SenseMS

SenseMS: A User-centered Approach to Enrich the Messaging Experience for Teens by Non-verbal Means A.K. Amin, B.T.A. Kersten, O.A. Kulyk, P.H. Pelgrim, C.M. Wang, P. Markopoulos User System Interaction, Technical University Eindhoven Den Dolech 2, 5612 AZ Eindhoven, The Netherlands

{a.k.amin, b.t.a.kersten, o.a.kulyk, p.h.pelgrim, c.wang, p.markopoulos}@tue.nl The main reasons for the popularity of SMS are: it is fast, cheap, easy, and convenient [9]. SMS also helps teens to control costs, avoiding long and unwanted conversations and they consider the use of SMS as more convenient, because they can do it in private, silently in public places or late in bed. Furthermore, the sender can choose how much time to spend on composing, editing, and redoing the message. It is possible to rethink about messages before finally sending them, because it is not real time communication [22].

ABSTRACT This paper reports the user study and design of a concept to improve mobile messaging for teens. A study of current mobile phone use by teens (16-18) showed that, while they prefer communicating by Short Message Service (SMS), they miss expressiveness in this application. An enhanced SMS application, SenseMS, is designed to support affective communication. An evaluation of a SenseMS prototype has shown that enhancing text messages with contextual information and human embodiment can result in a more pleasant experience for both the sender and receiver. Especially for negative emotions, contextual and emotional information are essential for interpreting the message. The study also showed different usage scenarios, in that SenseMS is preferred for emotion-rich messages, whereas SMS is more appropriate for fast and emotion-poor messages.

On the other hand, SMS also has disadvantages. Text messaging lacks expressiveness [3] and human embodiment (demonstrated by the usage of avatars in comparable applications such as Instant Messaging on a PC-platform [28]). Thus teens lack support for non-verbal aspects. These non-verbal aspects account for a great proportion of the total communication in face-to-face contact (body movements, facial expressions, and context) or even in a telephone conversation (paralanguage; pitch, regional accent, groans, pauses, hesitations, but also sarcasm, truthfulness, and emotions). Due to the fact that a receiver tends to interpret the intentions of the sender based on the non-verbal cues he receives, the absence of these non-verbal cues brings difficulties for receivers to interpret a message correctly. This means that teens may have difficulties with understanding the meaning behind SMS messages, interpreting the sender’s intentions, understanding the emotions of the sender and identifying sarcasm/humour.

Categories and Subject Descriptors: H.5.2 [Information Interfaces and Presentation]: User Interfaces – user-centered design. General Terms: Design. Keywords: Teens, mobile phone, non-verbal communication, avatar, text messaging, SMS.

1.INTRODUCTION 1.2 Context

As teens (13-18 years old) grow older they become more socially active and their needs to stay in contact increases [18]. They use many ways to stay in contact with each other, e.g. face-to-face, instant messaging, e-mailing, (mobile) calling, and Short Message Service (SMS). Especially in the usage of mobile phones teens have become true forerunners [20]. For example, recent research shows that already 98% of 14-years old Dutch teens have a mobile phone [13].

Besides non-verbal aspects, another important aspect of communication is context. In face-to-face communication people share the same physical space. Therefore they share an intuitive awareness of relevant circumstances, objects, or conditions in the surrounding environment, which will greatly enhance communication and prevents misunderstanding [24]. According to King [15] context consists of the psychological context (personality, needs, and desires), relational context (relationship between the persons), situational context (the psycho-social of “where” you are, e.g. on leisure time or official), environmental context (the physical of “where” you are, e.g. location and time), and cultural context. It is assumed that the relational context and the cultural context are already understood, because SMS messages are typically used in communication between friends, lovers, or family members [16]. However, the psychological, situational, and the environmental context are not self-evident. Therefore these aspects will be explored further.

1.1 Mobile Text Messaging by Teens Short Message Service (SMS) is one of the most popular mobile phone applications among teens [13]. They use it for arranging times to chat, to change plans that are already made, to coordinate with friends and family, to just chat or gossip [9], and to agree on when and where to meet [19]. 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. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. MobileHCI’05, September 19–22, 2005, Salzburg, Austria. Copyright 2005 ACM 1-59593-089-2/05/0009...$5.00.

1.3 Related Services Because of the success of SMS, manufacturers have introduced enhanced related services, like Multimedia Messaging Service

161

life style, with the environment they live in and with their experiences.

(MMS). MMS is supposed to make communication richer; by offering the possibility to the user of sending multimedia messages that include pictures, sounds, and animations. Also text can be enhanced in different manners, e.g. modifying the font size.

The workshops were divided into several parts, which consisted of diverse tasks. Some of the assignments were individual, while others had to be performed in a group. A good variation was needed to enhance creativity. The first part was called “This is me”. In this assignment participants had to make a collage to represent themselves. The second part was called “Them and me”. For this assignment they had to add different means of communications (telephone, SMS, face-to-face, MMS, chat, postcard, and letter) they have with persons around them by putting icons around these persons. Also they had to add the frequency of contact (frequent/middle/infrequent). The third part was a group assignment and was called “Holiday”. In this assignment the participants were split in two groups and each group had to make a message-collage about a different holiday scenario. Subsequently the scenarios were exchanged between the groups and each group had to interpret and respond to each other’s message.

Nevertheless, since its introduction on the Dutch market in 2002, the uptake of MMS has not been what was expected. In Europe, only 3% of the people with a mobile phone used MMS in 2003 [10]. The main reasons for this low uptake seems to be that: MMS requires special phones [2], picture quality is not satisfactory for sharing pictures [14], the screen size is too small [14], there is no compatibility between operators [10], and sending an MMS is too expensive [26]. We also do not know if attaching arbitrary pictures contributes to the understanding of a text message. Our research would then focus on what kind of information would be meaningful to have along with text messages.

1.4 Outline of the paper This paper presents an exploration into the enrichment of messaging with SMS. The aim is to support the communication of emotions during informal social communication between teens, using their mobile phones. In the remainder of this paper there will be a description of what was carried out to reach this goal. Therefore the following items will be discussed: related work, a user study, and the implementation and testing of the prototype. The paper concludes with a discussion and directions for future work.

3.1 The Participants In total 9 teens (4 girls and 5 boys) participated in the workshops. They were all aged between 13 and 18 years old. The workshops were split according to gender, because we wanted to gain insight into differences in communication behaviour between girls and boys.

3.2 Workshop Results Many of the behaviours and preferences, described in the literature, were confirmed during the workshops. For example, teens demonstrated that SMS is a very important means of communication for them. The main reasons for this were that it is fast and easy. Furthermore, it is common for teens to add emoticons to their messages to express their emotions and intentions. For the receiver emoticons increase the understanding of the content of the message and it allows the senders to express themselves. During the assignments a lot of pictures and cartoons of facial expressions were used for two reasons: as a selfrepresentation (girls tend to pick out girl/woman pictures and boys choose boy/man pictures) and to convey how they were feeling at that time, e.g. surprise, happy, angry, etc. (see figure 1).

2. RELATED WORK There are already several researches that investigate mobile applications for teens, though not explicitly on supporting messaging with non-verbal elements. Some studies showed that messaging is very popular among teens [5][20]. One example is the Internet messaging for teens with expressive avatars [27]. On the mobile platform, a prototype called ExMS emphasizes on avatars for messaging services [23]. It was reported that the avatar is used to express two types of emotions: real emotions and fictional expressions/behaviour. Berg et al. [3] have attempted to design a mobile messaging application, focusing on presence awareness and the form factor of the device that is appealing for teens. In addition to this, they saw the importance of communicating identity, mood, and feelings within the teen social networks. Also Blom et al. [4] explored location-based information for SMS, showing that location-based information in SMS is popular among teens but mostly affective to stimulate conversation with non-acquainted users.

3.

USER STUDY

To understand teens better, a user study was conducted in the form of a workshop. The study involved users in two participatory design workshops, where several of the activities were inspired by the Cultural Probes technique [12]. The material that was used during the workshop included: a workshop booklet with attractive colourful pages and different creative materials: drawing materials, teen magazines, emoticons, and stickers. During the workshops the probes were presented as assignments in a colourful booklet, which were adapted to teens’ interests and preferences. The aims of the assignments were: to gather inspirational responses from users as input for design ideas and to attain empathy with target users by getting acquainted with their

Figure 1. The teen’s craftwork: Facial expressions in a message from a teen boy (left) and a result from “holiday” assignment; the picture articulates landscape/place, activity, situation and again facial expressions.

162

The workshop also showed that context awareness is an important aspect in communication for teens. In the ‘Holiday’ assignment, they used pictures of the environment and icons to represent where they were (e.g. on the beach or at the hotel, see figure 1). It was also noticed that often during phone calls (e.g. in the train or in shops) the teens first ask each other where they are.

4.1 The Stamp The stamp (see figure 3) is the first thing the user sees when he/she receives a SenseMS. It provides the situational and environmental context (by means of a background) and the psychological context (the emotion of the avatar). In addition, the sender’s name, the time and date the message was sent are displayed. This way the receiver of the message has a lot of information at a glance, before the actual text message is even read.

An important finding is that when teens had the option, they preferred to include text, instead of only icons or pictures. Most teens make use of all sorts of colours not only when drawing pictures but also in text (see figure 2). This can be regarded as a way to more personalize and emphasize messages. This feature is also commonly used in chatting applications for the same reason.

The avatar’s facial expressions can be adjusted according to the user’s preference (see figure 4). In the prototype some of the basic emotions are implemented (happy, smiling, neutral, sad, and angry [7]). The background picture can be customized as well (see figure 4).

Figure 2. Various colouring tools used by the teens during the workshop (left), colourful text made by a teen girl (right).

4.

Figure 3. The stamp with the avatar, the background, date and time sent, and the sender of the message.

DESIGN CONCEPT

From the results of the users study four elements are chosen as a basis of the design. These elements represent the psychological, situational, and environmental information that should be conveyed along with sending a message: 1.

2.

3.

4.

Facial expression: A crucial part of non-verbal communication is body language. Facial expressions especially help in the interpretation of the intention of the communication content. Using smileys is an example of facial expressions in text format (like ;-) representing a blinking smiley). Self-presentation: The use of avatars is one way for people to visually present themselves in the virtual world. Also they give the possibility to be represented in a real or surreal (e.g. caricatures, cartoons, idols) way. Especially among teens, avatars are popular and widely used. Message personalization: When teens use Internet text applications (chat and Instant Messaging) they are very creative in expressing themselves with various text colours and sizes. Context awareness: The workshop showed that teens expressed the location information in the environment to elaborate their messages.

Figure 4. Changing the background picture (upper right), the facial expression of the avatar (upper left), and colourful text and word suggestions for a selected word (lower left).

4.2 Enhanced text The second part of the SenseMS is the enhanced text. The sender of a SenseMS can customize the text appearance, by changing the background colour, the font type, and the font colour. In addition, the application gives suggestions for the presentation of special words. For example, if a user wants to type the word “jarig”

These elements were realized in two parts of the design: the stamp and the text message itself.

163

version of the prototype, with automatic retrieval of the background and text effects.

(birthday), the SenseMS application will automatically present this word in different text effects (including small pictograms (see figure 4)). It is expected that in this way these words gain extra attention. The interaction with this feature has some similarity with T9™; the suggestions pop up as a user starts typing a specific word.

5.1 Experimental Set-up 5.1.1 Subjects In total 3 pairs of teens (8 teens, aged 16-18, and good friends of each other) participated in the evaluation; 1 pair of girls, 1 pair of boys, and 1 mixed pair of a girl and a boy.

4.3 The SenseMS Prototype

We focused on designing to support affective communication, which will probably fulfil the needs of teenage girls better than boys. Nevertheless, boy participants are recruited to help highlight potential differences between boys and girls.

The two elements of the SenseMS application were implemented in Flash on a Pocket PC (ASUS MyPal A620BT) (see figure 5). Although implemented in a PDA, the interaction with the application is designed to be as close as possible with a mobile phone.

5.1.2 Questionnaire In total 4 questionnaires were used, which were bundled into one booklet. Open-ended questions as well as closed questions were used. Closed questions were scored by means of a 5-point Likert scale, which participants used to indicate their agreement or disagreement with a statement. The “Receiver” questionnaire compared SMS with SenseMS regarding the understanding of messages (emotions, background, and intention) and teen’s preferences. The “Sender” questionnaire concerned usability issues regarding the creation of a SenseMS. The “General” questionnaire consisted of a combination of already existing questionnaires, which were sometimes adapted to match the SenseMS application and our research purpose. The questionnaires that were included are: QUIS 5.0 [6] for assessing the subjective usability of the application; items that were not applicable, e.g. about noise level, were omitted. Fun [25] for assessing the experienced fun; items covering the topics of engagement, expectations, and endurability were included. FEASP [1] to measure emotions felt during the session; the parts of the questionnaires relating to fear, anger, and pleasure were included to assess the feelings experienced during both the sending and receiving of a SenseMS. IPO-SPQ [8] for assessing the social presence experienced by participants; social presence is the feeling of the remote person being there with them during the communication and provides a way of describing the ‘warmth’ and ‘sociability’ of a communication experience. The parts of the questionnaire pertaining to the realism and immediacy of the communication were outside of the scope for this study and were omitted. In addition, 2 questions, constructed by the researchers, were included to measure the cool-aspect of the application.

Figure 5. The SenseMS application on the Pocket PC.

5. EVALUATION Research has shown that within the target group there are no major differences between the sexes in mobile ownership [20], that is the knowledge and experience are about the same; boys and girls are very skilled with their current mobile phone. However, the psychological characteristics of boys and girls can vary significantly just as the usage of mobile phones. Boys emphasize on technical functions. They tend to describe events rather than emotions when sending a message [17]. Girls emphasize more on the actual communication, they send more SMS messages in comparison to boys [9][20], and they are more focused on the emotions when sending a message [17]. The behaviours and interests of the different age groups within the target group also vary significantly. Most teens of 16-18 years old in the Netherlands change schools and start having a weekend/vacation job. As a consequence the geographical area in which their friends are located expands, so they have a higher need to stay in contact with these friends [18]. On the other hand, a weekend/vacation job gives them the opportunity to buy new gadgets for themselves. Based on these differences it was decided to narrow down the focus and test the designs with late teens (1618 years old).

5.1.3 Procedure First a slide show was shown to introduce the participants to the SenseMS application. This slide show was a scenario of how a teen uses SenseMS during the day. In addition, the teens were given some introductory tasks to get familiar with the application. After this training session the pairs were split into a sender and a receiver. When the sender completed a message, the roles were switched. Every participant created in total 3 messages and these messages were later used during the debriefing.

An evaluation was conducted to test the extent to which SenseMS has addressed the teens’ needs and preferences. To set up the experiment, participants were asked, prior to the experiment, to provide actual SMS messages they had sent to their friends, to name locations they regularly visit and to choose their favourite avatar. This information was used to design a personalized

164

most to their feeling of being closer to the other person. All participants shared the opinion that they could send a cool message with SenseMS. This was mainly because of the pictures of the background and the avatar. A t-test showed that there are only three significant differences between the means for boys and girls (Preference SenseMS t=2.95 p