Chapter 5.docx

Chapter 5

EMOTIONAL INTERACTION       

5.1 Introduction 5.2 Emotions and the User Experience 5.3 Expressive Interfaces 5.4 Frustrating Interfaces 5.5 Persuasive Technologies and Behavioral Change 5.6 Anthropomorphism and Zoomorphism 5.7 Models of Emotion

Objectives The main aims of this chapter are to:       

Cover how emotions relate to the user experience. Provide examples of interfaces that are both pleasurable and usable. Explain what expressive interfaces are and the effects they can have on people. Describe how technologies can be designed to change people's attitudes and behavior. Give an overview on how anthropomorphism has been applied in interaction design. Present well-known models and frameworks of emotion and pleasure. Enable you to critique the persuasive impact of an online agent on customers.

5.1 Introduction An overarching goal of interaction design is to develop products that elicit positive responses from users, such as feeling at ease, being comfortable, and enjoying the experience of using them – be it a washing machine or a flight deck. Designers are also concerned with how to create interactive products that elicit specific kinds of emotional responses in users, such as motivating them to learn, play, or be creative or social. There has also been much interest in designing websites that people can trust, and that make them feel comfortable about divulging personal information when making a purchase. Taken together, we refer to this emerging area as emotional interaction. In this chapter we look at how and why the design of interactive products may cause certain kinds of emotional responses in people. We begin by looking in general at expressive interfaces, examining the role of an interface's appearance to users and how it affects usability. We then examine how interactive products elicit positive effects, e.g. pleasure, and negative responses, e.g. frustration. How technologies are being designed and used to persuade people to change their behavior and attitudes is then covered. We look, in particular, at ubiquitous technology interventions that are being designed to improve health and well-being and reduce domestic energy and water consumption. Following this, we show how anthropomorphism has been used in interaction design and the implications of designing applications that have human-like qualities. A number of virtual characters and robot pets are described that have been developed to motivate people to learn, buy, and listen and we consider how useful and appropriate they are. Finally, we present three models that are well known in interaction design that conceptualize the user experience in terms of emotion, pleasure, and user experience: (i) Norman's (2004) emotional design model; (ii) Jordan's (2000) pleasure model for product design; and (iii) McCarthy and Wright's (2004) technology as experience framework.

5.2 Emotions and the User Experience Emotional interaction is concerned with how we feel and react when interacting with technologies. It covers different aspects of the user experience, from how we feel when first finding out about a new product to getting rid of it. It also looks at why people become emotionally attached to certain products (e.g. virtual pets), how social robots might help reduce loneliness, and how to change human behavior through the use of emotive feedback. Consider the different emotions you experience for a common everyday activity – shopping online for a product, such as a new phone, a washing machine, or a vacation. Firstly, there is the realization of needing or wanting it, and then the desire and anticipation of purchasing it. This is followed by the joy or frustration of finding out more about what products are available and deciding which to choose from potentially hundreds or even thousands (by visiting numerous websites, such as comparison sites, reviews, recommendations, blogs). This entails matching what is available with what you like or need and whether you can afford it. The thrill of deciding on a purchase may be quickly followed by the shock of how much it costs and the disappointment that you can't afford it. The process of having to decide again may be accompanied by annoyance as you can't find one that is as good as your first choice. You think about other options, such as seeking advice from an expert in a shopping mall, but you have an aversion to sales assistants and don't trust their advice, because you think they have their own interests (making money), rather than yours, at heart. So you carry on looking, getting more tired and frustrated. When you do make a decision you experience a sense of relief. You click merrily though the various options (such as color, size, warranty) and then the dreaded online payment form pops up. You type in all your details and press the final payment button. A window then appears saying that your credit card number is incorrect. So you type it in again very slowly. And you notice you need to type the threenumber security code in again. Finally, when all is done you let out a big sigh. But as you walk away from your computer doubts start to form in your mind – maybe you should have bought the other one … This rollercoaster set of emotions is what many of us experience when shopping online, especially for expensive products, where there are a myriad of options to choose from and where we want to be sure that we make the right choice. ACTIVITY 5.1 Wufoo is a company specializing in building online forms, with the intention of transforming what are usually boring and tedious tasks into more fun activities. How do the forms in Figure 5.1 compare with others you have had to fill in? Comment One difference with these forms is the way minimalism, balance, and aesthetics have been used in the design. As commented by one of Wufoo's creators, Kevin Hale: “The inspiration for our color palette came from our competitors. It was really depressing to see so much software designed to remind people they're making databases in a windowless office and so we immediately knew we wanted to go in the opposite direction. My goal was to design Wufoo to feel like something Fisher-Price would make. We were determined to make sure Wufoo was fun.”

Figure 5.1 Examples of Wufoo's online forms

Emotional interaction is about considering what makes us happy, sad, annoyed, anxious, frustrated, motivated, delirious, and so on, and translating this knowledge into different aspects of the user experience, from when we first want something to when we no longer interact with it or need to replace it. However, it is not straightforward to achieve as people's moods and feelings are constantly changing. There are also many reasons that might cause someone to be happy or sad, such as the sun shining or someone else winning a game. A good place to start understanding how emotions affect behavior and how behavior affects emotions is to examine how people express themselves and read each other's expressions. These include understanding the relationship between facial expressions, body language, gestures, and tone of voice. For example, when people are happy they typically smile, laugh, and open their bodies up. When they are angry they shout, gesticulate, and screw up their face. A person's expressions can trigger emotional responses in others. So when someone smiles it can cause others to feel good and smile back. Emotional skills, especially the ability to express and recognize emotions, are central to human communication. Most of us are highly skilled at detecting when someone is angry, happy, sad, or bored by recognizing their facial expressions, way of speaking, and other body signals. We are also very good at knowing what emotions to express in a given situation. For example, when someone has just heard he has failed an exam we know it is not a good time to smile and be happy. Instead we try to empathize. In addition to creating user experiences that elicit, avoid, or encourage certain kinds of emotional reactions, another approach, called affective computing, has attempted to develop computer-based systems that recognize and express emotions in the same way humans do (Picard, 1998). A long-standing area of research in artificial intelligence and artificial life has

been the creation of intelligent robots that behave like humans and other creatures. An early classic was COG, where a group of researchers attempted to build an artificial two-year-old. An offspring of COG was Kismet (Breazeal, 1999), which was designed to engage in meaningful social interactions with humans (see Figure 5.2). Besides being important theoretical research, the approach also has practical aims. For example, one application is to consider how to design technologies that can help people feel better, and which are “capable of soothing a crying child or of perhaps artificially preventing strong feeling of loneliness, sadness, frustration, and a host of other strong negative emotions, (Picard and Klein, 2002, p. 23). However, it should be stressed that designing computers to artificially care for humans, through listening, empathizing, and being cheerful, is not viewed as a replacement for human care but as an aid.

Figure 5.2 Kismet the robot expressing (a) surprise and (b) disgust

5.3 Expressive Interfaces Expressive forms like emoticons, sounds, icons, and virtual agents have been used at the interface to (i) convey emotional states and/or (ii) elicit certain kinds of emotional responses

in users, such as feeling at ease, comfort, and happiness. Icons and animations have been used to indicate the current state of a computer or a cell phone, notably when it is waking up or being rebooted. A classic from the 1980s and 1990s was the happy Mac icon that appeared on the screen of the Apple computer whenever the machine was booted (see Figure 5.3a). The smiling icon conveyed a sense of friendliness, inviting the user to feel at ease and even smile back. The appearance of the icon on the screen was also very reassuring to users, indicating that their computer was working correctly. This was especially true for situations where users had to reboot their computer after it had crashed, and where previous attempts to reboot had failed (usually indicated by a sad icon face – see Figure 5.3b). After nearly 20 years, the happy Mac icon was laid to rest although the sad Mac icon now shows its face on an iPod if its software needs restoring (see Figure 5.3c). Apple has since switched to the use of more abstract icons to indicate starting up and busy with a process, showing a swirling clock or a colorful beach ball. Other ways of conveying the status of a system are through the use of:   

 

Dynamic icons (e.g. a recycle bin expanding when a file is placed in it and paper disappearing in a puff when emptied). Animations (e.g. a beach ball whirling to say the computer is busy). Spoken messages, using various kinds of voices, telling the user what needs to be done (e.g. GPS navigation system instructing you politely where to go after having taken a wrong turn). Various sonifications indicating actions and events (e.g. whoosh for window closing, schlook for a file being dragged, ding for new email arriving). Vibrotactile feedback, such as distinct cell phone buzzes that specifically represent special messages from friends and family.

Figure 5.3 Smiling and sad Apple icons for the classic Mac (a) and (b); and for the iPod (c)

One of the benefits of using these kinds of expressive embellishments is that they provide reassuring feedback to the user that can be both informative and fun. They can, however, sometimes have the opposite effect on people, who find them intrusive, causing them at times to get annoyed and even angry. This is especially so for cutesy ones. How enjoyable they are perceived to be varies considerably across cultures. In South Korea and Japan, for example, cute cartoon characters such as those depicted in Manga comics have huge cult followings. Moreover, their influence has become widespread in the design of websites, video games, cell phone wallpapers, etc. These include the use of large-eyed creatures such as those used in Pokemon and Pikachu games, and bright colors and pastels (Marcus, 2002). Users themselves have also been inventive in expressing their emotions at the computer interface. A universal method is the use of emoticons, originally keyboard symbols that were combined in various ways to convey feelings and emotions by simulating facial expressions such as smiling, winking, and frowning on the screen. The meaning of an emoticon depends on the content of the message and where it is placed in the message. For example, a smiley face placed at the end of a message can mean that the sender is happy about a piece of news

she has just written about. Alternatively, if it is placed at the end of a comment in the body of the message, it usually indicates that this comment is not intended to be taken seriously. Many emoticons are designed to be interpreted with the viewer's head tilted over to the left (a result of the way the symbols are represented on the screen). Some of the best-known ones that are commonly used are presented in Table 5.1. Nowadays, people can select from sets of ready-made ones that are often 3D, very colorful, and cute such as those shown in Figure 5.4.

Table 5.1 Some commonly used emoticons

Figure 5.4 A collection of graphical smiley icons

ACTIVITY 5.2 How often do you use emoticons when communicating with your friends via different social media (texting, emailing, chatting, tweeting, or Facebooking)? Do you use them sparingly or a lot? Does it vary depending on the social media you are using? Does using them help you express your emotions effectively? Comment The use of emoticons has become ubiquitous in all social media but people vary as to how much they use them across different media and how they are feeling at that moment. It also depends on who the recipient is and the topic of the conversation/message. However, it seems that we use them most when chatting (e.g. when using IM, Skype, FB chat) – to tease, negotiate, charm, and cajole. Some of us prefer to use punctuation (!!) to stress or simply leave things hanging (…).

The style of an interface, in terms of the shapes, fonts, colors, balance, white space, and graphical elements that are used and the way they are combined, can also influence its emotional impact. Use of imagery at the interface can result in more engaging and enjoyable experiences (Mullet and Sano, 1995). The design of aesthetically pleasing interfaces has also become of central concern to interaction design. Empirical studies have shown that the aesthetics of an interface can have a positive effect on people's perception of the system's usability (e.g. Tractinsky, 1997). When the look and feel of an interface is pleasing – e.g. beautiful graphics, nice feel to the way the elements have been put together, well-designed fonts, elegant use of images and color, a good sense of balance – users are likely to be more tolerant, e.g. they may be prepared to wait a few more seconds for a website to download. Furthermore, good-looking interfaces are often more satisfying and pleasurable to use. A key concern, therefore, is to strike a balance between designing aesthetic and usable interfaces (Tractinsky et al, 2000). ACTIVITY 5.3 Figure 5.5 shows two differently designed dialog boxes. Describe how they differ in terms of style. Of the two, which one do you prefer? Why? Which one do you think Europeans would like the most and which Americans? Comment Marcus, a graphic designer, created the two designs in an attempt to provide appealing interfaces. Dialog box A was designed for white American females while dialog box B was designed for European adult male intellectuals. The rationale behind Marcus's ideas was that European adult male intellectuals like “suave prose, a restrained treatment of information density, and a classical approach to font selection, e.g. the use of serif type in axial symmetric layouts similar to those found in elegant bronze European building identification signs.” In contrast, white American females “prefer a more detailed presentation, curvilinear shapes and the absence of some of the more brutal terms … favored by male software engineers.” When the different interfaces were empirically tested by Teasley et al (1994), their results did not support Marcus's assumptions. In particular, they found that the European dialog box was preferred by all and was considered most appropriate for all users. Moreover, the round dialog box designed for women was strongly disliked by everyone. The assumption that women like curvilinear features did not hold in this context. At the very least, displaying the font labels in a circular plane makes them more difficult to read than when presented in the conventionally accepted horizontal plane.

Figure 5.5 Square and round dialog boxes designed by Marcus (1993): (a) dialog box designed for white American women; (b) dialog box designed for European adult male intellectuals

5.4 Frustrating Interfaces In many situations, computer interfaces may inadvertently elicit negative emotional responses, such as anger and disgust. This typically happens when something that should be simple to use or set turns out to be complex. The most common examples are remote controls, printers, digital alarm clocks, and digital TV systems. Getting a printer to work with a new digital camera, trying to switch from watching a DVD to the TV, and changing the time on a digital alarm clock in a hotel can be very trying. This does not mean that developers are unaware of such usability problems. Several methods have been devised to help the novice user get set up and become familiarized with a technology. However, these have sometimes backfired, since the design solution itself has ironically become a source of annoyance and frustration. For example, one technique that was popularized in the 1990s was the use of friendly agents at the interface. The assumption was that novices would feel more at ease with a companion and would be encouraged to try things out, after listening, watching, following, and interacting with it. Microsoft pioneered a class of agent-based software, Bob, aimed at new computer users (many of whom were viewed as computer-phobic). The agents were presented as friendly characters, including a pet dog and a cute bunny. An underlying assumption was that having these kinds of agents on the screen would make users feel more comfortable with using the software. An interface metaphor of a warm, cozy living room, replete with fire and furniture, was also provided (see Figure 5.6) – again intended to convey a comfortable feeling.

However, Bob never became a commercial product. Why do you think not? Contrary to the designers’ expectations, many people did not like the idea of Bob at all, finding the interface too cute and childish. However, Microsoft did not give up on the idea of making their interfaces more friendly and developed other kinds of agents, including the infamous Clippy (a paper clip that has human-like qualities), as part of their Windows 98 operating environment. Clippy typically appeared at the bottom of a user's screen whenever the system thought the user needed help carrying out a particular task (see Figure 5.7). It, too, was depicted as a cartoon character, with a warm personality. This time, Clippy was released as a commercial product but it was not a success. Many Microsoft users found it very trying and intrusive, distracting them from their work. When it was finally retired, numerous websites posted jokes and witty comments, celebrating its demise.