PRODUCT EMOTION

15

PRODUCT EMOTION PIETER M.A. DESMET Delft University of Technology, Delft, The Netherlands

1. INTRODUCTION Emotion is a central quality of human existence, and most of our behavior, motivation, and thought is enriched with and influenced by emotions. Our relationship with the world is essentially affective, which means that all our interactions imply and involve emotions, whether these are interactions with the social world or with the material one. A product, or using a product, can elicit disappointment, attraction, shame, pride, disgust, contempt, admiration, satisfaction, fear, anger, and any other emotion a person may also experience in response to events, people, or actions of people. Ignoring the emotional side of product experience would be like denying that these products are designed, bought, and used by humans. The focus of the current chapter is on that emotional side, and the term ‘product emotion’ will be used to refer to all emotions experienced in response to, or elicited by, seeing, using, owning, or thinking about consumer products. Those who are involved in design processes may sometimes be inclined to regard the concept of product emotion as intangible, and therefore unsuitable for a structured approach. Several reasons can explain this attitude. First, the concept of product emotion is broad and somewhat indefinite, because products can elicit many different kinds of emotions. We can admire the latest stylish music player, be irritated by an annoying mobile phone, be fascinated by the transparent fragility of a porcelain cup, and so on. And although the touch of melancholy felt when coming across a long forgotten childhood teddy bear seems incomparable to the thrill of driving a motorcycle, both responses belong to the same wide spectrum of product emotions. Secondly, emotions are subjective, and individuals will differ with respect to their emotional responses to a given product. We all know from experience that one person may be attracted to the same car model that is disliked by another. Even more so, the same person may experience different emotions Product Experience Copyright © 2008 Elsevier Ltd.

379

380

Product Experience

to a given product at different points in time. Thirdly, products often evoke ‘mixed’ emotions. Rather than eliciting one single emotion, products can elicit multiple emotions simultaneously because many different product aspects can have an emotional impact, such as the general product appearance, particular details, implicit and explicit expectations, and associated, remembered and fantasized meanings. Given the subjective and mixed nature of product emotions, only few one-to-one relationships between product design and emotional responses can be identified. However, one of the main themes of my discussion is that emotions are less intangible when they are analyzed on the level of the underlying process. The process that elicits emotions is primarily universal, and lawful relationships in the conditions that underlie emotions can be identified. The current chapter discusses a model of product emotion that was developed to describe some of these universalities by specifying the structure of the process that elicits emotions. Although the focus will be on tangible products, the model also applies to intangible services. The model represents a psychological view on product emotion because it represents the cognitive mechanisms that intervene between seeing, using, owning, or thinking about a product, and the emotional outcome. By focusing on the underlying process, the model aims to impose some structure on the limitless number of possible emotion-eliciting situations associated with product design. The model builds on my previous writings about product emotions, and introduces some new insights. The main aim is to set up a general framework that might help to structure the discussion of questions related to the topic of product emotion, i.e. how product emotion can be distinguished from other product experiences, why and how product emotion is elicited, and to what extent designers can influence or ‘design for’ these emotions.

1.1. Scope and structure of the chapter The chapter consists of three sections. The first section is a general analysis of the phenomenon emotion, focusing on some basic questions: What is emotion, how can it be differentiated from other affective phenomena, what manifestations are associated with it, why we experience it, and how is it elicited. The review aims to provide some basic answers to these questions that, on the one hand, are drawn from recent insights in emotion theory, and on the other hand, are not estranged from the everyday notion of what (product) emotion is. Some of the latest insights in emotion psychology are discussed, building on fundamental contributions to the science of emotion. Besides emotion, mood and affect disposition are also briefly discussed because they influence, and are related to, emotion. Although the summary does not do justice to the complex and vigorous nature of the discussions documented in the research literature, it does provide some background knowledge, and a foundation for the model and sources of product emotions discussed in the subsequent sections. The second section introduces the key variables of a basic product emotion model that was developed on the basis of an appraisal approach to emotion. In addition, two alternative approaches to product emotion are discussed: The pleasure approach of Jordan (2000), and the process-level approach of Norman (2004). The third section discusses sources of product emotions and uses examples to illustrate the different ways in which product emotion can be elicited.

2. AFFECT AND EMOTION Human beings have always shown a passionate interest in expressions of, and reflections on, the experiential phenomenon we call emotion. There is no beginning in counting the

Product emotion

381

attempts to conceptualize and define emotion that have been introduced in the (research) literature, and every year new ones are published. We should remember that although the social sciences have embraced the concept, it originated from the domain of our everyday experience, and therefore is a layman’s rather than a scientific concept. In other words: We all have emotions, and thus, from experience we all know what they are. This applies to the emotions we have in response to products just as much as it applies to those we may have in response to significant events in our life. Our implicit knowledge of emotion comes with the advantage that design and emotion scholars do not need to overly bother themselves with theorizing or defining the phenomenon, but can instead focus their attention on design-relevant questions, such as how and when products elicit emotions, and how products can be designed to have a positive emotional impact on their users. At the same time, one should be aware of relying too much on the common sense concept of emotion because these intuitive concepts differ between people, which introduces the risk of using product emotion as a container term for anything related to product design that involves some element of subjective experience (for a discussion, see Demir, Desmet and Hekkert, 2006). Before we turn our attention to products, we will therefore focus on some basic questions about the nature of emotions: What is emotion? How can we distinguish emotion from other affective phenomena like mood? How can we differentiate between emotions? What is the function of emotion?

2.1. Core affect When describing their emotions experienced in response to buying, owning or using products, people often use words like emotion, mood, feeling, and sentiment, as if these are alike. Many designers, and design researchers, tend to do the same. Because these words do refer to different phenomena, our conceptual exploration of product emotion requires a basic understanding of their similarities and differences. A first and basic distinction is between states that are affective and those that are not (see Clore and Ortony, 1988). In emotion theory, the term affect, or affective state, is generally used to refer to all types of subjective experiences that are valenced, that is, experiences that involve a perceived goodness or badness, pleasantness or unpleasantness. This approach fits the view of Zajonc (2000), who proposed that affect is the simplest raw, and universal (nonreflective) positive or negative feeling. Emotion research shows a long tradition in using valence as a bipolar dimension to describe and differentiate between affective states (e.g. Bradley and Lang, 1994; Plutchik, 1980; Wundt, 1905). Russell (1980; 2003) introduced the concept of ‘core-affect’ by combining the affect dimension with physiological arousal in a circular two-dimensional model. According to Russell the experience of core-affect is a single integral blend of those two dimensions, describable as a position on the circumplex structure in Figure 15.1. The horizontal axis represents valence (that ranges from unpleasant to pleasant), and the vertical axis represents arousal (that ranges from calm to excitement). The words in the figure represent emotions that have been found to be often experienced in response to the appearance of consumer products (see Desmet, 2002). We always experience core affect: From the moment we wake up to the moment we fall asleep our core affect constantly responds to a wide variety of internal (e.g. hormonal changes, nutritional deficiencies) and external causes (e.g. events, people, objects, weather). Core affect can be neutral (the central point), moderate, or extreme (the periphery). Changes can be short lived or long lasting, and can be in the focus of attention (in the case of intense core affect), or a part of the background of the person’s experience (in the case of mild core affect). Core affect can be experienced in relation to a

382

Product Experience

Activated

Irritation Disgust Alarm

Astonishment Eagerness Curiosity

Disappointment Contempt Unpleasant Jealousy

Pleasant

Calm

Boredom Sadness Isolation

Inspiration Desire Love

Awaiting Deferent Calm

Fascination Admiration Joyfulness

Satisfaction Softened Relaxed

FIGURE 15.1 Circumplex model of core affect with product relevant emotions (Desmet, 2002; adapted from Russell, 1980).

particular stimulus, as in emotion, but can also be experienced without a relation to a particular stimulus, as in moods (see the discussion in the next sections). Occasionally the cause of a change in core affect is obvious, but at other times, we can undergo a change in core affect without knowing why. Core affect theory and other dimensional theories of emotion offer a simple, yet powerful way to organize product affect, because all possible affective responses to (seeing, buying, using, owning, thinking about, repairing, etc.) products can be described in terms of core affect. The activated unpleasantness for the heated irritation in response to a failing computer, the calm pleasantness for the soothing experience of sliding into a warm bath, the activated pleasantness for the exhilaration of ice skating, and the calm unpleasantness for the sadness from the memory of a broken crystal vase, can all be plotted on the circumplex model. A second important strength of dimensional structures is that they allow for dynamic continuous affect reports. Dimensional models have therefore been found to be especially useful for capturing the continuous changes in emotional experience that occur during listening to music (see Schubert, 1996), using a product (see Laurans and Desmet, 2006), or looking at a TV ad (see Aaker, Stayman and Hagerty, 1986).

2.2. Attributed affect: Emotions Core affect can account for an infinite number of emotional states and provides a basis for discussing similarities and differences among affective states. Note, however, that not all affective states are emotions. Two eliciting conditions distinguish emotions from other types of affective states. An emotion is experienced when (1) some stimulus excites a substantial and acute change, or jump, in core affect (Ekman, 1994a; Russell, 2003); and when (2) this change is attributed to some antecedent or ‘object’ (Frijda, 1994b). All emotions, including product emotions, imply and involve a relation between the person who experiences them and a particular object (Frijda, 1994b). Note that this object is not necessarily a physical object, but can also be a person, an animal, a company, etc. One is not just angry, but angry at someone, afraid of something, proud of something, in love with someone, and so on. So we are angry at our printer for a bad quality print, satisfied with our toaster for making the perfect toast, and bored by our telephone for its tedious ring tone.

Product emotion

383

The distinction between stimulus and object is important because they are not necessarily the same. One day my sister and I were picking flowers in the meadow next to our parents’ house. At one point in time my sister bent over to pick a flower and, without us realizing it, her back touched the electric cattle fence. The resulting shock gave her a big fright, and, not knowing what hit her, she became angry with me because she concluded that I must have kicked her in the back. Although the electric fence was the actual emotional stimulus, the object of her emotion was me. This example illustrates that the object is whatever one believes is making him or her experience the emotion, and that in some cases we can misattribute an acute change in core affect due to a particular stimulus to another object (see Schwarz and Clore, 1983). One can, for example, be angry at a colleague for some mistake that he was actually not responsible for. The same can also apply to product emotions. One can be angry at the television set because it appears to be broken as it shows a white signal. But the objective facts might tell another story; perhaps the television only appears to be broken because of an interruption in the signal supply; or one should actually blame oneself for accidentally disconnecting the signal cable. One can be irritated by the call centre operator when the irritation should actually be attributed to the company who hired the person. The retail industry is known for experimenting with intentionally generated misattribution effects. Supermarkets, for example, sometimes distribute the fragrance of freshly baked bread to generate pleasant feelings that the customer tends to misattribute to the bread that is for sale.

2.3. Free floating affect: Moods Mood typically refers to a diffuse affective state that is of lower intensity than emotion, but considerably longer in duration (see Davidson, 1994). Whereas emotions are acute, moods evolve gradually over time. One can be sad or cheerful for several hours or even for several days. As opposed to emotions, moods do not involve a particular attributed object (e.g. one is not cheerful at something), but are directed at the surroundings in general, or in the words of Frijda (1986) at the ‘world as a whole’. Whereas emotions are usually elicited by an explicit cause, moods typically result from a combination of causes (e.g. bad weather, and bad coffee, and traffic jam, etc.). There are at least three reasons why, besides emotion, mood is relevant for product design. The first reason is that products serve important mood-manipulating functions. People seek for opportunities to alter unpleasant and maintain pleasant moods, and will engage in activities to deliberately influence their mood state. These activities can involve and require particular products, such as shopping, taking a bath, playing a video game, making or listening to music, etc. In some cases products are literarily used as instruments for mood management because they are consumed for their direct positive effect on mood, like eating or drinking particular foods (candy, ice-cream), drugs (stimulants), or drinks (alcohol). The second reason is that mood has an influence on emotional responses. Although mood and emotion can be distinguished in terms of underlying conditions and manifestations, they are not independent. Moods involve a general tuning of emotional responses (see Frijda, 1993). This means that unpleasant moods lower the threshold for experiencing unpleasant emotions, and pleasant moods lower the threshold for experiencing pleasant emotions. A person who is assembling a wardrobe closet in a bad mood, will probably be more ready to get irritated when the task is more complex than anticipated than a person who is in a cheerful mood. Many salesmen will attempt to have a positive influence on the mood of a customer (by, for example, looking attractive, offering drinks, making jokes and compliments) because mood will most likely influence the emotional response to the actual object of purchase. The third reason is that mood

384

Product Experience

has an influence on human-product interaction, and as a consequence, also on product emotion (see Wensveen, 2005). We all know that our motivation and behavior is influenced by our mood. In the case of a good mood one will, for example, respond differently to a colleague’s request than in the case of a bad mood. The person who is assembling a closet in a bad mood may try to use additional force when confronted with ill-fitting parts, whereas a cheerful person might have taken the time to explore the situation before using force. Where the first person may break the part, and experience anger, the second may discover that he or she should try another part, and experience relief.

2.4. Dispositional affect: Attitudes Affect dispositions are relatively enduring, affectively colored beliefs, preferences, and predispositions toward objects, persons, or events (Frijda, 1986; Ortony, Clore and Collins, 1988; Russell, 2003). Some of our affect dispositions (also referred to as attitudes or sentiments) are innate, and others are acquired. Examples of innate affect dispositions are the preference for sweet and aversion for bitter tastes (Rozin and Fallon, 1987), and preferences for particular odours and for particular facial features and expressions (Etcoff, 1999). Humans also have affect dispositions that are acquired through conditioning and learning. One can acquire a taste for wines, particular fashion styles, social activities, etc. These acquired affect dispositions show substantial cultural and personal differences. The Dutch salted candy ‘drop’ (liquorice) is often experienced as disgusting by those who are not familiar with the taste. Note that affect dispositions can easily be confused with emotions because they use similar terminology (e.g. ‘I love chocolate,’ or ‘I hate jogging’). Nevertheless, being afraid of dogs (affect disposition) and being frightened by a dog (emotion) are essentially different states (Frijda, 1994b). Of course, we also have affect dispositions regarding products, such as a dispositional love for 2CVs, or product categories, such as a dispositional dislike for mobile phones. The same applies to brands or particular companies, or (consequences) of product usage (e.g. ‘I love the smell of freshly washed laundry that is drying in the afternoon sun’). These dispositions are highly relevant for product emotion because, as will be discussed later in this chapter, they are one of the key parameters in the emotion eliciting process.

2.5. Differentiating between emotions Their valenced nature distinguishes emotion from non-affective states, and the acute and attributed change in core-affect distinguishes it from other affective states. An important limitation of core affect theory and other dimensional models is that dimensions are not fully sufficient to differentiate between emotions. Emotions that are clearly different may be similar in terms of core affect, like anger and fear that are both high in activity and unpleasantness. In order to effectively distinguish between emotions, the dimensional perspective must be integrated with a categorical perspective (Russell, 1993). Researchers, working with a categorical approach, focus on the (measurable) manifestations or components that distinguish various emotions from one another. Most current categorical theories of emotions consider emotions as multicomponential phenomena that involve behavioral reactions, expressive reactions, physiological reactions, and subjective feelings. Behavioral reaction (e.g. running or seeking contact) is the action or behavior one engages in when experiencing an emotion. Emotions initiate behavioral tendencies like approach, inaction, avoidance, and attack (Arnold, 1960). Fear makes one want to run, love makes one want to approach or caress, and so on. Expressive reactions (e.g. smiling or frowning) are the facial, vocal, and postural expressions that are

Product emotion

385

part of the emotion. Each emotion is associated with a particular pattern of expressions (Ekman, 1994b). Anger, for example, comes with fixed stare, contracted eyebrows, compressed lips, vigorous and brisk movements, and, usually, a raised voice, almost shouting (Darwin, 1872; Ekman and Friesen, 1978). Physiological reactions (activation or bodily arousal, e.g. increases in heart rate) are the changes of activity in the autonomic nervous system that are part of the emotion. Emotions show a variety of physiological manifestations, such as pupil dilatation and sweat production. Subjective feeling (e.g. feeling happy or feeling angry) is the conscious awareness of the emotional state one is in, i.e. the subjective emotional experience. Although feelings are sometimes seen as a separate affective phenomenon, most researchers (Lazarus, 1991a; Ortony et al., 1988) agree that feeling is the subjective representation of emotion (or any other affective state). Note that feelings are often expressed in terms of one of the other components that constitute emotion. One can express the behavioral impact of an emotion (‘I was so angry, I felt like throwing my computer out of the window’), the expression (‘the game was so sad, I felt like crying’), or the physiological reaction (‘I was trembling from fear when I noticed the smoke coming from my kitchen’).

2.6. Function of emotion In everyday life emotions are often (mis)used as an excuse to justify irresponsible behavior: ‘I am sorry for shouting at you, but you must understand that my anger had taken control of me’, or ‘I know I shouldn’t speed, but I was so thrilled by the sports car’s power that I just couldn’t help myself’. It may therefore seem odd to discuss the function of emotions when basically considering them to be dysfunctional: As primitive responses whose purpose is much better served by our more advanced mental capacities (see Frijda, 1994b). Nevertheless, and contrary to this view, most contemporary emotion theorists view emotions as coherent, organized, and functional rather than dysfunctional systems (Smith and Kirby, 2001). In fact, the basic Darwinian presupposition that emotions fulfill some sort of function is probably shared by all psychologists (Frijda, 1994a). Darwin proposed that the best way to understand emotion is by considering its evolutionary history and contribution to the survival of the species and the individual. The key to understand why we have emotions with their particular characteristics lies in the ‘functional requirements’ created by the environments in which all animals live. ‘All organisms, in order to survive and maintain their populations, must find and ingest food, avoid injury, and reproduce their kind’ (Plutchik, 1984, p. 201). In order to survive, it is not sufficient for an organism to simply understand its situation; it has to be motivated to do something about it, and that is precisely what emotions do: They physically prepare and motivate the individual to contend with the adaptational implications of the eliciting situation. Fear comes with a tendency to flee, and anger with the tendency to attack, fascination with the tendency to explore. Frijda (1986) argues that these action tendencies, or states of readiness to respond, should be thought of as adaptive responses to events that have been important to us as a species in our evolutionary past, as well as in our current environment. The important advantage of action tendencies over rigid reflexes and fixed action patterns is that they enable flexibility, both in event interpretation and in response choice. Emotions ‘decouple’ behavior from the perception of the stimulus so that reconsideration is possible (Scherer, 1984). Fear creates a tendency to flee, but depending on the particular situation (the individual is cornered) an aggressive attitude to intimidate the attacker might be the better alternative. Emotions are functional because they establish our position vis-à-vis our environment, pulling us toward certain people, objects, actions, and ideas, and pushing us away from others (Frijda, 1986). This

386

Product Experience

basic principle applies to all emotions; the intense emotions that we may experience in situations that threaten basic survival needs, and the subtle emotions that we may experience in response to seeing or using a product. Pleasant emotions pull us to products that are (or promise to be) beneficial, whereas unpleasant emotions will push us away from those that are (or promise to be) detrimental to our well-being.

3. APPROACHES TO PRODUCT EMOTION An explanation of product emotion should meet three basic requirements. First, it should reflect both the individual and temporal variability in emotional responses elicited by products. Whereas one person may be attracted by a glass dinner table, another may feel contempt towards the same table, and whereas one person may be disappointed by the performance of a mobile phone, another may be pleasantly surprised by its innovative design. In addition, an individual’s emotional response to a given product may change over time. One may be, for example, initially satisfied with a new couch, but experience dissatisfaction after using it for some time. Secondly, the explanation should reflect the differentiated nature of product emotion. Products do not elicit mere like (attraction or pleasure) and dislike (aversion or pain) responses, but distinct emotions, such as astonishment, inspiration, fascination, boredom, sadness, jealousy, and many others. Moreover, we often do not feel a particular single emotion towards a product, but a combination of ‘mixed’ (and sometimes paradoxical) emotions. One person can be proud of a new pair of shoes and happy with the reaction of his or her partner, and at the same time irritated by the lack of comfort and afraid of damaging the delicate leather. Thirdly, the explanation should clarify the role of the product as a stimulus in the mechanisms that bring about product emotion. Although this third requirement seems obvious, it is important to mention because most emotion research is focused on the manifestations of emotions as such (e.g. facial expressions, arousal, or experience), or on the processes that underlie these manifestations (e.g. cognitive, biological, or physical), without detailing the relationship between these manifestations and the stimuli that elicit them. Although in the last decade the design research community has shown an increased awareness of the phenomenon of product emotion, the main literature sources indicate a surprisingly small interest in proposing general explanations of product emotion that meet the above stated basic requirements. Next to the appraisal approach discussed in this chapter, two basic approaches have been introduced, discussed, and applied: The pleasure approach (introduced by Jordan, 2000), and the process-level approach (introduced by Norman, 2004). Jordan uses a psychological pleasure-framework to explain various types of product pleasure, and Norman explains product emotion with a neurobiological emotion-framework that distinguishes several levels of information processing. Before introducing the appraisal approach, I will first briefly review these two alternative approaches to give a broader overview of theoretical perspectives in the design and emotion research domain. Readers will find that although they show some essential differences, these approaches are not mutually exclusive and share some basic assumptions and theoretical considerations.

3.1. Pleasure approach to product emotion In the field of human factors, Patrick Jordan (2000) introduced an influential pleasurebased approach. His main proposition is that traditional usability-based design approaches are limited and even dehumanizing because they tend to focus only on the fit of a product

Product emotion

387

to a person’s physical and cognitive characteristics. As an alternative, he advocates addressing the relationship between people and products holistically, judging the quality of designs on the basis of the wider relationships between products and the people for whom they are designed. Given this intention, he proposed a pleasure-based approach to human factors, in which pleasure with products is defined as the emotional, hedonic, and practical benefits associated with products (Jordan, 1999). The approach draws on a pleasure framework introduced by Tiger (1992) that distinguishes four conceptually distinct types of pleasure that people may seek: Physical; social; psychological; and ideological pleasure. Physio-pleasure has to do with the body and with pleasures directly derived from the sensory organs (such as touch, taste, and smell). Products are perceived with the sensory organs and therefore a direct source of physio-pleasure or displeasure. For example, a mobile phone can generate physio-pleasure because of its soft touch and elegant appearance. Socio-pleasure is the enjoyment derived from relationships with others. This type of pleasure is relevant for those products that facilitate social interactions. Some of Jordan’s examples are products that attract comments (like a piece of jewellery), or act as a focal point for social gatherings (like a coffee machine). Psycho-pleasure is related to people’s cognitive and emotional reactions, and has to do with the cognitive demands of using products. A text processor that is easy to operate provides a higher level of psycho-pleasure than one that is cumbersome and illogical, because the former processor enables the user to complete the task more easily than the latter. Ideo-pleasure is related to people’s values (i.e. pleasures from ‘theoretical’ entities such as books). Ideo-pleasure experienced in response to products is related with the values that the products embody. A product made from biodegradable materials, for example, might be seen as embodying the value of environmental responsibility. This, then, would be a potential source of ideo-pleasure to those who are particularly concerned about environmental issues. In his book, Jordan (2000) comprehensively discusses and illustrates how products can bring about each of these types of pleasures, and how to link the pleasures to particular aspects of product design (and displeasures to inadequacies with respect to certain product properties). The role of emotion in the framework is not discussed explicitly. The framework focuses on the general bipolar pleasantness dimension that applies to all affect (see Figure 15.1), and therefore does not explain the differentiated nature of product emotions. In addition, it seems more likely that emotions can be associated with all types of pleasure instead of with only the psycho-pleasure. In that sense it is not an approach to product emotion, but to the more general concept of product affect. The main value of the framework is that it can be used as a tool in the design practice. Jordan stresses that it is not a theory that is intended to give an insight into why people experience pleasure, but a tool that can make it easier for those involved in the design process to consider the full spectrum of the sorts of pleasures that products can bring. Given this intention, an important contribution of the work is that it clearly illustrates the layered nature of product affect, and some of the important variables that are involved in the underlying process, such as the sensory quality of the product (physio-pleasure), the social context in which the product is used (socio-pleasure), task-related concerns of the user (psycho-pleasures), and values of the user (ideo-pleasures). The following sections will demonstrate how the appraisal approach offers a theoretical basis for explaining the role of these variables in the process that elicits product emotion.

3.2. Process-level approach to product emotion Donald Norman (2004) proposed a framework of product affect that distinguishes between three types of affect and three corresponding design focuses. He used the multi-level

388

Product Experience

analysis of information processing discussed by Ortony, Norman and Revelle (2005) as the theoretical fundament for his framework. In this analysis, affect is considered at each of three levels of information processing: The reactive; the routine; and the reflective level. The most elementary, reactive, level involves fixed action pattern responses, such as reflexes or simple fleeing behavior. These are biologically determined responses, and an example is the impulse to immediately reject a bitter substance by spitting it out. At this level, no emotions are experienced but only simple affect, that is, an unelaborated positive or negative value that is restricted to the here and now (and does not persist in the absence of the stimulus). The second level is called routine because it is concerned with the execution of well-learned routine behaviors and skills. Routine level processes are able to quickly engage in or inhibit actions to correct for simple deviations from expectations. An example is riding a bicycle, an activity that involves routine actions to control balance and speed. At this level, primitive emotions are experienced, that is, emotions that have not yet been interpreted and cognitively elaborated, such as basic happiness and fear. The third level, the reflective level, is the most sophisticated because it involves all higher-level cognitive processes. In general, this level comprises consciousness together with all advanced cognitive skills, such as the ability to form generalizations, to make plans, and to solve problems. This level is responsible for the rich emotional experience that we assume is unique to humans. These are cognitively elaborated emotions that can implicate representations of the present, the future, or the past, such as relief and disappointment. Norman (2004) discusses how each of the three levels of processing are involved in affective product experience. His main claim is that each level involves a distinct type of product affect and a corresponding design focus. The first type, i.e. visceral affect, is perceptually based and corresponds with ‘visceral design’ that is concerned with product appearance. The second type, i.e. behavioral emotion, is expectation-based and corresponds with ‘behavioral design’ that is concerned with the pleasure and effectiveness of use. The third type, i.e. reflective emotion, is intellectually based and corresponds with ‘reflective design’ that is concerned with self-image, personal satisfaction, and memories. Similar to the pleasure approach, the process-level approach involves several distinct, theoretically independent, sources of product affect. Whereas Jordan distinguishes responses on the basis of differentiated needs, Norman distinguishes responses on the basis of levels of processing in the brain. Although in that sense different, the types of affect seem to correspond at least to some extent: Visceral affect corresponds with physio-pleasure, behavioral emotion with psycho-pleasure, and reflective emotion with socio- and ideo-pleasure. Although the cursory translation to three types of emotional design (i.e. visceral, behavioral, and reflective design) overly reduces the complex nature of design and design activities, the distinction between three levels of processing with each associated affective phenomena (simple affect, primitive, and complex emotions) is an important contribution to the design and emotion discourse, because it clarifies and illustrates the role of cognition in the process of product emotion, and provides us with a basis for explaining why and how products elicit emotional responses. The process-level approach differs with the appraisal approach in what types of responses are considered to be elicited by appraisal processes. Whereas Norman argues that appraisal is involved only in the reflective level, many contemporary appraisal theorists propose that appraisal does not necessarily involve high-level cognitive processing, and therefore is not only involved in reflective emotions, but also in reactive and behavioral emotional responses (see e.g. Lazarus, 1991a). In following these models, the model of product emotions proposed next includes appraisals that generate responses at all three levels of processing.

389

Product emotion

3.3. Appraisal approach to product emotion In the tradition of the cognitive theories of emotion that are discussed by, for example, Lazarus (1991a), Frijda (1986), Ortony et al. (1988), Scherer (2001), and Smith and Ellsworth (1987), Desmet (2002) introduced a basic model of product emotions, shown in Figure 15.2. The model is ‘basic’ because it applies to all possible emotional responses elicited by (buying, using, or owning) products, and because it identifies three key variables in the process of emotion eliciting: (1) concern; (2) stimulus; and (3) appraisal. The proposition that the function of emotion is to motivate adaptive behavior in response to one’s circumstances implies that the eliciting conditions must involve some evaluation of the event’s significance for this person’s well-being. This evaluation is often referred to as ‘appraisal’ to signify that it is non-intellectual and automatic rather than conscious and deliberate (Arnold, 1960). A product appraisal has three basic possible outcomes: The product is (potentially) beneficial; harmful; or not relevant for personal well-being. These three general outcomes result in a pleasant emotion, an unpleasant emotion, or the absence of an emotion, respectively. The point of reference in the appraisal process is a concern, that is, a more or less stable preference for certain states of the world (Frijda, 1986). The third variable is the stimulus, which, in the case of product emotion, is not necessarily the product itself. The emotion can also be elicited by an event related to the product, such as a consequence of the product, the behavior of the product in interaction, or an associated object or person, like the manufacturer or the user. The basic model indicates that an emotion is not elicited by the product as such, but by an appraised concern match or mismatch, and that the study of product emotions requires an understanding of the concerns that underlie them. Each of the three variables is discussed in more detail below. The next section (4) provides examples of how these variables combine to elicit product emotions. Appraisal According to appraisal researchers, all emotions are preceded and elicited by an appraisal (Roseman, 1991). Appraisal is an evaluative process that serves to ‘diagnose’ whether a situation confronting an individual has adaptational relevance and, if it does, to identify the nature of that relevance and produce an appropriate emotional response to it (Lazarus, 1991b). One who is confronted with a fire alarm will most likely experience fear with a corresponding tendency to flee because the fire alarm signals a potentially Expression action tendency arousal feeling

Emotions

Appraisal

Latent/active Attitudes goals standards

Concern

Actual/associated Object

Product interaction consequence

FIGURE 15.2 Basic model of product emotions (adapted from Desmet, 2002).

390

Product Experience

harmful situation with particular behavioral consequences. This example illustrates that appraisals are inherently relational (e.g. Scherer, 1984). Rather than exclusively reflecting either the properties of the stimulus (e.g. a fire), the situation (e.g. the office), or the person (e.g. asthmatic condition), appraisal represents an evaluation of the properties of the stimulus and the situation as it relates to the properties of the individual (Smith and Lazarus, 1990). In short, appraisal is an evaluation of the significance of a stimulus for one’s personal well-being. It is this personal significance of a product, rather than the product itself, which causes the emotion. Because appraisal mediates between products and emotions, different individuals who appraise the same product in different ways will experience different emotions. One who is stressed may respond with irritation to the ring tone of his or her mobile phone because he or she appraises it as undesirable, whereas another person may appraise the same event as desirable. Concerns Emotions arise from encounters with events that are appraised as having beneficial or harmful consequences for the individual’s concerns, that is, his or her major goals, motives, well-being, or other sensitivities (Frijda, 1986; Lazarus, 1991a). Concerns are the dispositions that we bring into the emotion process, and stimuli are construed as emotionally relevant only in the context of one’s concerns (Lazarus, 1991a). In order to understand emotional responses to consumer products, one must understand the users’ concerns given the context in which the product is or will be used. The number and variety of human concerns is endless. Types of concerns reported in the research literature are, for example, drives, needs, instincts, motives, goals and values (for a discussion, see Scherer, 2001). Some concerns, such as the concern for safety and the concern for love, are universal, while others are context-dependent, such as the concern for being home before dark or the concern for securing a good seat for your friend at the cinema. Concerns can be latent (sleeping) or active (awake). Concerns are asleep as long as the circumstances pose no threat or possibility to their fulfillment. The concern of physical well-being, for example, will stay asleep until something threatens the physical well-being, like an upcoming headache, or a news item warning for the risk of eating unhealthy food. Sleeping concerns are latent in the sense that they only come to the foreground when the factual circumstances differ from satisfaction conditions. The concern of flexible communication can be awakened when one realizes that one’s mobile phone is missing. Our implicit expectations of how products should function are sleeping concerns that are awakened if the product does not meet that expectation (see Chapter 16). One expects, for example, that a door lock will not unlock spontaneously. A lock that meets this implicit expectation will not rouse an emotion. A lock that does not, however, will awaken the concern of security and elicit an unpleasant emotion for being appraised as insecure. Stimulus According to Frijda (1986) any perceived change, or event, has the potential to elicit an emotion. This can be someone approaching us, a car passing at high speed, the sound of a vase breaking, etc. In the case of products, the stimulus ‘event’ can be: Perceiving the product, using the product, and the consequences of (using) the product. Perceiving the product is the most straightforward stimulus event. Seeing, touching, hearing, and smelling an object can be a strong emotional stimulus. The second type of stimulus event is product usage. Using a product is an event in itself, which can be separated in many sub-events of action and reaction of both user and product. Each of these events can be an emotional stimulus. For example, the TV doesn’t respond to the remote control, the oven starts to produce a scent of freshly baked cookies, the alarm clock sets of, a drawer

Product emotion

391

runs unexpectedly smoothly, and so on. The third type of stimulus event is the consequences of (using the) product. The consequence of wearing a fashionable new suit can be positive remarks of colleagues; the consequence of using a laptop can be that the work is done more efficiently; the consequence of eating too much ice-cream can be a tummy ache. Each of these consequences can act as emotional stimuli. Note that the absence of an expected consequence can also elicit an emotional response. Those who expect a friend for dinner, will be disappointed when the friend does not show up, and those who buy an auto-bronzing lotion, will be dissatisfied when the product does not tan their skin. The model in Figure 15.2 distinguishes between actual and associated stimuli. This is an important distinction because not only events that actually happen can elicit emotions, but also events that are remembered, imagined, anticipated, dreamed, or fictional (Ekman, 1994a). Imagining what would happen if I arrive late for an important meeting is sufficient to arouse an emotion. Or, remembering a joyful event can be sufficient to experience joy all over again. This distinction between actual and associated stimuli also applies to products. Products can represent personal meaning (e.g. souvenirs and nostalgia), or cultural meaning (e.g. symbols and branding). One can, for example, be attracted to a record player because it represents a childhood memory, or feel contempt towards a pair of shoes because the brand is associated with hooligans. The same applies to product usage. One can anticipate usage events (‘I expect this handle to break when I push it too hard’), or fantasize about usage events (‘my computer thinks he knows what I want, but he does not have a clue’). One can also imagine, anticipate, or fantasize about possible consequences of usage. For instance, a person may feel desire towards a new abdominal work-out device because he or she anticipates that with this device the perfect body is within reach.

4. SOURCES OF PRODUCT EMOTION The key factor of an appraisal perspective on the elicitation and differentiation of emotion is the assumption that people constantly evaluate (actual or imagined) stimulus events for their personal significance. This significance is operationally defined by a number of evaluative issues or criteria that constitute the meaning structure in which the evaluation takes place (Scherer, 2001). In this view, the overall outcome of the appraisal process is the combined outcome of these evaluative issues. Each of these issues is addressed by a particular ‘appraisal check’. In the case of products, the appraisal checks relate to issues such as: Does this product help me to attain my ambitions? Can I afford it? Will my neighbors approve? Is it safe? Am I successful in operating it? etc. Many appraisal researchers have developed models that attempt to specify the evaluative issues that initiate specific emotions (see, e.g. Frijda, 1986; Lazarus, 1991a; Roseman, 1991; Scherer, 1988; Smith and Ellsworth, 1987). These distinct evaluative issues are in turn associated with distinct types of concerns. Emotion theorists often distinguish between three concern types: Attitudes; goals; and standards (see e.g. Ortony et al., 1988). This concern distinction has been found effective for understanding product emotion (see, Desmet, 2002). The different types of concerns and corresponding appraisal checks can be combined with different types of stimuli to represent various sources of product emotions. Figure 15.3 visualizes how the three concern types combine with three levels of product stimulus to nine sources of product emotion. These nine sources represent combinations of the variables in the basic model in Figure 15.2, and can be used as a framework for understanding product emotion. The columns represent three types of concerns with associated appraisal checks, and the

392

Product Experience

Attitudes

Goals

Standards

Product

Attracted to the sensuous shape of this product

Desire for owning a mobile phone

Admiring the designer for making an innovative design

Usage

Enjoying the gestures required for making an espresso

Frustrated at not being able to set the timer of a DVD recorder

Being angry with the product for breaking down

Consequence

Being inspired by an art gallery visit

Satisfied by increase in health resulting from using steam pan

Being proud of weight loss because of product

FIGURE 15.3 Nine sources of product emotion. The horizontal axis shows different types of stimuli, and the vertical axis shows different types of concerns.

rows represent three types of stimulus events that can evoke product emotions. Note that although many emotional responses can be explained with these nine sources, it is not claimed that these represent all possible sources of product emotions.

4.1. Product emotion related to affect dispositions The first source of product emotion is represented by an appraisal check of intrinsic pleasantness: How pleasant or unpleasant is this stimulus event? This appraisal is an evaluation of whether a stimulus is pleasurable or painful (or whether a stimulus is likely to result in pleasure or pain), and determines the fundamental pleasure response: Liking feelings, generally encouraging approach, versus disliking feelings, leading to withdrawal or avoidance. The concerns that serve as point of reference in this appraisal are our attitudes, which are the affect dispositions that were discussed in a previous section of this chapter. We have attitudes towards product types (‘I don’t like microwave ovens’) aspects or features of products (‘I like red cars’), towards style (‘I like Italian design’), towards usage (‘I like cars that have a firm drive’), and towards consequences of products (‘I like feeling relaxed after drinking a beer’). The intrinsic pleasantness check applies to all three stimulus levels. In the case of the product level, the stimulus is the (visual, tactile, olfactory, auditory, gustatory) manifestation of a product. Products are objects, and all objects (including their properties and features) are appraised as pleasant or unpleasant. As a result, one is attracted to a sensuous shape of a perfume bottle, feels aversion towards the off-colored leather suitcase, or enjoys the taste of sweet and cold ice-cream. The second stimulus level, that is, using the product, can also involve sensations that are appraised as pleasant or unpleasant. The gestures that are required to operate an espresso machine, the expressive movements of playing the violin, and the forces that are felt when driving a motor cycle are appraised as intrinsically pleasant or unpleasant. In those cases the act of using the product, rather than the product as such, generates sensations that are pleasing or displeasing. The third stimulus level represents those emotions that are elicited by the consequence of (using or owning) the product. These consequences can also be intrinsically pleasant or unpleasant. For example, the consequence of eating too much cake is an unpleasant feeling in the stomach, the consequence of looking at art is a pleasant feeling of inspiration, and the consequence of using a massage device is the pleasant feeling of relaxation.

Product emotion

393

Emotions evoked by an intrinsic pleasantness check have in common that they are elicited independent of the motivational state of the person (i.e. particular goals or motives), and are related only to attitudes. Some researchers refer to these emotions as ‘aesthetic emotions’ because they are elicited by sensations that are pleasing or displeasing to our senses (see Lazarus, 1991b). In the design research literature these experiences are sometimes not considered to be emotions (see e.g. Hekkert, 2006; Chapter 10 this volume), but rather a separate type of product experience. This corresponds with the process-level approach of Norman (2004) that identifies these experiences as simple affect (at the reactive level of information processing). Note however that this source of product emotion also represents emotions that are elicited by associated meanings of the product. Many people will be disgusted by the appearance of a swastika sign; not because the appearance of the sign is displeasing to the senses but because it represents values, memories, or ideas that are appraised as intrinsically unpleasant. In the same way, one may experience attraction towards a particular object (brand, company, color, etc.) because it symbolizes or represents something that is appraised as pleasant. These experiences involve higher-level cognitive processes that Norman associates with reflective emotions rather than with simple affect.

4.2. Product emotion related to goals The second source of product emotion is represented by an appraisal check of motive consistency: To what degree does this stimulus event help me to attain my goals? Goals are the things one wants to get done and the things one wants to see happen. The goals that people try to satisfy are often assumed to be structured in a hierarchy ranging between abstract goals or aspirations, like the goal to have a successful life, and goals as concrete and immediate as the goal to catch a train. Many goals are directly and indirectly activated in the human-product relationship. For example, we buy, own, and use products because we believe they can help us to achieve things (a digital agenda to make us more organized), or they fulfill a need (a bicycle fulfills the need for transportation). The more directly products facilitate goal attainment, and the closer they propel the person toward reaching a goal, the higher the appraised motive consistency (Ellsworth and Scherer, 2003). In other cases, products can be obstructive for goal attainment, by putting goal satisfaction out of reach, delaying its attainment, or requiring additional effort (see Srull and Wyer, 1986). In the case of the first stimulus level, i.e. the product level, the goal at stake involves the product as such, which can be, for example, the goal to own a particular product. Other examples of goals that involve products are the goal to share, personalize, restore, discard, or repair a product. If a person has the goal to own a particular pair of shoes, this person will be disappointed to hear that the model is out of stock. One who is interested in owning an original model T-Ford will be happily surprised when accidentally bumping into one in some old barn. The motive consistency appraisal check also applies to using the product. When using products, people are involved in goal-directed behavior sequences. If this sequence is blocked in the interaction, people will typically experience frustration. One can be frustrated by product packages that are impossible to open (e.g. pre-packed slices of cheese), satisfied with products that are easy to operate (e.g. the selfexplaining Apple i-Pod interface), and pleasantly surprised by the accurate response of a stereo set’s volume to the remote control. The third stimulus level, the consequence of (using or owning) the product, may be appraised as motive consistent (e.g. being satisfied by an alarm clock because one is not late for work), or motive inconsistent (e.g. being dissatisfied with the new mattress because it increases instead of decreases the backache). Emotions are not only experienced in response to actual goal achievement, but also in response to anticipated goal achievement. One can be attracted to a particular computer

394

Product Experience

because one anticipates that this computer will satisfy the goal of productivity. An additional appraisal check involved in emotions elicited by anticipated events is the probability check (Scherer, 1984). This check assesses the likelihood of possible effects. This is of particular importance in cases in which both the probability of the event occurring or its consequences are in doubt, as in the case of emotions like hope and fear.

4.3. Product emotion related to standards The third source of product emotion identified in Figure 15.3 is represented by an appraisal check of legitimacy: How praiseworthy is the stimulus event? The concerns involved in this appraisal check are our standards. Standards are our beliefs, social norms or conventions of how we think things should be. Whereas goals refer to the state of affairs we want to obtain, standards are the states of affairs we believe ought to be (Frijda, 1986; Ortony et al., 1988; Scherer, 1984). For example, many of us believe that we should respect our parents, or wear clean clothes at work. Most standards are socially learned and represent the beliefs in terms of which moral and other kinds of judgmental evaluations are made (Ortony et al., 1988). Whereas goals are relevant for our personal well-being, standards are relevant for the preservation of our social structures (and thus indirectly also for our personal well-being). Social organization depends on shared rules (norms) about what behavior is acceptable and what is unacceptable. Such norms are sustained by appropriate emotional reactions of group members to behavior that violates norms, as well as to conforming behavior. We approve of things that comply with standards and disapprove of things that conflict with them. We not only have standards regarding human (inter)action, but also regarding products. In the case of the first stimulus level, the legitimacy of the product itself is at stake. We have standards of how products should be, and how they should be designed and produced. One can admire a chair for being more eco-friendly than a conventional chair. Or one can be irritated by a new car model because automotive companies should not introduce new versions too often. In those cases the products are appraised as the outcome of the action of some person or institute, and that particular action is appraised as either legitimate or as improper. The second stimulus level involves standards of performance, that is, standards of how products should behave when they are used. For instance, one shouldn’t hear a rattling sound when driving a brand new car, a computer should not crash without a warning, and the brakes of a bicycle should be reliable. One can experience emotions such as anger or disappointment when a product does not meet the standards of performance. The third level is related to the consequence of owning or using the product. Discrepancy with standards of ownership can lead to emotions such as contempt, and exceeding the standards may produce admiration. One can admire one’s perfectly mown lawn, a consequence of using a high quality lawn mower, or one can feel contempt towards some person’s run down car. One can also appraise the legitimacy of one’s own behavior with reference to internal standards, one’s internalized moral code or self-concept. These standards represent the self-ideal and are central for the experience of the so called self-reflexive emotions, such as pride, guilt, and shame. One can be ashamed of owning sex toys or proud of owning a signed baseball.

4.4. Mixed emotions The distinction between various sources of product emotion enables us to explain why products sometimes elicit mixed emotions. First, mixed emotions may be elicited within a particular source. One can experience mixed emotions in response to events that are

Product emotion

395

consistent with one goal and obstructing another (Weigert, 1991). Buying a digital agenda may correspond with the goal to be more time efficient, but at the same time conflict with the goal to be independent of digital devices. The same applies to all other sources of emotion. One may appraise the color of a product as pleasant, and, at the same time, the tactile quality as unpleasant. One may appraise the innovativeness of a product as legitimate, and, at the same time, the high price as illegitimate. Secondly, mixed emotions may be generated by emotional responses elicited by different types of sources. For example, the intrinsic pleasantness check is independent of the motivational state of the person, whereas motivational state is the decisive element in the goal conduciveness check. We all know from experience that an inherently pleasant product can block goal achievement, since something pleasant (like chocolate cake) can obstruct us in reaching a goal (trying to lose weight). One can experience shame from using a rollator (standard of being independent), and at the same time, be satisfied with the increased mobility it provides (goal of being mobile). Note that our emotional responses are not mutually independent because our attitudes, goals, and standards are related to each other. One can have a favourable attitude towards the color red (appraise a red product as intrinsically pleasant), can have the goal to own a red car (appraise a red car as motive consistent), and may have the standard that cars of a particular brand should be available in the color red (appraise a brand that does not offer red cars as illegitimate). Our concerns are the dispositions that we bring into the emotion process. Although dispositional, these concerns do evolve over time, and differ in different contexts. Children have different goals, attitudes, and standards than teenagers and adults. Concerns are not independent of the context of usage. For example, people have other concerns with respect to computers in the context of work than in a family context. In addition, there are many factors that have a constant influence on our concerns, such as marketing, technological innovation, peer group behavior, and fashion. The standard of performance for a laptop computer today differs from the one we had two years ago. Attitudes towards floral prints on clothes change with the season, the attitude towards smoking cigarettes is influenced by anti-smoking campaigns, etc. The product emotion system is not static but dynamic and interactive. Similar to the pleasure framework of Jordan (2000), the main value of the framework is that it can be used as a tool that helps in taking a structured approach to design for emotion. It represents some of the important variables in the process of product emotion, and can be used as a means of structuring thoughts and discussion as regard to emotion, or as a means of formulating relevant questions about the user or context of usage. One can explore the variables with the aim to generate positive outcomes on as many appraisal checks as possible. Alternatively, one can also explore these variables in order to deliberately design products that elicit mixed emotions by, for example, confronting its user with his or her sometimes contradictory concerns. The palette of emotions that can be experienced in response to products displays many different varieties: Some mild, some intense, some cursory, and some long lasting. It includes immediate emotions that relate directly to the interaction or context of interaction, like anger, confusion, contentment, enjoyment, and aversion, perceptible at the surface of our emotional experience, expression, and behavior. And the palette also includes emotions that are more indirect; those that operate at a deeper level, invisible and less explosive, but no less oppressive and influential. These are emotions like trust, resignation, compassion, empathy, melancholy, hope, and consolation. The framework of product emotion introduced in this chapter does not explain all these possible emotional responses, and future research may reveal additional types of concerns, types of stimuli, or dimensions that enable us to explain a wider variety of emotional responses. Even so,

396

Product Experience

the basic framework of product emotion does illustrate that we can identify relationships between our highly personal and subjective emotional responses, and some objective and universal principles in the processes that elicit them.

REFERENCES Aaker, D. A., Stayman, D. M. and Hagerty, M. R. (1986). Warmth in advertising: Measurement, impact, and sequence effects. Journal of Consumer Research, 12, 365–381. Arnold, M. B. (1960). Emotion and personality, volume 1: Psychological aspects. New York: Colombia University Press. Bradley, M. M. and Lang, P. J. (1994). Measuring emotion – the self-assessment mannequin and the semantic differential. Journal of Behavior Therapy and Experimental Psychiatry, 25(1), 49–59. Clore, G. L. and Ortony, A. (1988). The semantics of the affective lexicon. In: V. Hamilton, G. H. Bower and N. H. Frijda (Eds.) Cognitive perspectives on emotion and motivation, pp. 367–397. Dordrecht: Kluwer Academic Publishers. Darwin, C. (1872). The expression of the emotions in man and animals. London: Murray. Davidson, R. J. (1994). On emotion, mood, and related affective constructs. In: P. Ekman and R. J. Davidson (Eds.) The nature of emotion, fundamental questions, pp. 51–56. Oxford: Oxford University Press. Demir, E., Desmet, P. M. A. and Hekkert, P. (2006). Experiential concepts in design research; a (not too) critical review. In: M. A. Karlsson, P. M. A. Desmet and J. van Erp (Eds.) Proceedings of the 5th conference on design and emotion, September 27–29. Gothenburg, Sweden: Chalmers University of Technology. Desmet, P. M. A. (2002). Designing emotions. Unpublished doctoral thesis, Delft University of Technology. Ekman, P. (1994a). Moods, emotions, and traits. In: P. Ekman and R. J. Davidson (Eds.) The nature of emotion, fundamental questions, pp. 56–58. Oxford: Oxford University Press. Ekman, P. (1994b). Strong evidence for universals in facial expressions – a reply to Russell’s mistaken critique. Psychological Bulletin, 115(2), 268–287. Ekman, P. and Friesen, W. V. (1978). Facial Action Coding System. Palo Alto: Consulting Psychologists Press. Ellsworth, P. C. and Scherer, K. R. (2003). Appraisal process in emotion. In: R. J. Davidson, K. R. Scherer and H. H. Goldsmith (Eds.) Handbook of affective sciences, pp. 575–595. Oxford: Oxford University Press. Etcoff, N. (1999). Survival of the prettiest: The science of beauty. New York: Doubleday. Frijda, N. H. (1986). The emotions. Cambridge: Cambridge University Press. Frijda, N. H. (1993). Appraisal and beyond. Cognition and Emotion, 7, 225–231. Frijda, N. H. (1994a). Emotions are functional, most of the time. In: P. Ekman and R. J. Davidson (Eds.) The nature of emotion, fundamental questions, pp. 112–122. Oxford: Oxford University Press. Frijda, N. H. (1994b). Emotions require cognitions, even simple ones. In: P. Ekman and R. J. Davidson (Eds.) The nature of emotion, fundamental questions, pp. 197–202. Oxford: Oxford University Press. Hekkert, P. (2006). Design aesthetics: Principles of pleasure in product design. Psychology Science, 48, 157–172. Jordan, P. W. (1999). Pleasure with products: human factors for body, mind and soul. In: W. S. Green and P. W. Jordan (Eds.) Human factors in product design: Current practice and future trends, pp. 206–217. London: Taylor and Francis. Jordan, P. W. (2000). Designing pleasurable products. London: Taylor and Francis. Laurans, G. and Desmet, P. M. A. (2006). Using self-confrontation to study user experience. In: M. A. Karlsson, P. M. A. Desmet and J. van Erp (Eds.) Proceedings of the 5th conference on design and emotion, September 27–29. Gothenburg, Sweden: Chalmers University of Technology. Lazarus, R. S. (1991a). Emotion and adaptation. Oxford: Oxford University Press. Lazarus, R. S. (1991b). Progress on a cognitive motivational relational theory of emotion. American Psychologist, 46, 819–834. Norman, D. A. (2004). Emotional design. New York: Basic Books. Ortony, A., Clore, G. L. and Collins, A. (1988). The cognitive structure of emotions. Cambridge: Cambridge University Press. Ortony, A., Norman, D. A. and Revelle, W. (2005). The role of affect and proto-affect in effective functioning. In: J. M. Fellous and M. A. Arbib (Eds.) Who needs emotions: The brain meets the machine. Oxford: Oxford University Press. Plutchik, R. (1980). Emotion: A psychobioevolutionary synthesis. New York: Harper and Row. Plutchik, R. (1984). Emotions: A general psychoevolutionary theory. In: K. R. Scherer and P. Ekman (Eds.) Approaches to emotion, pp. 197–219. Hillsdale, NJ: Lawrence Erlbaum.

Product emotion

397

Roseman, I. J. (1991). Appraisal determinants of discrete emotions. Cognition and Emotion, 5, 161–200. Rozin, P. and Fallon, A. (1987). A perspective on disgust. Psychological Review, 94, 23–41. Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39, 1161–1178. Russell, J. A. (1993). Force-choice response format in the study of facial expression. Motivation and Emotion, 17, 41–51. Russell, J. A. (2003). Core affect and the psychological construction of emotion. Psychological Review, 110, 145–172. Scherer, K. R. (1984). On the nature and function of emotion: a component process approach. In: K. R. Scherer and P. Ekman (Eds.) Approaches to emotion, pp. 293–318. Hillsdale, NJ: Lawrence Erlbaum. Scherer, K. R. (1988). Criteria for emotion-antecedent appraisal: a review. In: V. Hamilton, G. H. Bower and N. H. Frijda (Eds.) Cognitive perspectives on emotion and motivation, pp. 89–126. Dordrecht: Nijhoff. Scherer, K. R. (2001). Appraisal considered as a process of multi-level sequential checking. In: K. R. Scherer, A. Schorr and T. Johnstone (Eds.) Appraisal process in emotion: theory, methods, research, pp. 92–120. New York: Oxford University Press. Schubert, E. (1996). Measuring temporal emotional response to music using the two dimensional emotion space. Proceedings of the 4th International Conference for Music Perception and Cognition, pp. 263– 268. Montreal, Canada. Schwarz, N. and Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: informative and directive functions of affective states. Journal of Personality and Social Psychology, 45, 513–523. Smith, C. A. and Ellsworth, P. C. (1987). Patterns of appraisal and emotion related to taking an exam. Journal of Personality and Social Psychology, 52, 475–488. Smith, C. A. and Kirby, L. D. (2001). Toward delivering the promise of appraisal theory. In: K. R. Scherer, A. Schorr and T. Johnson (Eds.) Appraisal processes in emotion, pp. 121–140. Oxford: Oxford University Press. Smith, C. A. and Lazarus, R. S. (1990). Emotion and adaptation. In: L. A. Pervin (Ed.) Handbook of personality: Theory and research, pp. 609–637. New York: Guilford. Srull, T. S. and Wyer, R. S., Jr. (1986). The role of chronic and temporary goals in social information processing. In: R. M. Sorrentino and E. T. Higgins (Eds.) Handbook of motivation and cognition, pp. 503–549. New York: Wiley. Tiger, L. (1992). The pursuit of pleasure. Boston: Little Brown. Weigert, A. J. (1991). Mixed emotions: Certain steps toward understanding ambivalence. Albany: State University of New York Press. Wensveen, S. A. G. (2005). A tangibility approach to affective interaction. Unpublished doctoral thesis, Delft University of Technology. Wundt, W. (1905). Fundamentals of psychology (7th Ed.) Leipzig: Engelman. Zajonc, R. B. (2000). Feeling and thinking: closing the debate over the independence of affect. In: J. P. Forgas (Ed.) Feeling and thinking: The role of affect in social cognition, pp. 31–58. New York: Cambridge University Press.