rules. In Proceedings of the 19th ACM Conference
on Computer-Supported Cooperative Work &
Social Computing (CSCW ’16), 1376–1389. hps://
doi.org/10.1145/2818048.2819940
[15] Rohit Ashok Khot, Eshita Sri Arza, Harshitha Kur-
ra, and Yan Wang. 2019. FoBo: Towards Designing
a Robotic Companion for Solo Dining. In Extended
Abstracts of the 2019 CHI Conference on Human
Factors in Computing Systems (CHI EA ‘19).
ACM, Paper LBW1617, 6 pages. DOI: hps://doi.
org/10.1145/3290607.3313069
[16] Anastasia Kononova, Anna McAlister, and Hyun
Jung Oh. 2018. Screen overload: Pleasant multi-
tasking with screen devices leads to the choice of
healthful over less healthful snacks when com-
pared with unpleasant multitasking. Computers in
human behavior 80: 1–11. hps://doi.org/10.1016/j.
chb.2017.10.042
[17] Jean L. Kristeller and Ruth Q. Wolever. 2011.
Mindfulness-based eating awareness training for
treating binge eating disorder: the conceptual
foundation. Eating disorders 19, 1: 49–61. hps://
doi.org/10.1080/10640266.2011.533605
REFERENCES
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
755

[18] Henry Lin, Ron Wakkary, and Doenja Oogjes.
2019. e Tilting Bowl: Electronic Design for a Re-
search Product. In Proceedings of the 2019 on De-
signing Interactive Systems Conference (DIS ’19),
345–357. hps://doi.org/10.1145/3322276.3323701
[19] Ricky Maughan. 2018. One in three Americans
can’t eat without their phone. Retrieved October 3,
2019 from hps://nypost.com/2018/01/23/one-in-
three-americans-cant-eat-without-their-phone/
[20] Melissa Mazmanian and Simone Lanee. 2017.
Okay, One More Episode: An Ethnography of
Parenting in the Digital Age. In Proceedings of the
2017 ACM Conference on Computer Supported
Cooperative Work and Social Computing, 2273–
2286. hps://doi.org/10.1145/2998181.2998218
[21] Yash Mehta, Rohit Ashok Khot, Rakesh Patibanda,
and Florian Mueller. 2018. Arm-A-Dine: Towards
Understanding the Design of Playful Embodied
Eating Experiences. In Publication of the Annual
Symposium on Computer-Human Interaction in
Play.
[22] Robb Mitchell, Alexandra Papadimitriou, Youran
You, and Laurens Boer. 2015. Really eating togeth-
er: a kinetic table to synchronise social dining
experiences. In Proceedings of the 6th Augment-
ed Human International Conference, 173–174.
hps://doi.org/10.1145/2735711.2735822
[23] William Odom, Ishac Bertran, Garnet Hertz,
Henry Lin, Amy Yo Sue Chen, Ma Harkness, and
Ron Wakkary. 2019. Unpacking the inking and
Making Behind a Slow Technology Research Prod-
uct with Slow Game. In Proceedings of the 2019
on Creativity and Cognition (C&C ’19), 15–28.
hps://doi.org/10.1145/3325480.3326567
[24] William Odom, Ron Wakkary, Youn-Kyung
Lim, Audrey Desjardins, Bart Hengeveld, and
Richard Banks. 2016. From Research Prototype
to Research Product. In Proceedings of the 2016
CHI Conference on Human Factors in Comput-
ing Systems (CHI ’16), 2549–2561. hps://doi.
org/10.1145/2858036.2858447
[25] Kenton O’Hara, John Helmes, Abigail Sellen,
Richard Harper, Martijn ten Bhömer, and Elise
van den Hoven. 2012. Food for Talk: Phototalk in
the Context of Sharing a Meal. Human–Computer
Interaction 27, 1-2: 124–150. hps://doi.org/10.108
0/07370024.2012.656069
[26] Rose E. Oldham-Cooper, Charloe A. Hardman,
Charloe E. Nicoll, Peter J. Rogers, and Jerey M.
Brunstrom. 2011. Playing a computer game during
lunch aects fullness, memory for lunch, and later
snack intake. e American journal of clinical
nutrition 93, 2: 308–313. hps://doi.org/10.3945/
ajcn.110.004580
[27] Simone Rebaudengo. 2017. Design for Living with
Smart Products. O’Reilly Media, Inc. Retrieved
October 3, 2019 from hps://www.oreilly.com/
library/view/design-for-living/9781491997390/
[28] Jonathan van’t Riet, Siet J. Sijtsema, Hans Dage-
vos, and Gert-Jan De Bruijn. 2011. e importance
of habits in eating behaviour. An overview and
recommendations for future research. Appetite
57, 3: 585–596. hps://doi.org/10.1016/j.ap-
pet.2011.07.010
[29] Roland Sigrist, Georg Rauter, Robert Riener, and
Peter Wolf. 2013. Augmented visual, auditory,
haptic, and multimodal feedback in motor learn-
ing: a review. Psychonomic bulletin & review 20,
1: 21–53. hps://doi.org/10.3758/s13423-012-0333-
8
[30] Katy Tapper. 2018. Mindfulness and craving:
eects and mechanisms. Clinical psychology
review 59: 101–117. hps://doi.org/10.1016/j.
cpr.2017.11.003
[31] e Australian Communications and Media Au-
thority. 2017. Communications report 2017–18 |
ACMA. Retrieved May 1, 2019 from hps://www.
acma.gov.au/theACMA/communications-report
[32] Janet M. Warren, Nicola Smith, and Margaret
Ashwell. 2017. A structured literature review
on the role of mindfulness, mindful eating and
intuitive eating in changing eating behaviours: ef-
fectiveness and associated potential mechanisms.
Nutrition research reviews 30, 2: 272–283. hps://
doi.org/10.1017/S0954422417000154
[33] Jun Wei, Xuan Wang, Roshan Lalintha Peiris,
Yongsoon Choi, Xavier Roman Martinez, Remi
Tache, Jerey Tzu Kwan Valino Koh, Veronica
Halupka, and Adrian David Cheok. 2011. CoDine:
an interactive multi-sensory system for remote
dining. In Proceedings of the 13th internation-
al conference on Ubiquitous computing, 21–30.
hps://doi.org/10.1145/2030112.2030116
Pictorials Session 1: Urban Experiences, Eating, and Communication
TEI ’20, February 9–12, 2020, Sydney, NSW, Australia
756