If designers give users the means to create or experience interactive prototypes with them, users can contribute and form realistic expectations and experiences. That is especially relevant in case of novel interactive solutions and children as users. In 2017, two prototyping workshops concerning a ective objects were organised with children and designers across Europe: one in Germany in a summer camp, the other in Italy during a science festival. The former workshop helped designers inspect children's expectations for a ective objects. The latter workshop helped designers assess other children's experience with the prototypes created by the former children. The paper presents the prototyping workshops with in total 81 children, aged 11 to 16, and re ects on what designers learnt from them.
Emotion communication is mostly done non-verbally, through facial expressions,
body postures or objects that are extensions of our bodies. A ective objects
are interactive objects that enable users to communicate emotions. A ective
objects are often created by teams of adults, and experienced by children1 later
in the design process. At the same time, the design community also advocates
for a more active role of children in the early design of interactive objects, with
diverse degrees of participation [
Various techniques can be used to include children in the early design;
prototyping is one, which can be used in di erent design stages and for di erent
goals. For instance, Moser et al. [
Starting from that body of work, this paper pursues the idea to include children into the early design of a ective objects for them through prototyping. 1 A child is a person below the legal age of majority.
If designers include children into the making of prototypes, children can express their expectations for a ective objects. If designers enable children to interact with early prototypes, children can experience novel design ideas of a ective objects for them. The paper reports on two such prototyping workshops with designers and children, organised across diverse ages (from 11 to 16), cultures and countries. In one workshop, in a Germany forest, children make their prototypes and designers investigate children's expectations of a ective objects for children. In the other one, in a school in Italy, other children experience those prototypes and designers investigate their experience.
The exploratory interrelated research questions of the workshops are two: (R1) What a ective objects do children create or engage with? (R2) Can prototyping enable children to communicate their ideas of a ective objects?
The paper begins with an overview of related work and background concerning the design of interactive objects with children, especially in relation to emotion communication. Then the paper continues with its core matters. It presents the workshop in Germany, with 55 13{16 years old children from Germany, Czech Republic and non-EU countries, and the follow-up workshop in Italy, with 26 11{14 years old children from di erent cultures as well. In its conclusions, the paper re ects on the results of the workshops in relation to the above research questions. 2
Emotions play an essential role in our interpersonal relationships. Besides the
body, objects embody meanings, memories, thoughts and, more generally,
emotions [
Children aged 11{18, who are pre-teens or teens, face considerable emotional
changes [
Workshops that investigate expectations for a ective objects are not many,
especially with pre-teens or teens [
Later on, the interactive prototypes can be used to gain direct insight into
children's experience with them. Experience prototyping, in the original
acceptation of the term [
A workshop was organised with designers and children to investigate children's expectations of a ective objects for them. It lasted c. 2 hours in the evening and asked participant children to ideate, conceptualise and prototype their own a ective objects. It was conducted in a German forest during a youth summer camp, within a semi-open location with long dining tables. All children participated voluntarily and were told about the workshop goals. Their parents or referent adults approved of their participation, after being informed about the workshop goals and methods.
Group F M Nationality Group F M Nationality G1 G2 G3 G4 7 0 Mixed 0 7 Mixed 6 0 Czech 6 0 Mixed
G5 G5 G6 G7 0 8 Mixed 0 8 Mixed 0 7 Mixed 0 7 Mixed
Tables for groups were located on both sides of a central table, working as hub, with the design material, e.g., Littlebits toolkits. Another area had a theatre-like stage for the nal presentation and sharing of a ective objects, placed in front of the hub table.
The workshop involved a total of 55 children, 26 females and 29 males, aged 13{16. Children were from Czech Republic (31), Germany (18) and non-EU countries (3). Common languages were German or English. In total, 8 groups were created by the camp organisers according to camp organisation constraints, with separate female groups and male groups, but with mixed groups in terms of nationality. See Figure 1 for a group. Table 1 speci es them all.
The workshop also involved design facilitators, one per group, trained to the goal and methods of the workshop. There was also a design moderator, who usually remained at the hub table for delivering material and feedback to the group elected representative. The workshop was organised in an ideation and a conceptualisation phase, a prototyping phase and a sharing phase.
In the ideation and conceptualisation phase, paper-based generative toolkits
are often used. Tiles cards are paper-based generative IoT toolkits [
However, Tiles or similar cards, alone, may be insu cient with inexperienced
participants for conceptualising IoT objects. Thereby, designers also use further
generative toolkits for guiding children towards the prototyping phase. In [
During the workshop, rstly, children discussed about the assigned mission: creating a ective objects that communicate one's mood or emotions. Then they chose what objects to make interactive: either everyday objects, represented on adapted Tiles cards, or natural elements of the forest location, e.g., a stone. At the same time, they chose input and output cards, adapted from the Tiles primitive deck for the available Littlebits sensors (for input cards) and actuators (for output cards). Children were asked to re ect about such cards together and place them in the correct area of the conceptualisation framework.
For the prototyping phase, children moved to the hub for collecting prototyping sensors and actuators. The moderator handed the toolkits out according to children's questions and what speci ed in their conceptualisation framework.
In the sharing phase, children moved to a theatre-like stage. There each group in turn enacted their prototype to explain its ideation, in a silent-theatre performance. This was chosen because of the multi-lingual environment, so as to enable all participants to equally share their ideas and engage in the process. While one group was presenting, with the help of facilitators, the others were trying to understand the exact idea behind the object, and what emotions the object was communicating. 3.3
This section reports on the results of the workshop in Germany. The a ective object prototypes, created by children, are listed per group in Table 2 and described below. Other results are reported in relation to the two main questions of the paper, given in Section 1: concerning children's expectations for a ective objects, rstly, and the workshop organisation, secondly.
Data come from children's artefacts, i.e., their conceptualisation frameworks and prototypes, besides designers' documentation, i.e., photos, videos and independent observations. In particular, at the end of the workshop, designers compared their documentations, and organised them in relation to children's expectations for a ective objects and the workshop organisation. 25
Description of A ective Objects. The A ection Watch should count each a ectionate touch (e.g., hugging, caressing), create happiness, and display how frequently its user gets in physical contact with others. The Hungry Fork should express negative emotions due to hunger: when the button on the fork is pressed, and lights the red LED which is embedded into the user's t-shirt, the user expresses the fact that hunger stresses him/her and he/she needs food from others. The Sadness Bottle should reveal its user's sadness through a blue LED when the user hugs it and touches it; hugging, as care giving, is translated into an affective object which communicates the sadness of its user. The Cheer-up Pillow is an assistive object for coping with negative emotions: when pressed, it makes a massage by vibrating and, when the slider is used, it also blows fresh air on the face. See Figure 2. The Laughing Bracelet responds to laughs and magnies its user's positive emotional state: the loudest the user laughs, the loudest the bracelet responds. The Confusion Stone, when held in hand and squeezed, should recognise the user's state of confusion (a negative situation), and warn the user and others with a buzzing sound. When the Happiness Clip is touched, it lights up and makes a buzz sound to represent the happiness of the user. Similarly, when the Happiness Candle is put on its platform, it should measure the happiness of the user and show it with numbers on its digital display. Results Concerning A ective Objects. Children's prototypes and designers' documentation were given to three researchers: two from interaction design, one from industrial design. Researchers inspected them in relation to children's expectations for a ective objects, by considering the following analysis criteria: (1) the used physical objects and their a ordance; (2) emotions conveyed by affective objects. Researchers worked independently, after agreeing on the criteria and a common report format in two initial meetings. Researchers met again for comparing their results. Results were discussed; in case of low agreement, they were revised. The nal results are as follows.
Researchers judged if children's prototypes used only everyday objects, or also natural elements from the surrounding natural environment (forest); the a ordance of the used physical objects. Only 1 out of 8 prototypes (the Confusion Stone) used an object from nature. A ordance was judged low in the following sense: only 37; 5% prototypes (3 out of 8) used physical characteristics of the physical object. In general, according to researchers, children tended to choose objects that they were using daily and represented on cards (e.g., pillow, fork, bottle) rather than re ecting on physical objects and their a ordance.
Criteria concerning emotions were taken from the emotion design
literature [
workshop organisation come from designers' documentation. According to this, all children managed to create their prototypes in the alloted time and with the given material; designers mainly helped in the sca olding of children's ideas.
Moreover, paper-based generative toolkits helped children create realistic expectations and keep the design aligned with the assigned mission. Most impotantly, they helped children structure their ideas and hence facilitate the communication with designers. For instance, designers gave all groups feedback concerning the chosen sensors and actuators in relation to the if-then rule, inserted into the conceptualisation framework, before groups of children started prototyping their a ective objects. 4
A later workshop was organised with designers and other children in a school in Italy, on the occasion of a science event, with the main aim of investigating children's experience of the a ective object prototypes that were created in the former workshop. 4.1
The science event organisers and teachers invited children to the event, and the workshop had 26 participants, aged 11{14. There were 15 males and 11 females. Participants were selected by the science event organisers and school 27 personnel; all children participated voluntarily and were told about the workshop goals. Their parents or referent adults approved of their participation, after being informed about workshop goals and methods.
The school is multi-cultural, and 38:46% of participating children had di erent cultural backgrounds than Italian. Demographics data were collected through a survey by the science event organisers, concerning children's age and gender, besides cultural origins (if di erent than Italian or not).
Four designers were also present, besides a moderating teacher. One of the designer was the design moderator present in the former workshop in Germany. Children were divided by their teacher in groups of 5{6 children, heterogeneous for cultural backgrounds. Each group participated separately in the workshop. The workshop was split into three phases, with dedicated physical areas; a brief introduction to the prototyping of a ective objects; the sharing of the former workshop experience; the collection of children's preferences for a ective objects.
The rst phase served to introduce children to sensors and actuators for prototyping a ective objects. Each group, in turn, moved towards the related area with two designers. A designer showed children examples of simple a ective objects, and invited children to re ect on how the objects used sensors and actuators; the other designer mainly acted as observer and note-taker. This phase served to expose children to the prototyping of a ective objects.
The second phase was for sharing the results of the workshop experience in Germany. Each group in turn moved to the sharing area, with a long table for displaying the a ective objects of the former workshop. In this area, children found the moderating designer of the former workshop and another one, mainly acting as observer and note-taker. The moderating designer acted as narrator and \bridge" between the two workshops. She explained children's origin and ideation of each object, and how to interact with it. Then children were invited to touch and use it, and ask questions about it.
In the nal phase, children ranked a ective objects with a survey form. Each object was represented on a sticker. Each child chose three stickers to rank his or her three preferred objects in the survey form. This has medals for the rst position (the rst preferred object), the second position (the second preferred object) and the third position (the third preferred object). 4.3
As in the case of the workshop in Germany, results concerning the workshop in Italy are given in relation to the two research questions of Section 1: concerning rstly children's experience of a ective objects, and secondly the workshop organisation.
Data come from children's survey rankings, besides designers' documentation| photos, videos and independent observations. Like in the rst workshop, designers compared their documentations, and organised them in relation to children's experience with a ective objects and the workshop organisation.
objects were analysed by gender. The Happiness Pillow was ranked rst by 81% of participants. The result is independent of gender: 82% females and 80% males chose it. The A ective Watch was second for 38%: 45% were females and 33% males. Males and females di erently assigned the third place: 45% of females chose the Laughing Bracelet; 33% of males chose the Hungry Fork.
Those results are backed up by designers' documentation, concerning the behaviour of children with objects or their verbatim comments. According to that, the Happiness Pillow was very engaging, e.g., children kept on squeezing it. The A ection Watch created physical contacts among participants (such as holding hands). This seems to have engaged more females than males. In turns, males seemed to be more interested in the functioning of objects, e.g., they questioned the moderating designer about the employed sensors or actuators. Results Concerning the Workshop Organisation. According to designers' documentation, children generally grabbed how to interact with protypes, after being exposed to their prototyping int the workshop introduction phase; during the sharing phase, little explanation was needed by the moderating designer, who mainly reported the generation of the objects and what they were meant for. The fact that the designer was previously exposed to children's ideas enabled her to quickly interact with children about prototypes.
The interactivity of prototypes also engaged children into experiencing them, e.g., children asked to interact with all prototypes. While interacting with the prototypes, some children spontaneosly voiced their opinions about them, e.g., they criticised some prototypes, or suggested novel application scenarios. For instance, the Sadness Bottle was criticised because of the chosen physical object (its material and shape); a group of participants came up with an idea of a soft object that could be hugged, whereas \a glass bottle cannot". While trying out 29 the Confusion Stone, another group conceived a novel usage: the stone could help children express their stress or fear and ask for the help from others \very far away". 5
This paper presented two workshops for including children in the design of affective objects. Prototypes were the main communication medium within and across workshops, among children and with designers. The former workshop was held in a Germany multi-cultural summer camp and mainly served to investigate children's expectations for a ective objects. The latter workshop was in a multicultural school in Italy, on the occasion of a science event; children experienced the prototypes of the former workshop, and expressed their ideas about them.
Data were gathered and analysed for both workshops from complementary perspectives, e.g., children's and designers'. The reported results enabled us to answer the main research questions of the paper: what a ective objects do children create or engage with? Is prototyping adequate for communicating children's ideas? This section ends by re ecting on the speci c results related to the research questions: rstly, a ective objects; secondly, the prototyping of a ective objects itself, as a communication means among children, and with designers. 5.1
According to the results of the workshop in Germany, the majority of children (7 groups out of 8) used everyday physical objects, and not nature elements, as if children had grabbed \the rst available thing". Overall, such results seem to indicate that children were not constrained and yet not guided towards their choice of physical objects. Thus the workshop results do not give clear indications concerning children's expectations of physical objects for a ective objects. As far as emotions are concerned, some indications emerge from that workshop. Children realised as many a ective objects for positive emotions as for negative ones, and mainly for communicating emotions to others, independently of gender. The group with refugees made a bottle that shows others their sadness, thereby externalising and sharing their status. Only male groups made a ective objects with utilitarian values (half of them). That may be explained with the age range of participants (aged 13{16) and re ect a gender di erence.
The later experience workshop in a multi-cultural school in Italy used the affective object prototypes created by children in the former workshop. According to its results, two objects were preferred by all children: the Cheer-up Pillow was the rst choice, independently of gender; the A ective Watch was the second, preferred by females. A possible reason for children's preference for the haptic pillow was that it created a haptic sensation, which was unexpected in a pillow and hence appreciated by children. As for the A ective Watch, it was leading children to touch each other rather than creating a simulation of touch, which might be the reason why it engaged more females than males.
30
In each workshop and across workshops, prototypes were the main
communication means. According to the reported results, interactivity facilitated prototype
sharing and engagement in the former workshop. It also enabled participants in
the later workshop to rapidly experience prototypes in an ecological setting and
engage with them, in line with previous ndings albeit with adults [
The material for the workshops included ad-hoc paper-based ideation cards
and a conceptualisation framework, besides prototyping kits. The material helped
in rapidly guiding children towards their creation or experience of prototypes.
Moreover, the material facilitated children's communication with designers, and
hence it helped designers convey children's ideas across workshops. At the same
time, the material might have partly limited children's expression, e.g., in the
former workshop, children always tended to use everyday objects, represented
on cards. Children's imagination could also have been hampered by the limited
range of available sensors and actuators. In the future, an intermediate
workshop might be introduced between the two workshops with children, to continue
prototypes with more advanced micro-electronics kits and fully realise children's
expectations and experience of a ective objects, e.g., a making workshop with
university students, like in [
The presence of the same designer across workshops was also relevant to
convey the ideas behind object prototypes more completely, in line with the
results reported in [
We would like to thank the Gaisthal camp leaders, Andrea Dieck and Daniel Philipp Adlmueller, as well as other camp team members who helped with the rst workshop. We also acknowledge the support of the Italian Education Body of the Province of Bozen-Bolzano for the workshop in Italy, as well as the nancial support of the GOST and SOOfA projects. 31