COGNITION AND EMOTION 2009, 23 (7), 13071351

The dynamic architecture of emotion: Evidence for the component process model
Klaus R. Scherer
Swiss Centre for Affective Sciences, University of Geneva, Geneva, Switzerland

Emotion is conceptualised as an emergent, dynamic process based on an individuals subjective appraisal of significant events. It is argued that theoretical models of emotion need to propose an architecture that reflects the essential nature and functions of emotion as a psychobiological and cultural adaptation mechanism. One proposal for such a model and its underlying dynamic architecture, the component process model, is briefly sketched and compared with some of its major competitors. Recent empirical evidence in support of the model is reviewed. Special emphasis is given to the dynamic aspect of emotion processes, in particular the sequence of appraisal checks and the synchronisation of response systems, as well as the capacity of the model to predict individual differences in emotional responding.

Keywords: Emotion theories; Component processes; Appraisal; Response synchronization; Emotion categories and words.

Although there is now increasing consensus on a componential approach to emotion and the need to consider appraisal as one of the central underlying mechanisms (Frijda, 2007a, 2007b; Scherer, 2005b), four central design features of emotions often receive little more than lip service and continue to be seriously under researched: (i) the dynamic, recursive nature of emotional processing; (ii) the central, and causal, role of multilevel cognitive processing of both antecedent events and response options; (iii) the important distinction between the emotion episode as a multicomponent process as a whole and one of its components: subjective feeling or conscious emotional experience; and (iv) the essential role of individual differences in both
Correspondence should be addressed to: Klaus R. Scherer, Swiss Centre for Affective Sciences, University of Geneva, 7, Rue des Battoirs, 1205 Geneva, Switzerland. E-mail: Klaus.Scherer@unige.ch Preparation of this paper was supported by the Swiss National Science Foundation through the National Center for Competence in Research in the Affective Sciences grant. The author acknowledges precious comments and suggestions by Tatjana Aue, Geraldine Coppin, Didier Grandjean, and Sylvia Kreibig. # 2009 Psychology Press, an imprint of the Taylor & Francis Group, an Informa business www.psypress.com/cogemotion DOI: 10.1080/02699930902928969

1308

SCHERER

cognitive event appraisal and response dispositions. In recent years, it has been mostly the protagonists of appraisal theories (Ellsworth & Scherer, 2003; Roseman & Smith, 2001; Scherer, 1999a; Schorr, 2001) who have attempted to model these design features in a principled fashion. One of the theoretical models in that tradition, the component process model (CPM) of emotion, specifically focuses on the dynamic unfolding of emotion (Scherer, 1984a, 1984b, 1986, 2001, 2004). This contribution aims to demonstrate the utility of this model for our understanding of individual differences in emotion components and dynamics and to review the empirical evidence that has accumulated over the years in support of the models predictions.

DESCRIPTION OF THE CPM

Figure 1 shows the architecture of the model, including the dynamic, recursive emotion processes following an event that is highly pertinent to the needs, goals, and values of an individual. An important feature of the model is that it does not include overt instrumental behaviour. In line with the dominant functional or adaptational approach pioneered by Darwin (1872/ 1998), emotion is seen as a reaction to significant events that prepares action readiness and different types of alternative, possibly conflicting, action tendencies but not as a sufficient cause for their execution (see also Frijda, 2007b). The assumption is that even highly emotional behaviours such as aggression or flight are prepared by emotions such as anger or fear but that their execution is multiply determined, with emotion being only one, albeit potentially important, factor. As shown in the flow diagram, the CPM suggests that the event and its consequences are appraised with a set of criteria on multiple levels of processing (the appraisal component). The result of the appraisal will generally have a motivational effect, often changing or modifying the

Figure 1. The dynamic architecture of the component process model.

DYNAMIC ARCHITECTURE OF EMOTION

1309

motivational state before the occurrence of the event. Based on the appraisal results and the concomitant motivational changes, efferent effects will occur in the autonomic nervous system (e.g., in the form of cardiovascular and respiratory changes) and in the somatic nervous system (in the form of motor expression in face, voice, and body). All of these components, appraisal results, action tendencies, somatovisceral changes, and motor expressions are centrally represented and constantly fused in a multimodal integration area (with continuous updating as events and appraisals change). Parts of this central integrated representation may then become conscious and subject to assignment to fuzzy emotion categories as well as being labelled with emotion words, expressions, or metaphors. As recent descriptions of the model can be found elsewhere (Scherer, 2001, 2004, 2005a), in the following section only those elements directly relevant to the focus on individual differences and dynamic unfolding of emotion processes are discussed in detail. To aid the comprehension of the reader and to allow interpretation of some of the figures, Table 1 synthetically recapitulates some of the central elements of the CPM. As shown in the table, the model suggests that there are four major appraisal objectives to adaptively react to a salient event: (a) How relevant is this event for me? Does it directly affect me or my social reference group? (relevance); (b) What are the implications or consequences of this event and how do they affect my well-being and my immediate or long-term goals? (implications); (c) How well can I cope with or adjust to these consequences? (coping potential); (d) What is the significance of this event for my self-concept and for social norms and values? (normative significance). To attain these objectives, the organism evaluates the event and its consequences on a number of criteria or stimulus evaluation checks (SECs; shown in column 1 of Table 1), with the results reflecting the organisms subjective assessment (which may well be unrealistic or biased) of consequences and implications on a background of personal needs, goals, and values (see Ellsworth & Scherer, 2003; Sander, Grandjean, & Scherer, 2005; Scherer, 2001, for further details and references). It is important to note that the appraisal process does not necessarily require a complex cognitive calculus but often occurs in an automatic, unconscious, and effortless fashion. A powerful example is the appraisal process underlying emotional attention. Recent research using functional brain imaging in human subjects has revealed low-level neural substrates by which sensory processing and attention can be modulated by the affective significance of stimuli. The amygdala plays a crucial role in providing both direct and indirect top-down signals on auditory and visual sensory pathways, which can influence the representation of emotional events, especially when related to threat (Grandjean et al., 2005; Pourtois, Grandjean, Sander, & Vuilleumier, 2004; Vuilleumier, 2005).

1310

TABLE 1 Synthetic recapitulation of central elements of the component process model (CPM) of emotion Stimulus Evaluation Checks (SECs) Organismic/Social functions Component patterning

SCHERER

Relevance (A stimulus event is considered as requiring attention deployment, further information processing, and potential action) Novelty (Abrupt onset, familiarity, predictability) Goal relevance (Does the event have consequences for my needs or goals?)

Novel and goal relevant: Orienting, Focusing/ Alerting

Orienting response; EEG alpha changes, modulation of the P3a in ERPs; heart rate deceleration, vasomotor contraction, increased skin conductance responses, pupillary dilatation, local muscle tonus changes; brows and lids up, frown, jaw drop, gaze directed; interruption of speech and action, raising head (possibly also preparatory changes for subsequent effort investment given relevance appraisal at this stage, in particular increased cardiac contractility as indicated by, e.g., decreased pre-ejection period) Sensitisation; inhalation, heart rate deceleration, salivation, pupillary dilatation; lids up, open mouth and nostrils, lips part and corners pulled upwards, gaze directed; faucal and pharyngeal expansion, vocal tract shortened and relaxation of tract walls (wide voice*increase in low frequency energy, F1 falling, slightly broader F1 bandwidth); centripetal hand and arm movements, expanding posture, approach locomotion Defence response, heart rate acceleration, increase in skin conductance level, decrease in salivation, pupillary constriction; slight muscle tonus increase; brow lowering, lid tightening, eye closing, nose wrinkling, upper lip raising, lip corner depression, chin raise, lip press, nostril compression, tongue thrust, gaze aversion; faucal and pharyngeal constriction, vocal tract shortened and tensing of tract walls (narrow voice*more high frequency energy, F1 rising, F2 and F3 falling, narrow F1 bandwidth, laryngopharyngeal nasality, resonances raised); centrifugal hand and arm movements, hands covering orifices, shrinking posture, avoidance locomotion

Intrinsic pleasantness (Is the Pleasant: event intrinsically pleasant or Incorporation/ unpleasant, independently of Recommending my current motivational state?) Unpleasant: Rejection/Warning

TABLE 1 (Continued) Stimulus Evaluation Checks (SECs) Organismic/Social functions Component patterning

Implications (Following attention deployment, the pertinent characteristics of the stimulus event and its implications or consequences for the organism are determined) Outcome probability (How likely is it that the consequences will occur?) Discrepancy from expectation (How different is the situation from what I expected it to be?) Conduciveness (Is the event Conducive: Relaxation/ conducive or obstructive to Stability reaching my goals?)

DYNAMIC ARCHITECTURE OF EMOTION

Trophotropic shift, rest and recovery; decrease in respiration rate, slight heart rate decrease, bronchial constriction, increase in gastrointestinal motility, relaxation of sphincters; decrease in general muscle tone; relaxation of facial muscle tone; overall relaxation of vocal apparatus (relaxed voice*F0 at lower end of range, low-tomoderate amplitude, balanced resonance with slight decrease in high-frequency energy; comfort and rest positions; plus elements from pleasantness response (however, if a conduciveness appraisal is accompanied by plans for further action, an ergotropic shift is to be expected) Ergotropic shift, preparation for action; corticosteroid and catecholamine, particularly adrenaline secretion; deeper and faster respiration, increase in heart rate and heart stroke volume, vasoconstriction in skin, gastrointestinal tract, and sexual organs, vasodilatation in heart and striped musculature, increase of glucose and free fatty acids in blood, decreased gastrointestinal motility, sphincter contraction, bronchial dilatation, contraction of m. arrectores pilorum, decrease of glandular secretion, increase in skin conductance level, pupillary dilatation strongly increased muscular tonus; frown, lids tighten, lips tighten, chin raising; gaze directed; overall tensing of vocal apparatus (tense voice*F0 and amplitude increase, jitter and shimmer, increase in high frequency energy, narrow F1 bandwidth, pronounced formant frequency differences); strong tonus, task-dependent instrumental actions; plus elements of unpleasantness response

Obstructive: Activation/ Reactivity

1311

1312

TABLE 1 (Continued) Stimulus Evaluation Checks (SECs) Organismic/Social functions Component patterning

SCHERER

Urgency (How urgently do I need to react?) Coping potential (Once nature of event and consequences are known sufficiently well, organism checks its ability to cope with the consequences to be expected) Agent and intention (Who was responsible and what was the reason?) Control (Can the event or its No or low control: consequences be controlled Readjustment/Withdrawal by human agents?)

Trophotropic dominance; decrease in respiration rate and depth, heart rate decrease, increase in glandular secretion, particularly tear glands, bronchial constriction; hypotonus of the musculature; lip corner depression, lips parting, jaw dropping, lids drooping, inner brow raise and brow lowered, gaze aversion; hypotonus of vocal apparatus (lax voice*low F0 and restricted F0 range, low amplitude, weak pulses, very low high-frequency energy, spectral noise, format frequencies tending toward neutral setting, broad F1 bandwidth); few and slowed movements, slumped posture Shift toward ergotropictrophotropic balance; increase in depth of respiration, slight heart rate decrease, increase in systolic and diastolic blood pressure, changes in regional blood flow, increased flow to head, chest, and hands (reddening, increased skin temperature in upper torso), pupillary constriction; balanced muscle tone, tension increase in head and neck; eyebrows contracted, eyes widened, lids tightened, eyes narrowed, lips tight and parted, bared teeth or lips tight, pressed together, nostril dilation; stare; chest register phonation (full voice*low F0, high amplitude, strong energy in entire frequency range); agonistic hand/arm movements, erect posture, body lean forward, approach locomotion

Power (Do I have sufficient power to exert control if possible?)

High control/High power: Assertion/Dominance

TABLE 1 (Continued) Stimulus Evaluation Checks (SECs) Organismic/Social functions Control possible/Low power: Protection/Submission Component patterning Extreme ergotropic dominance; faster and more irregular respiration, strong increase in heart rate and heart stroke volume, increase in systolic and decrease in diastolic blood pressure, increase in pulse volume amplitude, vasoconstriction in skin (pallor, decreased skin temperature), gastrointestinal tract, and sexual organs, increase in blood flow to striped musculature, decreased gastrointestinal motility, sphincter contraction, tracheobronchial relaxation, contraction of m. arrectores pilorum, decrease of glandular secretion, secretion of sweat (increase in skin conductance level), pupillary dilatation; muscular hypertonus, particularly in locomotor areas, trembling; brow and lid raising, mouth stretch and corner retraction, switching between gaze direction and aversion; head register phonation (thin voice*raised F0, widely spaced harmonics with relatively low energy); protective hand/arm movements, fast locomotion or freezing

DYNAMIC ARCHITECTURE OF EMOTION

Adjustment (If control is impossible, how well can I adjust to the consequences?) Normative significance (Overall assessment of the event with respect to compatibility with self-concept, values, social-norms, and moral rules Compatibility with internal and external standards (Does the event or my behaviour correspond to (a) my selfconcept or my values, is it just given my entitlement; (b) social norms, values, beliefs about justice, or moral principles Requirements met or surpassed: Relaxation, Bolstering self-esteem, Norm confirmation Ergotropic shift plus elements of pleasantness and high power response

Incompatible: Activation, Self-consciousness, Highlighting norms

Ergotropic shift plus elements of unpleasantness and low power response (peripheral blood flow to face, blushing; body movements: active avoidance of communicative contact)

1313

1314

SCHERER

All of the appraisal criteria can be processed at different levels of processing such as (a) a low-level neural circuit as described above, in which the checking mechanisms are mostly genetically determined and the criteria consist of appropriate templates for pattern matching and similar mechan isms (cf. the notion of biological preparedness, e.g., for snakes, Ohman, 1986; or baby faces, Brosch, Sander, & Scherer, 2007); (b) a schematic level, based on memory traces from social learning processes and occurring in a fairly automatic, unconscious fashion; (c) an association level, involving various cortical association areas, which may occur automatically and unconsciously or in a deliberate, conscious fashion, and (d) the conceptual level, involving propositional knowledge, and underlying cultural meaning systems, requiring consciousness and effortful calculations in prefrontal cortical areas. The different levels continuously interact, producing topdown and bottom-up effects (see Leventhal & Scherer, 1987; Power & Dalgleish, 1997; Scherer, 2005a; van Reekum & Scherer, 1997). The appraisal mechanism requires interaction between many cognitive functions and their underlying neural circuits to compare the features of stimulus events to stored schemata, representations in memory and selfconcept, and expectations and motivational urges of high priority. In addition, this process controls attention deployment and relies heavily on implicit or explicit computation of probabilities of consequences, coping potential, and action alternatives. As shown in Figure 2, the architecture assumes bidirectional influences between appraisal and various cognitive functions. For example, minimal attention needs to be given for appraisal to start, but a relevance outcome will immediately deploy further attention to the stimulus. Stimulus features are compared with schemata in memory but strongly relevant stimulus features will, following appropriate appraisal, be stored as emotional schemata in memory. Event consequences are compared with current motivational states, but particular appraisal outcomes will change motivation and produce adaptive action tendencies. These bidirectional effects between appraisal and other cognitive functions are illustrated by the arrows in the upper part of Figure 2. As shown in Figures 1 and 2, the fundamental assumption of the CPM is that the appraisal results drive the response patterning in other components by triggering efferent outputs designed to produce adaptive reactions that are in line with the current appraisal results (often mediated by motivational changes). Thus, emotion differentiation is the result of the net effect of all subsystem changes brought about by the outcome profile of the SEC sequence. These subsystem changes are theoretically predicted on the basis of a componential patterning model, which assumes that the different organismic subsystems are highly interdependent and that changes in one subsystem will tend to elicit related changes in other subsystems. As illustrated in Figure 2, this process, similar to appraisal, is highly recursive,

DYNAMIC ARCHITECTURE OF EMOTION

1315

Figure 2. Schematic summary of the component process model (reproduced with permission from
Sander, Grandjean, & Scherer, 2005). To view this gure in colour, please visit the online version of this issue.

which is what one would expect from the neurophysiological evidence for complex feedback and feedforward mechanisms between the subsystems (see neural architecture discussion following). As shown in Figure 2, the result of each consecutive check is expected to differentially and cumulatively affect the state of all other subsystems. The CPM makes specific predictions about the effects of the results of certain appraisal checks on the autonomic and somatic nervous systems, indicating which physiological changes and which motor expression features are expected. These predictions are shown in column 3 of Table 1 (see Scherer, 1987, for further details and justification). The predictions are based on specific motivational and behavioural tendencies expected to be activated in the motivation component in order to serve the specific requirements for the adaptive response demanded by a particular SEC result. In socially living species, adaptive responses are required not only for the internal regulation of the organism and motor action for instrumental purposes (organismic functions), but also for interaction and communication with conspecifics (social functions). The assumed functions linked to specific appraisal outcomes are shown in column 2 of Table 1.

1316

SCHERER

How does the CPM, via the mechanism described above, predict specific emotions? Contrary to basic emotion theories (see Scherer, 2009; Scherer & Ellgring, 2007, for reviews), the CPM does not assume the existence of a limited set of discrete emotions or affect programmes, but considers the possibility of an infinite number of different types of emotion episode. As shown below, the beginning of an emotion episode is defined by the synchronisation or coherence of organismic subsystems beyond a certain threshold and its end by the coherence dropping below threshold. The nature of the emotion episode is exclusively determined by the pattern of appraisal results and the specific patterning over time driven by the recursively generated appraisal results (see Scherer, 2004). This is why I have suggested (ever since the first full presentation of the CPM; Scherer, 1984a, p. 311) that the kaleidoscopic arrangement of appraisal results can generate many different kinds of emotion processes without clear categorical boundaries. However, in that chapter I also suggested the existence of certain modal outcomes that occur more frequently due to event contingencies and psychobiological prewiring. Prime candidates for such frequent and important contingencies are major patterns of adaptation in the life of animate organisms that reflect frequently recurring patterns of environmental evaluation and adaptation (see also Ortony & Turner, 1990; Plutchik, 1980). I have suggested using the term modal emotions for emotion episodes resulting from predominant SEC outcomes that are due to general conditions of life, constraints of social organisation, and similarity of innate equipment (Scherer, 1984a, 1994). These modal emotions are predicted by the CPM on the basis of prototypical appraisal profiles, e.g., modal anger often occurring as a result of an appraisal profile that includes novelty, high goal relevance, other agent and intent, high outcome probability, dissonant expectations, goal obstructiveness, high urgency, high control and power, as well as injustice or immorality. In contrast, the fear profile, while similar with respect to relevance and obstructiveness, is characterised by low control and low power appraisals. The detailed CPM predictions are shown in Table 5.4 in Scherer (2001). The elements of the model described so far are highly comparable to other appraisal theories, although some of the labels for particular appraisal criteria and details of the assumed mechanisms may vary (see Ellsworth & Scherer, 2003; Roseman & Smith, 2001). However, the CPM differs from other appraisal theories with respect to two aspects, which are of central importance to our focus on individual differences and the temporal dynamics of emotional processes*the sequence assumption and the notion of integration and synchronisation of components, including, in particular, the experiential feeling component.

DYNAMIC ARCHITECTURE OF EMOTION

1317

Sequential processing of appraisal checks

As shown in Figure 2, the CPM claims that the SECs are processed in sequence, following a fixed order, consisting of four stages in the appraisal process that corresponds to the appraisal objectives described above. This sequence assumption is justified in terms of systems economy and logical dependencies*the results of the earlier SECs need to be processed before later SECs can operate successfully, that is, yield a conclusive result. Expensive information processing should occur for only those stimuli that are considered relevant for the organism. In consequence, relevance detection is considered to be a first selective filter that a stimulus or event needs to pass to merit further processing. Extensive further processing and preparation of behavioural reactions are indicated only if the event concerns a goal or need of major importance or when a salient discrepancy with an expected state is detected, suggesting that the implications for the organism are assessed next in the sequence. Further, the causes and implications of the event need to be established before the organisms coping potential can be conclusively determined, as the latter must be evaluated for a specific situational demand structure. I have suggested that the microgenetic unfolding of the emotion antecedent appraisal processes parallels both phylogenetic and ontogenetic development in the differentiation of emotions. The earlier SECs, particularly the novelty and the intrinsic pleasantness checks, are present in most animals, including newborn humans, and one can argue that these low-level processing mechanisms take precedence as part of our hard-wired detection capacities and occur rapidly after a stimulus event occurs. More complex evaluation mechanisms are successively developed at more advanced levels of phylogenetic and ontogenetic development: Natural selection operates toward more sophisticated information processing ability in phylogenesis, and maturation and learning increase the individuals cognitive capacity in ontogenesis (see Scherer, 1984b, pp. 313314; Scherer, Zentner, & Stern, 2004b). This sequence assumption is often criticised as being overly restrictive and inconsistent with the idea that massively parallel processing of information occurs in different systems. This criticism overlooks the CPMs postulate that external or internal event changes maintain a recursive appraisal process until the monitoring subsystem signals termination of or adjustment to the stimulation that originally elicited the appraisal episode. Thus, the checking process repeats the sequence continuously, constantly updating the appraisal results that change rapidly with changing events and evolving evaluation. The level of processing can be expected to move up during this sequential course, given both the nature of the computation and the likelihood that lower levels have been unable to settle the issue. The normative significance

1318

SCHERER

of the event, that is its consequences for the self and its normative or moral status, is expected to be appraised last, as it requires comprehensive information about the event and comparison with high-level propositional representation. The proposed mechanism is highly compatible with the assumption of parallel processing as all SECs are expected to be processed simultaneously, starting with relevance detection. It should be noted that the essential criterion for the sequence assumption is not the occurrence of the event and the start of the specific appraisal process but the time at which a particular check achieves preliminary closure, that is yields a result that warrants efferent commands to response modalities, as shown by the descending arrows in Figure 2 (see Scherer, 2004, for details). In this way the assumption of continuously parallel processing is entirely compatible with the notion of a sequence of primary outcomes. The sequence theory postulates that, for the reasons outlined earlier, the result of a prior processing step (or check) must be in before the consecutive step (or check) can produce a conclusive result with efferent consequences. It is indeed feasible to assume that the results of parallel processes for different evaluation criteria will be available at different times, given differential depth of processing.

Integration and central representation of components

The CPM assigns a special status to the feeling component in the emotion process, which should not be confused with the complete componential package of emotion (see Scherer, 2005b), as it monitors and regulates the component process and enables the individual to communicate its emotional experience to others. Subjective experience needs to integrate and centrally represent all information about the continuous patterns of change and their coherence in all other components, especially if it is to serve a monitoring function. Thus, feeling is an extraordinarily complex conglomerate of information from different systems. Scherer (2004) has presented a CPM account of how feelings integrate the central representation of appraisaldriven response organisation in emotion. This proposal is briefly summarised in this section. Psychologists generally assume that feelings are conscious phenomena and that the only way to access and measure feelings is via verbal report (but see Lacewing, 2007). However, this widespread assumption holds only for the visible tip of a huge iceberg. The CPM (Scherer, 2004, 2005a) conceptualises the problem, as shown in Figure 3, with a Venn diagram in which a set of overlapping circles represents the different aspects of feeling. The first circle (A) represents the sheer reflection or representation of changes in all synchronised components in some form of monitoring structure in the central nervous system (CNS), which receives massive

DYNAMIC ARCHITECTURE OF EMOTION

1319

Figure 3. The reection of component emotion processes in a monitor system, circle A, the
emergence of consciousness, circle B, and categorisation and verbalisation, circle C (reproduced with permission from Scherer, 2004). To view this gure in colour, please visit the online version of this issue.

projections from both cortical and subcortical CNS structures (including proprioceptive feedback from the periphery). To my knowledge, there is currently neither an established technical term nor a precise definition for this type of integrated unconscious representation. Clearly, to stand out from the stream of booming, buzzing neural activation, an organising principle must be at work. Given my emphasis on dynamic synchronisation, I suggest that this principle might correspond to coherence spreading. This mechanism has been described for sensory perception (Henkel, 2000) as a highly nonlinear selection operation between independently operating disparity units that is achieved quickly and reliably by neural hardware. Coherence spreading is robust because only the data available within the coherent cluster are fed forward to later processing stages, and only the tiny coherent portion of the neural information streams is transmitted for further analysis. Although several neural realisations of this process may be possible, the simplest and most common would rely on the synchronisation properties of pools of neural oscillators (for example, the phase at low frequencies in subcortical regions may tune the rhythms of high-frequency bursts related to cortical activities). If this mechanism works for exogenous perception, it may very well do the job for endogenous interoception. In consequence, I suggest that the content of circle A be called a neural coherence cluster. The second circle (B), which only partially overlaps with the first, represents that part of the integrated central representation that becomes conscious. This circle corresponds most directly to what is more generally called feelings or qualia. I have suggested that the degree of synchronisation of the components (which might in turn be determined by the

1320

SCHERER

pertinence of the event as appraised by the organism) generates awareness (Scherer, 2005a; see also Grandjean, Sander, & Scherer, 2008). This suggested architecture depends in large part on the processes of synchronisation and integration within and between components (Grandjean et al., 2008). Within-component integration is required because different structures and processes interact in a complex fashion during emotion episodes. Thus, information integration within the cognitive component is required, given the parallel and sequential processing for different evaluation criteria on different levels. As shown in Figure 3, the proprioceptive (both intero- and exteroceptive) feedback information from different response components (such as vocal and facial expression or psychophysiological changes) is integrated with the representation of the appraisal results to yield a neural coherence cluster. I suggest that the very process of multicomponent integration and synchronisation constitutes the critical defining feature of emotion episodes. In trying to understand integration at different points in the emotion episode and the rules likely to underlie this process, the key issue concerns the relative weight given to the different components*appraisal, physiological responses, motor expression, motivation, and action tendencies. Research to examine the nature of this multimodal integration has hardly started. Although there is renewed interest in interoception and its neural basis (Craig, 2002; Wiens, 2005), few attempts have been made to examine how the peripheral representation is integrated with the representation of other subsystems. An important feature of emotion processes is that they are also dynamically integrated over time. The term emotional state is misleading, as it suggests a static, unitary phenomenon rather than a flow of continuously changing component states that constitute emotion episodes. Although humans can focus on micro-momentary changes of feeling, we tend to become aware of our feelings in experiential chunks that provide phenomenal unity to a particular feeling quality or quale. Temporal integration, in the sense of experiential chunking, is most likely determined by the period during which a certain level of component synchronisation persists. Contrary to discrete or basic emotion theories (Ekman, 1984, 1992; Izard, 1977, 1993; Tomkins, 1984), the CPM does not share the assumption of a limited number of innate, hard-wired affect programmes*even for basic emotions such as anger, fear, joy, sadness, and disgust. Rather, as shown above, the emotion process is considered as a continuously fluctuating pattern of change in several organismic subsystems that become integrated into coherence clusters and thus yields an extraordinarily large number of different emotional qualities, virtually as many as there are different integrations of appraisal results and consequent response patterns (Scherer,

DYNAMIC ARCHITECTURE OF EMOTION

1321

1984b, 2001). In consequence, there should be an almost unlimited number of qualia. Before proceeding, it may be useful to examine this concept in greater detail, as it defines the content of circle B. Lewis (1929) pioneered the use of the term qualia in its generally agreed modern sense as follows:
There are recognizable qualitative characters of the given, which may be repeated in different experiences, and are thus a sort of universals; I call these qualia. But although such qualia are universals, in the sense of being recognized from one to another experience, they must be distinguished from the properties of objects. . . . The quale is directly intuited, given, and is not the subject of any possible error because it is purely subjective. (p. 121; see also Metzinger, 2004)

Dennett (1988) has identified four properties that are commonly ascribed to qualia: (i) ineffable (cannot be communicated or apprehended by any means other than direct experience); (ii) intrinsic (nonrelational properties, which do not change depending on the experiences relation to other things); (iii) private (interpersonal comparisons of qualia are systematically impossible); and (iv) directly or immediately apprehensible in consciousness (to experience a quale is to know one experiences a quale and to know all there is to know about that quale). Although there is a lot of philosophical debate about the concept, and despite Dennetts (2001) warning about the functional use of the term qualia in neuroscience, I feel that its considered use is preferable to the complete absence of an appropriate term for what Tolman (1935) meant by immediate experience or raw feels (see also Kirk, 1996). I feel that it fits the conscious representation of an integrated multimodal coherence cluster as personal, subjective experience of an emotion episode rather well. It is particularly important to insist on Lewis description of qualia as universals being stable over experiences of a similar type and thus memorisable and recognisable. In consequence, it is the qualia that constitute the core of affective experience, the most fundamental, raw, untouched (in the sense of further processing) representation of the underlying appraisal and consequent response processes. This is very different from Russells (2003, and 2009 this issue) notion of core affect, which is defined as a single point in a twodimensional valence)arousal space. Qualia, in contrast, are seen as central representations of integrated multimodal coherence clusters and thus their formation should in principle be predictable, observable, and explainable* although researchers currently lack the required theoretical modelling tools as well as the appropriate methodology. Qualia are reflections of complex multidimensional feature spaces; which, like any other high-dimensional data set, can be projected into lower-dimensional space (see below; Scherer, 2005b).

1322

SCHERER

Categorisation and verbalisation of feeling

Although a richly textured conscious feeling that fits the situation a person experiences like a glove is highly functional for fine-tuned adaptation and regulation, it is less well suited for cognitive manipulation, memorisation, or communication. Of course the same is true for all perception, which is why categorisation plays such a major role in all of cognitive processing. How can one envisage feeling categorisation on the basis of the component process architecture? Rosch (1973) proposed two general principles for the formation of categories that have been widely accepted: (i) cognitive economy, reducing the infinite differences among stimuli to behaviourally and cognitively manageable scope; and (ii) reflection of perceived structures and regularities in the world. In addition, Rosch proposed a distinction between vertical (inclusiveness of a category) and horizontal (differentiation of categories at any one layer) levels of categorisation. Finally, she proposed that the use of prototypes, containing the most representative attributes inside a category, will increase the flexibility and distinctiveness of categories along the horizontal dimension. Shaver and his collaborators (Shaver, Schwartz, Kirson, & OConnor, 1987) have shown the utility of this approach for understanding the emotion vocabulary, distinguishing between superordinate (positive, negative emotions), basic (joy, anger, sadness), and subordinate (irritation, rage) category levels. But categorisation clearly occurs before verbalisation, and numerous preverbal categories for emotional feelings are to be expected. In fact, these prelinguistic groupings form the basis for the acquisition of concepts and words in very young children (e.g., Hamlin, Wynn, & Bloom, 2007; Hirschfeld, 2001). In each of these cases, the perception of some regularity in the differences between objects or actions seems to be involved. One would imagine that the same is true for the formation of prelinguistic emotion categories. Individuals would need to perceive some regularity in the differential patterning of different emotional experiences in order to form such preverbal qualia categories. Not only on the ontogenetic level, but equally so at the microgenetic level, prelinguistic categories must precede verbalisation (if only to reduce the candidate sets of words for which the semantic features need to be checked against felt experience), and one needs to pose the question of the processes that have led to the formation of such preverbal emotion categories. Most likely, the lowest level of inclusiveness is that of the preverbal qualia, as they represent qualitative types of experiences that can be recognised and remembered. This level can be expected to be idiosyncratic and fuzzy. Using Roschs principles, one would expect that more inclusive categories are formed through a constraint satisfaction procedure between cognitive economy optimisation and the practical importance of certain structures and regularities

DYNAMIC ARCHITECTURE OF EMOTION

1323

encountered during emotion episodes. It is thus of primary importance for an understanding of qualia categorisation to identify these structures and regularities. Even though the content of the qualia is likely to consist of an integrated representation of the individual components, one might imagine that some components, especially those that best differentiate types of qualia, are more important than others in determining categorisation. One candidate for determining category formation is somatovisceral feedback, which adherents of a peripheralist tradition founded by James and Lange (Mandler, 1975; Schachter & Singer, 1962) tend to emphasise (Barrett, 2006, uses the term embodied categories). While somatovisceral feedback is certainly one of the determinants of categorisation, an exaggerated emphasis on this factor, to the exclusion of other factors, may prove to be an impediment in trying to understand the process. The reason is that certain types of physiological activation occur in a similar form in several emotions (e.g., increase in heart rate and muscle tension) because sympathetic activation serves urgent action preparation, a feature shared by several emotions. In consequence, it seems unlikely that this modality is the only determinant of category formation (as many critics of the James/Lange position, have argued over the last century). The perception of bodily changes may well add colour to the emotional experience (as James, 1890, argued), for example, concerning the intensity and vividness of the feeling, but it is unlikely to account for the bulk of differentiation and categorisation of feelings. One can reasonably argue that appraisal configurations (Scherer, 1997a, 2001), core relational themes (Lazarus, 1991; Smith & Lazarus, 1993), or action tendencies (Frijda, 1986, 1987) are more promising candidates. Thus, I assume that qualia representing integrated appraisal configurations and action tendencies (probably including the accompanying somatovisceral response patterns as part of the integrated package), which occur relatively frequently and are of central importance to the individuals wellbeing, the modal emotions described earlier, will serve as the basis for categorisation of qualia clusters into more inclusive categories on Roschs vertical dimension. As suggested early on (Scherer, 1984a, p. 311), I believe that it is those modal categories that are generally labelled with a single word or a brief expression in most languages of the world and that the availability of such linguistic labels imposes a large degree of separateness and discreteness to particular types of experiences. I also think that these categories correspond to both the subordinate and basic levels of categorisation in Shavers system. I now turn to the issue of the labelling of ones preverbal feelings and their communication. This process is one of the most neglected and least well understood components of emotion research. Presumably, the same experiential chunks that form the coherence clusters described earlier serve as the basis for verbalisation. Again, the partial overlap of circle C in Figure 3

1324

SCHERER

indicates that the use of linguistic labels or expressions to describe the conscious part of feeling rarely covers the complete conscious experience, which may be due to the lack of appropriate verbal concepts or strategic communication intentions. On the other hand, the implications of the chosen verbal description may go beyond the content of the emotional experience, as the denotation and connotation of the concepts used in the verbalisation may add surplus meaning (or even modify the meaning to some extent). In any case, verbal report always relies on language and thus on the semantic fields of emotion words or metaphors to express conscious experience. The extraordinarily rich texture of the qualia concerned is likely to be expressed to only a small degree. Apart from capacity constraints (the constantly changing stream of conscious content cannot be appropriately represented by a discrete utterance), the available words and expressions in a language constrain the potential complexity of differentiation, despite the fact that most languages offer a choice of several hundred emotion terms. Given strong individual differences in category width and verbal ability (including alexithymia), this indeterminacy may account for much of the variance in emotion reports. An interesting question is whether the act of categorisation that is implied by verbal labelling, which will undoubtedly affect the representation of the emotional experience, may impoverish the rich qualia experience and mould it into socioculturally determined schemata. Another interesting question concerns the labelling mechanism. Does the integrated central representation described earlier first activate a preverbal category that can then be verbally labelled if so desired? Or does the individual check the semantic features of emotion words and expressions against the centrally represented experience if the need arises? Of course, languages differ somewhat with respect to emotion vocabulary, and it is thus imperative to better understand the nature and origin of the differences between the semantic fields of emotion terms in different languages across the world. Together with a large group of international collaborators, I have started to examine this issue with the help of the GRID approach (asking native speakers to evaluate a standard set of emotion terms on a feature vector of 144 items representing all components of emotion; Scherer, 2005b). Currently, data for about 30 languages have been collected. First results suggest that four dimensions are necessary to define the affective space onto which the meaning of major individual emotion terms can be projected* valence, power/control, arousal, and unpredictability, in this order (Fontaine, Scherer, Roesch, & Ellsworth, 2007).

Individual differences and regulation

Apart from its emphasis on components and dynamic unfolding of emotion, the CPM has, like all appraisal theories, the advantage of being particularly

DYNAMIC ARCHITECTURE OF EMOTION

1325

useful for the conceptualisation and prediction of individual differences in emotional reactions*even in cases in which the eliciting event is objectively the same. Take baggage loss in the airport, for example. One might think that this event would produce the same emotion in all passengers concerned by such a loss. But this is not the case. In a study of 109 airline passengers waiting in vain for their baggage to materialise (Scherer & Ceschi, 1997), we found that there were virtually no two passengers who reported feeling exactly the same pure emotion. Rather, all of them reported various mixtures or blends of several different emotions. When we systematically questioned these passengers about their appraisals of the event (using questions based on the CPM checks), we could show that specific appraisal results and predicted response types clustered closely together in threedimensional space (see Figure 6 in Scherer & Ceschi, 1997). In analysing the videotaped interaction between the passengers and the airline agents processing their claims, we could also show that certain types of appraisal differentiate the occurrence of true (Duchenne) smiles or false smiles, demonstrating that the predictive validity of appraisal goes beyond verbal report (Scherer & Ceschi, 2000). Similarly, in a series of experiments Kuppens and his collaborators have demonstrated that anger can occur in combination with different patterns of appraisals, varying as a function of situation but also person characteristics. Thus, while for some individuals frustration is sufficient for becoming angered, for others, the thwarting has to be appraised as norm violation as well as unfair and deliberate in order for them to experience anger (Kuppens, Van Mechelen, Smits, De Boeck, & Ceulemans, 2007).
TABLE 2 Voice type predictions for specific types of affect disorders (reproduced from Scherer, 1987) Vocal tract Pharynx setting Euphoria Mania Anhedonia Helplessness Anxiety disorder Agitated fear Chronic frustration Indifference, apathy Shame, guilt Wide Wide Very narrow Narrow Narrow Narrow Narrow Neutral Narrow Tension Medium tense Extremely tense Slightly tense Lax Medium/high tense Extremely tense Medium tense Slightly Slightly tense Register Slightly full Extremely full Neutral Thin Thin Very thin Medium full Neutral Thin

Note: For the discussion of voice types, see column 3 in Table 1.

1326

SCHERER

These examples show the utility of appraisal theories, and thus of the CPM, to model and predict individual differences. In particular, these theories predict that emotions are elicited entirely on the basis of an individuals subjective evaluations of the event and his or her role in it rather than its objective characteristics*given that the latter may be perceived differently and evaluated on the basis of the individuals goals and values as well as on his or her coping potential. In consequence, as the baggage-loss example shows, if one knows the results of an individuals event appraisal on the major checks, one can approximately predict (see Table 5.4 in Scherer, 2001) what kind of emotion he or she will most likely experience (or more precisely, what label the person is likely to use to refer to the experience). Furthermore, it can be predicted (based on the component patterning model; see Table 1, Scherer, 1987, and Tables 5.3 and 5.4 in Scherer, 2001) what motor expressions, action tendencies, and physiological changes can be expected to underlie this experience. Alternatively, again based on the component patterning model, if one observes particular motor expressions of an individual in a given situation, one can try to infer the results of the persons specific appraisal of an event (and predict the likely emotion, or the label used, on that basis; Scherer, 1988, 1992). To my knowledge, none of competing emotion theories is structurally able to account for individual differences in such a detailed fashion or to make concrete a priori predictions. Apart from this general approach to individual differences in emotional responses to given events or situations, our group has attempted to identify dispositional factors that may lead to stable tendencies to appraise events in a particular fashion. These tendencies or biases can systematically affect an individuals evaluations and cause him or her to stray from a possible modal response to a given event, to the point of pathological responding. In consequence, Scherer (1987) suggested conceptualising different types of emotional disorders on the basis of appraisal malfunctioning. The underlying assumption is that although appraisal is subjective and may vary from individual to individual, it must remain*within certain limits* appropriate to the objective situation (e.g., through reality testing) and to the coping potential that is commonly perceived to be within the individuals means. Violation of these appraisal reality constraints, as one might call them, will lead to the resulting emotion being considered as abnormal or disordered, at least by an individuals social environment, if not by him- or herself. For example, Scherer (1987) has suggested that one particular form of depression, helplessness, might be partly due to a consistent underestimation of ones coping potential. An important distinction is that an individual who truly lacks the means to deal with a particularly difficult situation would be described as dejected, whereas someone described as depressed is implicitly assumed to appraise the situation, particularly his

DYNAMIC ARCHITECTURE OF EMOTION

1327

or her coping potential, in an inappropriate or unrealistic fashion. Table 2 shows the predictions for vocal signatures of specific emotional disturbances as based on the combination of the appraisal bias predictions and the component patterning model. Kaiser and Scherer (1998) extended this analysis and suggested four possible types of long-term malfunction and the chronic emotionality disturbances that can result: (a) inappropriate or inadequate appraisal of a situation and events; (b) inappropriate or inadequate level of processing; (c) inappropriate or inadequate motor expression or signalling; and (d) inappropriate or inadequate relationships between aspects of feeling. Appraisal biases in the form of inappropriate or inadequate appraisal of a situation and events in specific can be linked to the
TABLE 3 Facial action units predicted as indicators of selected types of affect disorder Related action units (shorthand descriptions) 5 (lids up), 26 (open mouth), 38 (open nostrils); or 6 (crows feet wrinkles), 12 (lip corners up), 25 (lips part) 4 (brow lowering), 7 (lid tightening), 9 (nose wrinkling), 10 (upper lip raising), 15 (lip corners down & chin raised), 24 (lip press), 39 (nostrils closed) 4 (brow lowering), 7 (lid tightening), 17 (chin raised), 23 or 24 (lips tight or pressed together) Hypotonus of facial musculature 4, 5 (eyebrows contracted, eyes widened), or 7 (lids tight, eyes narrowed); 23, 25 (lips tight & parted, bared teeth) 15 (lip corner depression), 25 or 26 (lips part or jaw drop), 41 or 43 (lids droop or eyes closed), if tears 1'4 (inner brow raised & contracted) 1, 2, 5 (brows & upper lid raising), 20 (mouth stretch), 26 (jaw drop), or 32 (lip bite) 4 (eyebrows contracted), 1, 2, 5 (brows and upper lid raising), 20 (mouth stretch), 23, 24 (lips tight or pressed), 32, 37 (lip bite or wipe)

Affect disorders Euphoria

Inadequate appraisals Excessively positive pleasantness and conduciveness evaluation Excessively negative pleasantness evaluation

Anhedonia

Chronic dissatisfaction, frustration Indifference, apathy Mania

Strong bias toward habitual negative conduciveness evaluation Malfunctioning of conduciveness check Overestimation of power and control

Hopelessness

Underestimation of control

Helplessness

Underestimation of power

Anxiety disorders

Excessive concern about adequacy of power

1328
TABLE 4 Potential sources of systematic individual differences (dispositions, biases) in event appraisal and response dispositions, classified by emotion component and type of origin or processing Individual differences (IDs) Appraisal process Hardwired / constitutional Automatic Sensorimotor Genetic or cultural factors, brain circuitry biases (chemosensory signal sensitivity; speed and structure of cognitive system, e.g., category width, illusions; value systems) Reflexivity, impulsivity Vagal tone, temperament, stablelabile autonomic nervous system Learned / dispositional Schematic Unconscious Transient / voluntary Controlled Conscious SCHERER Momentarily dominant biases Personal learning history (conditioned perception and judgement tendencies; (interpersonal stances; hypothesis testing) dispositional appraisal biases due to wishful thinking, stable personal or situational schemata) Dispositional reaction and coping tendencies, personality Physiological response schemata Evaluation of adaptivity and success probability of action alternatives Adopting physiological control stances Strategic choice of expressive patterns, regulation Deliberate categorisation, availability of differentiated labels

Motivational change Physiological responses Motor expression Subjective feeling

Constitutional and cultural expressivity Motor attitudes, habitual expressions Temperamental affectivity Trait affectivity, nonverbal categorisation schemata, contagion/information effect of others

DYNAMIC ARCHITECTURE OF EMOTION

1329

efferent expressions generated by specific appraisals, as predicted in the component patterning model, which may give rise to inappropriate facial expressions (see Table 3). Van Reekum and Scherer (1997) further extended the analysis of appraisal biases as stable dispositions and distinguished between (a) appraisal biases with respect to the form or process of appraisal such as speed, thoroughness or completeness, width of the categories used in inference and classification, vigilance, or degree of top-down control of lower-level processing (e.g., sensitisation or automatisation); and (b) appraisal biases with respect to content such as agency or power attribution bias, over sensitisation, valence appreciation, over- or underestimation of relevance, conduciveness, control and coping potential, fairness, or moral appropriateness (see Table 1 for greater detail on these SECs). Van Reekum and Scherer also reviewed the literature on potential sources for individual differences in appraisal tendencies such as organismic predispositions, cognitive styles, need for cognition, personality traits (e.g., extroversion, repressionsensitisation, neuroticism, rigidity, dysphoria, worrying, sensation-seeking, openness), attitudinal structures, self-concept, or self-image. The main origin of these interindividual differences are probably strongly related to early differences (genetics and epigenetics) in the development of the central nervous system, particularly in terms of executive functions or perception biases. For example, Canli et al. (2006) found support for a model by which life stress interacts with the effect of 5-HTTLPR genotype on amygdala and hippocampal resting activation that may provoke a chronic state of negative cognitive bias including increased vigilance, threat, or rumination and may thus constitute a neural mechanism for epigenetic vulnerability for depression. These theoretical analyses are currently pursued and extended to other components of emotion. Table 4 shows a preliminary attempt, produced for the purposes of this article, to categorise the sources of individual differences in emotional responding by the components and their potential origins. There is not sufficient space to discuss these suggestions in detail. However, there is now copious evidence for the powerful role of dispositional factors (both constitutional and learned) for emotional reactions, in particular the role of personality factors and trait affect for both appraisal and motivational aspects (Griner & Smith, 2000; Matthews, Derryberry, & Siegle, 2000; Reisenzein & Weber, in press; Revelle & Scherer, 2009). Kuppens and his collaborators have illustrated these links for the case of anger. Thus, they demonstrated that trait anger is negatively correlated with agreeableness and perceived social esteem (especially for individuals who attach high importance to social relations; Kuppens, 2005) and showed situation-specific individual differences in the appraisal of threatened self-esteem and otherblame, which showed differential relationships with dispositional variables

1330

SCHERER

TABLE 5 Individual difference variables potentially biasing appraisal toward anger outcomes Appraisal dimension Relevance detection Novelty Intrinsic pleasantness Goal relevance Implication assessment Causal attribution Outcome probability check Discrepancy from expectation Goal/need conduciveness Urgency Individual difference variables

Speed of habituation, extent of inhibition Anhedonia (Germans & Kring, 2000) Human motivation, achievement motivation, affiliation motivation (McClelland, 1985) Explanatory style (Seligman et al., 1979) Optimismpessimism (Scheier & Carver, 1985) Openness to experience/conservatism (Costa & McCrae, 1992) Perfectionism (Stoeber & Otto, 2006) Realism

Coping potential determination Control Locus of control (Rotter, 1966); illusion of control Power Self-esteem (Kuppens, 2005; Kuppens & Van Mechelen, 2007); selfefficacy (Bandura, 1977) Adjustment Openness to experience (Costa & McCrae, 1992); flexibility, agreeableness (Kuppens, 2005) Normative significance evaluation (a) Internal standards Individual human values (Schwartz, 1992) (b) External standards Cultural values (Hofstede, 1984; Schwartz, 2006)
Note: Adapted from Wranik and Scherer (2009).

(unstable self-esteem, neuroticism, and BIS sensitivity in unpleasant evaluative situations vs. feeling lowly valued by others in non-evaluative situations; Kuppens & Van Mechelen, 2007). Based on this earlier work, Wranik and Scherer (2009) developed a model of how such appraisal biases may systematically privilege the occurrence of anger experiences, arguing that individuals differ in how they selectively attend to specific elements of a situation or event, how these elements are cognitively encoded, and how these encodings activate and interact with other cognitions and affects in the overall personality system (Mischel & Shoda, 1995). This may explain why some people experience anger more frequently or intensely or why they generally experience certain types of emotions under specific conditions. For example, a perfectionist may chronically overestimate the importance of events, an impatient person may overestimate the urgency of situations, or a person particularly sensitive to injustice will evaluate many situations as unjust. Table 5 illustrates this approach for the emotion of anger, listing individual difference variables that

DYNAMIC ARCHITECTURE OF EMOTION

1331

could systematically influence specific appraisal dimensions as postulated by the CPM (Scherer, 1984a, 1984b, 2001), including traditional personality traits, social-cognitive and motivational dispositions (e.g., self-efficacy, selfesteem, optimism), and differences in lower-level cognitive processing (e.g., inhibition, processing speed). The underlying idea is that these individual differences influence specific appraisal dimensions in a relatively stable manner and thus help explain why some people are more likely to experience certain types of emotions under specific conditions than other people do. Recently, Scherer and Brosch (2009) have made a first effort to develop a preliminary theoretical analysis of the effect of specific appraisal tendencies or biases on emotion dispositions and trait affect such as trait anxiety, trait anger, or trait positive/negative affect. Based on the theoretical suggestions outlined above, we developed the compilation shown in Table 6, illustrating possible links between certain appraisal tendencies or biases and specific emotion dispositions or trait affects. In addition, we made an attempt to illustrate how culture-based goal, belief, and value systems can produce appraisal biases by affecting the perception of events and the criteria used in their evaluation. So far, only individual differences with respect to dispositions, biases, and distortions in the use of the different appraisal check criteria have been discussed. Of course, individual differences go way beyond this aspect of emotion processing. Thus, individuals may differ with respect to the levels of analysis that are preferentially used in appraisal or in the rapidity of switching between levels or the capacity to integrate levels. Furthermore, individuals may differ as regards the rules underlying the integration of the results produced by different appraisal checks. For example, Kuppens and his collaborators showed that whereas some individuals react with approach behaviour to anger, others prefer avoidance (often coupled with social sharing; Kuppens, Van Mechelen, & Meulders, 2004) and that the relationships between appraisals and the resulting emotional experience can be very different for different angry individuals (Kuppens, Van Mechelen, & Rijmen, 2008). I have suggested using Andersons (1989) model of integration functions to understand the process whereby subjective appraisal results are transformed into an implicit response (Scherer, 2004). Individuals may develop, strongly affected by their current goals, specific integration rules for the combinations of specific appraisal criteria. For example, we found empirically (van Reekum et al., 2004) that different levels of coping potential have very different effects upon psychophysiological responses as a function of goal conduciveness, reflecting the fact that coping ability is of less relevance when things are going according to plan. In Andersons approach, this would be modelled by a configuration rule, predicting that the importance of one of the criteria depends on the level of another. As mentioned above, individual

1332

TABLE 6 Individual difference variables and cultural goal, belief, and value dimensions potentially biasing appraisal processes toward trait affect (reproduced from Scherer & Brosch, 2009) Potentially facilitating culturally dominant goal, belief, value dimensions Goa: Interdependent goal pursuits Bel: Human nature good Val: Conservatism, security, embeddedness, benevolence, harmony Goa: Independent goal pursuits Bel: Human nature bad, normativity Val: Conservatism, self-enhancement, autonomy, entitlement, mastery Goa: Independent goal pursuits Bel: Human nature bad, normativity Val: Conservatism, self-enhancement, autonomy, entitlement, mastery Goa: Interdependent goal pursuits Bel: Human nature good Val: Conservatism, embeddedness, benevolence, harmony Goa: Independent goal pursuits Bel: Human nature good Val: Embeddedness, benevolence, harmony, openness for change

SCHERER

Emotion disposition / Trait affect (emotional disorder) Trait sadness: Resignation, dejection, acquiescence (depression) Trait anger: Irritation, irascibility, choleric (hostility, psychoticism)

Appraisal tendencies or biases (motivational and cognitive) Mot: Strong attachment to people and property Cog: Low self-esteem, underestimation of control, coping, and adjustment potential; tendency to ruminate Mot: Strong goal orientation, high expectations Cog: High self-esteem, external attribution, blaming, overestimation of control, power, coping, and adjustment potential; exaggerated optimism Mot: Perfectionism Cog: Exaggerated sensitivity for novelty, uncertainty, and urgency (looming); low self-esteem, underestimation of control, coping, and adjustment potential; exaggerated pessimism Mot: High need for self-worth and social recognition; conformity; perfectionism Cog: Internal attribution Mot: Hedonism, realistic aims Cog: Optimism; high self-esteem, overestimation of control, coping, and adjustment potential

Trait anxiety: Worrier, apprehensiveness, neuroticism (general anxiety disorder) Trait shame/guilt: Embarrassment, unworthiness, disconcertment, abashment (clinical shame/guilt syndromes) Trait positive affect: Joyfulness, buoyancy, cheerfulness, good spirits (manic euphoria)

Note: Mot: motivational, Cog: cognitive; Goa: goal pursuit; Bel: beliefs about human nature; Val: value dimensions.

DYNAMIC ARCHITECTURE OF EMOTION

1333

differences in cognitive or motivational dispositions, as well as stable sociocultural factors and the immediate social context, may have a very powerful effect on the weighting of specific criteria and may, in and of itself, bias the integration of the appraisal criteria toward configuration rules. To model such individually different integration functions we may need to adopt nonlinear dynamic system analysis (Scherer, 2000a) as a more appropriate framework for emotion modelling than the classic assumption of linear functions dominating our current statistics toolbox (e.g., regression analysis). Similarly, individuals may vary in the process of integration of the other organismic subsystems involved in emotion, such as autonomic variability and muscular expression. Individuals are likely to differ with respect to their capacity for interoception and the use of proprioceptive cues in the different response components of the emotion process, which will affect feedback and integration mechanisms. The most important integration mechanism, elicited and organised by the process of synchronisation that I have proposed as the hallmark of the emotion phenomenon, is the integration of all the synchronised components (see Figure 3), which probably occurs outside of awareness. Anderson (1989, p. 147) suggested: What does attain consciousness is often, perhaps always, a result integrated across different sense modalities at preconscious stages. The nature of this integration process has not yet been addressed: Are the widely varying types of representations exchanged into a common currency? Or does even the final product of integration still consist of a heterogeneous amalgam of representations reflecting the specific nature of the various components? Individuals may differ in how this process works and they will most likely differ as regards the relative weights attributed to different components. Thus the qualia of an emotion experience may be more strongly determined by appraisal results whereas another gives greater importance to self-perceived action tendencies. One individual may strongly weight internal physiological arousal whereas another places more emphasis on proprioceptive motor cues. The old distinction between internalisers vs. externalisers may well be relevant here (see Cacioppo et al., 1992). Another source for individual differences may be the threshold of synchronisation or coherence that is needed for an emotional process to become conscious. Finally, we do know that there are major individual differences with respect to linguistic labelling (circle C in Figure 3) based on factors such as pathology (alexithymia), verbal intelligence or experience (especially knowledge of vocabulary), socialisation (e.g., verbal interaction with mothers; Dunn & Brown, 1994), or past experiences.

1334

SCHERER

COMPARISON WITH OTHER EMOTION THEORIES

As one might expect, the CPM is highly compatible with other appraisal theories (see Ellsworth & Scherer, 2003; Scherer, 1999a). Although not all theorists subscribe to all aspects of the model, the issues concerned are fairly minor and can be settled by empirical evidence. The CPM is continuously updated on the basis of new findings. The CPM is also largely compatible with theories that stress the importance of particular aspects of emotion such as relationships to goals (Oatley & Johnson-Laird, 1987), causal attribution (Weiner, 1985), or action tendencies (Frijda, 1987, 2007b). The CPM also postulates a major role for these aspects, and adjustments can be made once new evidence comes in. There seem to be major differences with discrete or basic emotion theories in the tradition of Ekman (1992) and Izard (1993), especially about the small number of emotions addressed and the rather rigid notion of affect programmes leading to prototypical response patterns (see Scherer & Ellgring, 2007). However, these differences are much less decisive than they appear at first sight, as these theorists have stressed in their recent writings that they: (a) consider complex emotions in addition to basic emotions; (b) postulate emotion families that allow for many gradations within each family; (c) assume affect programmes to be flexible; (d) postulate interactions between subsystems; and (e) consider both automatic and elaborated appraisal as differentiating elicitors (see Scherer, 2009, for further detail). In contrast, the CPM architecture is incompatible with recent constructivist theories suggested by Russell (2003) and Barrett (2006). Implicitly, Barrett (2006, p. 31) considers the CPM as a natural kinds model, like most theories other than her own, and declares them as obsolete. It is to be hoped that the preceding description of the model has convinced the reader that it is anything but a natural kinds model, postulating fuzzy sets of modal emotions (as a result of a categorisation of qualia) on the background of an infinite variety of emotional processes and their qualia representations. Barrett has made the issue of feeling categorisation the centrepiece of her neo-constructivist theory claiming that emotions are exclusively constituted by the individuals categorisation (through a conceptual act) of core affect (valence)arousal), as determined by personal memories and cultural concepts. The CPM is in strong disaccord with several aspects of these claims, as is evident from comparing the model described above with the two constructivist approaches in this volume (Barrett, 2009 this issue; Russell, 2009 this issue). First, both Russell and Barrett redefine the word emotion to refer exclusively to categorised feeling, following William James, who also treated the terms emotion and feeling as synonyms (Scherer, 2005b, p. 699), thereby giving rise to a century of fruitless debate (Ellsworth, 1994). The

DYNAMIC ARCHITECTURE OF EMOTION

1335

CPM treats feeling as one of the components of the emotion process (albeit an important one, as feeling monitors the emotion episode through integration and representation of changes in the other components). Using the words emotion and feeling interchangeably (as is sometimes done in nave language) muddles the important conceptual distinction made by defining these words differentially as technical terms. In the context of a componential approach to emotion, such a semantic confusion is logically problematical, as it implies treating a part or component as equivalent to the whole (a pars pro toto problem). Second, both Russell and Barrett claim that core affect, a point in a low dimensional valence)arousal space, is the central psychological primitive for affective feeling and the basis for all additional processing, i.e., the construction of an emotion category. This claim is neither justified theoretically nor demonstrated empirically. It is not clear in what sense and why valence and arousal feelings are considered as more core, primitive, or basic than other internal representations. The core affect theorists may argue that people can reliably describe their feelings on these dimensions. However, this is easily explained by a projection from a highdimensional qualia space to a lower dimensional space (see above), especially if a person is invited to engage in such integration and projection by being given valence and arousal scales. This does not mean that the lowdimensional projection is core in the sense of being logically prior or more raw or primitive in the sense of the extent of processing. On the contrary, a low-dimensional projection is most likely a highly processed product. This view is buttressed by findings showing that when one asks people what they have felt during a certain emotional episode, they rarely spontaneously answer in terms of valence and arousal gradation. We asked a representative sample of the Swiss population what emotion they experienced on the previous day. They described the situation and labelled their subjective experience, their feelings, in their own words. Only a very small percentage of the more that 1000 respondents used general or positive valence labels (5.8%), and almost none used direct arousal terms (Scherer, Wranik, Sangsue, Tran, & Scherer, 2004a). Obviously, if we had provided them with scales for valence and activation of the experience, they would have gladly complied. But the low-dimensional description does not spontaneously come to mind*which is what one would expect if it were a primitive. Maybe Russell and Barrett mean the word core in the sense of the most important dimensions in low-dimensional space (valence and arousal). Since Wundt there have been many efforts to establish the set of dimensions that most economically accounts for the similarities and differences in emotional experience, and there has been considerable disagreement about the number and nature of the dimensions (see Gehm & Scherer, 1988). While there is little doubt that valence is central, arousal is a more difficult case as it is not

1336

SCHERER

clear what kind of arousal or activation is meant*mental activation, sympathetic arousal, parasympathetic arousal? These are all quite different. The two-dimensional valence by arousal space seems to be considered basic on the basis of countless factor analyses that show stability for only these two dimensions. However, it is questionable whether this is not an artefact of methodology (the choice of the objects for similarity ratings, especially verbal labels; see Scherer, 2000b, pp. 184185). As mentioned above, recent work, using a more adequate, theoretically anchored feature profile-based similarity assessment yields four reliable factors for many different languages, with arousal coming in only third, after a control/mastery/power factor (Fontaine et al., 2007). It seems reasonable to assume that control/ mastery/power are very prominent criteria for adaptive responses and should be part of a primitive or core feeling read-out. The same is true for unexpectedness or novelty, which weighs in as the fourth factor. Indeed, there is an enormous amount of literature showing the basicness and primitivity of novelty detection in all organisms. Why should this central factor in perceiving and evaluating the world not be represented in core affect? Particularly as it is one of the central determinants of emotional attention (as a central aspect of relevance; Brosch & Scherer, 2009). A central problem with the core affect notion as proposed by Russell (2003, and 2009 this issue) and Barrett (2006, and 2009 this issue) is that there is little attempt to describe the mechanisms whereby core affect is produced. The authors provide a general list of factors involved in this process, including appraisal. However, no hypotheses or mapping rules, comparable to the appraisal predictions, are suggested that could be empirically tested. It is thus not clear how core affect is differentiated. This omission is particularly worrisome, as an enormous amount of information needs to be compressed and integrated to yield a single point in low-dimensional space, reflecting only evaluation (valence) and (arousal) response information. How does this work? One also wonders what happens to the large number of factors that are supposed to influence core affect. Are they not represented in feeling space or do the representations get lost once projection into low-dimensional space has occurred? And would those not be important to fine tune the adaptive action? It is equally unclear how, out of the constant flux of valence by arousal variation, an attribution to an object (Russell, 2003) or the assignment of a conceptual category (Barrett, 2006, and 2009 this issue) occur. As core affect is supposedly primitive and primary, waxing and waning, there must be some quality of core affect, a threshold or another criterion, that triggers the attribution and categorisation processes. What are these? It cannot be the evaluation of the objects or events, because if it were, it would not be clear how the theories differ from appraisal theories, except for an under specification of the respective mechanisms.

DYNAMIC ARCHITECTURE OF EMOTION

1337

Third, regarding categorisation, Barrett (2006) essentially enumerates a list of well-known categorisation mechanisms but refrains from making any theoretical predictions that can be empirically tested. In her paper in this issue she announces four specific hypotheses, namely: (1) there are psychological primitives (e.g., core affect); (2) emotions are like books of recipes (not mechanisms) with psychological primitives as elements in a wellstocked pantry that can be used to make any number of different dishes; (3) cognitions and perceptions are emotional in nature; and (4) emotion words are powerful in affecting experience. While the latter can be taken for granted, there are serious issues with the first three. Most importantly, it is difficult to consider these as specific hypotheses. How can they be falsified, especially as much seems to rest on definitional matters (and what does it mean that cognitions are emotional)? Fourth, the constructivist approach, as exemplified by the papers in this issue, adopts an outright ideographic approach by asserting that each individual constructs the category felt to fit the core affect in a given context on the basis of idiosyncratic input. Although it is certainly the case that the categorisation of immensely variable qualia feeling involves many idiosyncratic features that will be difficult to predict and examine empirically, the complete abandonment of the nomothetic approach threatens to lead to the abandonment of theory-guided empirical investigation, the hallmark of a scientific approach (Scherer, 1995). Barrett (2009 this issue) uses the metaphor of a book of recipes for her theory. But recipes generally imply rules for combining ingredients, not free construction. And these rules can be investigated: The proof of the pudding lies in the testing.

REVIEW OF EMPIRICAL EVIDENCE

Appraisal theories of emotion have stimulated numerous empirical studies over the last 20 years and there is now substantial experimental evidence for many of the predictions made. For an overview, the reader is referred to reviews in separate chapters of a volume on appraisal theories (Scherer, Schorr, & Johnstone, 2001; in particular, chapters by Johnstone, van Reekum, & Scherer; Kaiser & Wehrle; Pecchinenda, Roseman, & Smith; Scherer; Smith & Kirby). The research reviewed in this volume demonstrates that appraisal theorists have not only relied on self-report, but have also made extensive use of objectively measured indicators of appraisal processes such as physiological parameters and expressive behaviour. The use of self-report in this domain is often prematurely dismissed with the claim that individuals are not aware of rapid, implicit evaluation processes. Although this claim is true, it does not necessarily mean that individuals cannot recover or reasonably infer some of this information later. In addition, some appraisal processes occur on

1338

SCHERER

schematic or propositional levels that are accessible to awareness. Even if, in recall and report, participants rely in part on established schemata on canonical appraisals of certain types of events, this does not mean that the information is necessarily wrong, as social schemata contain a representation of regularities that often amounts to more than a kernel of truth. In any case, when one needs to resort to self-report on cognitive processes, this usually means no other method is available to gain access to the processes of interest. Under such circumstances, an imperfect approximation to assessment that will not produce proof but possibly plausibility is preferable to not studying the phenomenon at all (see Locke, 2009). In what follows, I review pertinent work from the Geneva Emotion Research Group that has empirically examined the CPM prediction that the different appraisal checks occur in a fixed sequence during a series of recursive cycles (the results for each check are continuously updated). In concluding, I describe preliminary studies on the important role of coherence or synchronisation in the emotion episode. I limit the following review of the evidence for efferent effects of the appraisal checks and the experimental investigation of the sequence hypothesis of the CPM as described earlier. Because of the rapidity of appraisal processing, often in the millisecond range, and the inaccessibility of much of these processes to consciousness, let alone to verbal description, appraisal markers other than verbal report have to be used. In consequence, the first part of the review of work in our laboratory concerns our attempt to confirm the theoretical assumption that appraisals have reliable signatures in the domains of brain electric activity, peripheral physiological changes, and expressive signals.

The efferent effects of appraisal checks on somatovisceral changes and motor expression as markers of appraisal results
We used a computer game to study the psychophysiological signatures of appraisal outcomes by manipulating intrinsic pleasantness (pleasant vs. unpleasant sounds accompanying central events) and goal conduciveness (winning*reaching the next level in the game, or losing*the destruction of ones spaceship) of game events in a factorial design (Johnstone, van Reekum, Hird, Kirsner, & Scherer, 2005; van Reekum et al., 2004). Participants played the game while cardiac activity, skin conductance, skin temperature, and muscle activity, as well as emotion self-reports, were assessed. Self-reports showed that game events altered levels of felt pride, joy, anger, and surprise. Goal conduciveness had little effect on forearm muscle activity (extensor) but was associated with significant autonomic effects, including changes to interbeat interval, pulse transit time, skin conductance, and finger temperature. The results of the autonomic measures are consistent with the CPM

DYNAMIC ARCHITECTURE OF EMOTION

1339

prediction of a higher level of sympathetic arousal related to obstructive compared to conducive events (provided an assessment of high coping potential). The manipulation of intrinsic pleasantness produced significant changes in skin conductance activity only. The obvious explanation for the lack of effects of intrinsic pleasantness manipulations on autonomic nervous system physiology is that intrinsic pleasantness rarely produces strong action tendencies that require increased sympathetic arousal. Speech following obstructive events was higher in energy and had a higher level of fundamental frequency (f0, heard as pitch), as indicated by f0 floor (lowest 5% of f0 values), than was speech following conducive events. These results suggest that physiological arousal was higher following the destruction of a ship than it was following the completion of a game level. This interpretation is supported by measurements of skin conductance (a measure that reflects sympathetic autonomic nervous system arousal), taken in a concurrent study, which were higher following obstructive events than they were following conducive events (van Reekum et al., 2004). In summary, this experiment revealed that variations in the intrinsic pleasantness of an event cause changes to spectral energy distribution, but not to overall energy, f0, or the measured temporal parameters and that changes to the conduciveness of an event are associated with changes to the latter set of variables but not to spectral energy distribution. Aue, Flykt, and Scherer (2007) presented participants, in the context of a memory task, with pictures displaying biological and cultural threat stimuli or neutral stimuli (stimulus relevance manipulation) with superimposed symbols signalling monetary gains or losses (goal conduciveness manipulation). Facial electromyogram and heart rate (HR) served as dependent variables. Results for the facial muscle innervation showed differential efferent effects for both stimulus relevance and goal conduciveness appraisal. Biological threat stimuli were associated with increased activity over the cheek region, which was explained by the participants adoption of a response pattern that resembled the fear grin in chimpanzees (Van Hooff, 1972). Consistent with the idea of a stronger need for effortful processing, because younger in human history, the vision of cultural threat stimuli led to higher activity over the brow region than the vision of both biological threat stimuli and neutral stimuli. Furthermore, as expected, increased activity over the cheek region (zygomaticus, smiling) was observed in the winning condition and over the brow region (corrugator, frowning) in the losing condition (but the latter for neutral pictures only). HR, in contrast, was influenced by the stimulus relevance manipulation only. As expected, HR acceleration was positively related to goal relevance, with strongest resource mobilisation for biological threat and least for neutral stimuli. The importance of stimulus relevance was confirmed by Aue and Scherer (2008), who had participants view unpleasant and pleasant pictures (intrinsic

1340

SCHERER

pleasantness appraisal), asking them to concurrently perform either an arm extension or an arm flexion, leading to an increase or a decrease in picture size. Increasing pleasant stimuli and decreasing unpleasant stimuli were considered goal conducive; decreasing pleasant stimuli and increasing unpleasant stimuli were considered goal obstructive (goal conduciveness appraisal). Facial electromyogram and HR were measured as dependent variables. As predicted by the CPM, the two appraisals differentially affected zygomaticus and corrugator responses, showing similar patterns of changes (i.e., pleasant events produced similar changes as conducive events; unpleasant events produced similar changes as obstructive events). However, results for HR did not support CPM predictions. Unpleasant images were associated with lower*not higher*HR change scores than were pleasant images. Comparison with earlier studies suggests that the adaptive cardiovascular effects of both appraisals may be mediated by stimulus proximity and motivational factors in the experimental setting. These results suggest that the predictions for facial muscle responses that are related to appraisal of stimulus valence, but not directly involved in action tendencies, can be conveniently studied in experimental settings that provide only indirect exposure to stimuli. As predicted, the zygomaticus and corrugator muscles, in particular, have proven to be reliable signatures for valence appraisals, both intrinsic pleasantness and goal conduciveness (see Aue & Scherer, 2008). In contrast, CPM predictions for sympathetic arousal, based on naturally occurring events such as imminent threat requiring immediate action, cannot be generalised to experimental situations that do not require vigorous action (as in Aue & Scherer, 2008). As shown by Aue et al. (2007), powerful biologically anchored threat stimuli (such as snakes and spiders) or cultural danger symbols (such as guns) may be appraised as sufficiently relevant to produce such rudimentary action tendencies. Experimental paradigms that present such highly relevant stimuli or that use goal conduciveness manipulations that have immediate consequences in the laboratory setting, such as winning or losing money or reputation, do show the predicted autonomic effects (Kreibig, Gendolla, & Scherer, 2009). Johnstone et al. (2007) examined voice and skin conductance changes following the appraisal of difficulty and reward in a computer task in which participants lost or gained points under two levels of difficulty. The rate at which the vocal folds open and close (fundamental frequency; f0) and changes in skin conductance were higher for loss than for gain, particularly when difficulty was high, suggesting high sympathetic arousal in challenging situations. The results provide evidence for the specificity of the efferent physiological effects of particular appraisal check combinations. An alternative approach to testing CPM patterning conditions consists in the analysis of expression patterns produced by actors asked to portray emotions that are characterised by specific appraisal profiles. Banse and

DYNAMIC ARCHITECTURE OF EMOTION

1341

Scherer (1996) used digital acoustic analysis to measure the vocal parameters in emotion portrayals by a group of professional actors and confirmed a sizeable proportion of the CPM predictions. Scherer and Ellgring (2007) used the same corpus to analyse the facial action units consistently used by actors to express certain emotions. The results are consistent with the assumption that particular appraisal checks generate facial expression units that can be interpreted as a functional consequence of certain appraisal results (Scherer, 1992). Although this type of evidence is indirect, it adds plausibility to the assumption that there are specific efferent effects of appraisal checks on somatovisceral changes and motor expression.

Sequential processing of appraisal checks

As described earlier, the CPM postulates that the SECs are processed in sequence, following a fixed order. This assumption concerning the microgenetic process is based on logical, phylogenetic, and ontogenetic arguments (Scherer, 1984b, 2001; Scherer et al., 2004b). In a first indirect test of this assumption, Scherer (1999b) demonstrated that the recognition of emotions consistent with scenarios that provide the information corresponding to the SECs is faster and more accurate if the information is given in the predicted order. Tests that are more direct must use the signatures for appraisal checks, as discussed in the preceding section. Thus, Aue et al. (2007) tested the sequence assumption with a memory task in which participants were presented with pictures displaying biological and cultural threat stimuli or neutral stimuli (stimulus relevance manipulation) with superimposed symbols signalling monetary gains or losses (goal conduciveness manipulation). Results for HR and facial electromyogram showed differential efferent effects of the respective appraisal outcomes and provided direct evidence for sequential processing. Specifically, as predicted, muscle activity over the brow and cheek regions marking the process of relevance appraisal occurred significantly earlier than did facial muscle activity markers of goal conduciveness appraisal. HR, in contrast, was influenced by the stimulus-relevance manipulation only. Lancto and Hess (2007) empirically tested the CPM hypothesis that the t intrinsic pleasantness evaluation occurs before the goal conduciveness evaluation. In two studies, intrinsically pleasant and unpleasant images were used to manipulate pleasantness, and a specific event in a Pac-Mantype video game was used to manipulate goal conduciveness. Facial EMG was used to measure facial reactions to each evaluation. As predicted, facial reactions to the intrinsic pleasantness manipulation were faster than facial reactions to the goal conduciveness manipulation. The authors interpreted these results as providing strong empirical support for the sequential nature of the appraisal process.

1342

SCHERER

Delplanque et al. (2009) investigated the effects of odours on appraisal processes and consequent emotional responses. The main goal was to test whether an odour is detected as novel or familiar before it is evaluated as pleasant or unpleasant. Participants performed a recognition task in which they were presented with pairs of unpleasant or pleasant odours (sample and target odours). Within a pair, the sample and target were either identical or different to assess participants novelty detection; unpleasant and pleasant target odours were contrasted to examine participants appraisal of intrinsic pleasantness. The authors measured facial expressions using electromyography and physiological reactions using electrocardiogram and electrodermal activity in response to odours. The earliest effects on facial muscles and heart rate occurred in response to novelty detection. Later effects on facial muscles and heart rate were related to pleasantness evaluation, confirming the existence of a sequence of appraisal checks for odours eliciting an emotional reaction. Grandjean and Scherer (2008) systematically manipulated novelty, goal relevance, intrinsic pleasantness, and goal conduciveness SECs in visual stimuli to test the sequence hypothesis in two experiments with electroencephalographic (EEG) recordings (characterised by a high temporal resolution). Topographical analyses of the event-related potentials (ERPs) revealed a specific electrical map related to novelty (90 ms after the onset of the stimulus) preceding another topographical map related to taskgoal relevance, indicating that the occurrence of the novel map precedes the task goal relevance map by about 50 ms. To investigate the effects of manipulated appraisals not revealed by the topographical analyses, Grandjean and Scherer further analysed the global field power (GFP), which revealed early effects related to novelty and later effects related to the intrinsic pleasantness factor. For the second experiment, in which intrinsic pleasantness and goal conduciveness were manipulated, the results confirmed that neuronal processing of intrinsic pleasantness precedes the effects related to goal conduciveness checks (see Figure 4). The frequency analyses revealed late effects in the gamma band, indicating an effect of goal conduciveness on the so-called induced gamma (e.g., Tallon-Baudry, Bertrand, Delpuech, & Pernier, 1996) at about 600 ms after the onset of the stimuli, suggesting that a high level of cognitive processing is involved in this type of appraisal. The results of these two experiments support the CPM predictions and suggest that novelty and intrinsic pleasantness may be appraised early, on an unconscious, automatic, and possibly schematic level, whereas goal conduciveness tends to be evaluated later in the sequence, on a conscious, controlled, and possibly propositional level (as predicted by Leventhal & Scherer, 1987). The results of the two studies are summarised in Figure 5.

DYNAMIC ARCHITECTURE OF EMOTION

1343

Figure 4. Electroencephalographic brain activity maps as markers of specic checks in the appraisal sequence*results of the topographical analyses and the global eld power (GFP) analyses (see Grandjean & Scherer, 2008, for further details). Note: Five panels with different results for Experiment 1: (A) Topographical results are shown for the experimental condition familiar/non-relevant (the different shades of grey in the graph for the time course correspond to the different topographical maps observed); the different maps (light grey corresponding to the positive, dark grey corresponding to the negative part of the electrical eld, and the upper part of the topographical maps corresponding to the anterior part of the head) are not specic for this experimental condition. (B) Specic novelty map for the second section. (C) Specic task/goal relevance map for the third section. (D) Timing of the two specic event-related potential topographical maps: the novelty map and the task/goal relevance map after the onset of the stimuli. (E) Timing of the GFP for the negative, positive, and neutral levels of intrinsic pleasantness after onset of the stimuli. The superimposed red rectangle indicates the time windows for which the positive and negative levels are signicantly different from the neutral level ( pB.05). To view this gure in colour, please visit the online version of this issue.

Synchronisation and the emergence of consciousness

Scherer (2005a) suggested that the emergence of the reflections of component changes into consciousness (see Figure 3) may occur when the degree of synchronisation of the different subsystems reaches a critical

1344

SCHERER

Figure 5. Mental chronography of the sequence of appraisal checks (see Grandjean & Scherer, 2008,
for details). To view this gure in colour, please visit the online version of this issue.

threshold. This hypothesis is exceedingly difficult to test, given the problems surrounding the definition and measurement of consciousness, as well as the absence of established models to determine the degree of synchronisation. However, recently Dan Glauser and Scherer (2008) reported a first investigation into the processes involved in the emergence of a subjective feeling. Assuming that the oscillatory brain activity presumed to underlie the emergence of a subjective feeling can be measured by EEG frequency band activity (similar to that shown in the literature for the conscious representation of objects), emotional reactions were induced in participants by using visual stimuli. Episodes for which participants reported a subjective feeling were compared with those that did not lead to a conscious emotional experience, in order to identify potential differences between these two types of reactions at the oscillatory level. Discrete wavelet transforms of the EEG signal in gamma (3163 Hz) and beta (1531 Hz) bands showed significant differences between these two types of reactions. In addition, whereas betaband activities were widely distributed, differences in gamma-band activity were predominantly observed in the frontal and prefrontal regions. However, contrary to our hypothesis, more oscillatory activity is present when the task of emotional monitoring is maintained throughout the image presentation without resulting in a subjective feeling report. The results are interpreted in terms of the complexity and dynamics of the processes required to perform the affective monitoring task and report a conscious feeling. In future

DYNAMIC ARCHITECTURE OF EMOTION

1345

studies, it will be necessary to include modifications of the time window examined and an extension of the frequency range of the oscillations analysed. However, the data demonstrate that the monitoring of electric brain activity at the surface is a potential means to examine synchronisation and the emergence of consciousness.

The categorisation and verbalisation of emotional experience

As shown in Figure 3 and described earlier, the next step concerns categorisation and verbalisation of the conscious feeling. Again, one faces a complex phenomenon that has rarely been the object of controlled empirical investigations. Our own efforts have been mainly directed toward the study of the dimensionality of the affective space. However, unlike proponents of circumplex models (Russell, 2003), we do not believe that a two-dimensional valence)arousal space maps a primitive core affect system. On the contrary, we submit that a rich representation of qualia can be projected, if a person is required to do so, into lower dimensional space. We have evidence against the claim that there are only two major dimensions, valence and arousal, and that emotion categories are arranged as a circle in this space. As suggested above, such a circle, which is not a true circumplex in the sense of Guttman (1954), may be the result of a biased choice of categories for similarity judgement tasks (Scherer, 2005b, p. 720). If a large and/or representative number of emotion categories are used, they are distributed over the total space and at least three, if not four, dimensions are needed to minimally account for important distinctions in this set, for example, between fear and anger. We can show that following valence, a control or power dimension is more important for emotion quality distinction than is arousal (Fontaine et al., 2007; Gehm & Scherer, 1988).

Conclusion
This contribution outlined the emergent dynamic architecture of emotion and briefly sketched the current state of the component process model of emotion, compared it with some of its major competitors, and summarised recent research from our laboratory. The dynamic aspect of emotion processes was emphasised, in particular the sequence of appraisal checks and the synchronisation of response systems. Lack of space prevented a demonstration of the models architecture as being ideally suited to explain the mechanisms that underlie group and cultural differences in emotional experience, even in highly similar situations (see Scherer, 1997b). Much remains to be done. For example, the CPM needs extension in the direction of event characteristics, particularly in terms whether it is a real or imagined event and whether it happened (or did not happen, even if it was

1346

SCHERER

possible) in the past or expected for the future. But it is hoped that this overview sensitises emotion researchers to a number of central, and often neglected, issues concerning the nature of the emotion architecture and its implications for temporal dynamics and individual differences. It will have served its purpose if it contributes to a strengthening of theory-guided empirical research on emotion.

REFERENCES
Anderson, N. H. (1989). Information integration approach to emotions and their measurement. In R. Plutchik & H. Kellerman (Eds.). Emotion: Theory, research, and experience: Vol. 4. The measurement of emotion (pp. 133186). New York: Academic Press. Aue, T., Flykt, A., & Scherer, K. R. (2007). First evidence for differential and sequential efferent effects of goal relevance and goal conduciveness appraisal. Biological Psychology, 74, 347357. Aue, T., & Scherer, K. R. (2008). Appraisal-driven somatovisceral response patterning: Effects of intrinsic pleasantness and goal conduciveness. Biological Psychology, 79, 158164. Bandura, A. (1997). Self-efcacy: The exercise of control. New York: Freeman. Banse, R., & Scherer, K. R. (1996). Acoustic proles in vocal emotion expression. Journal of Personality and Social Psychology, 70, 614636. Barrett, L. F. (2006). Emotions as natural kinds? Perspectives on Psychological Science, 1, 2858. Barrett, L. F. (2009). Variety is the spice of life: A psychological construction approach to understanding variability in emotion. Cognition and Emotion, 23(7), 12841306. Brosch, T., Sander, D., & Scherer, K. R. (2007). That baby caught my eye. Attention capture by infant faces. Emotion, 7(3), 685689. Brosch, T., & Scherer, K. R. (2009). Relevance and attention: An appraisal theory perspective on the emotional modulation of attention. Manuscript submitted. Cacioppo, J. T., Uchino, B. N., Crites, S. L., Snydersmith, M. A., Smith, G., Berntson, G. G., et al. (1992). Relationship between facial expressiveness and sympathetic activation in emotion: A critical review, with emphasis on modeling underlying mechanisms and individual differences. Journal of Personality and Social Psychology, 62, 110128. Canli, T., Qiu, M., Omura, K., Congdon, E., Haas, B. W., Amin, Z., et al. (2006). Neural correlates of epigenesis. Proceedings of the National Academy of Sciences, USA, 103, 1603316038. Costa, P. T., & McCrae, R. R. (1992). Manual for the Revised NEO Personality Inventory (NEOPI-R). Odessa, FL: Psychological Assessment Resources. Craig, A. D. (2002). How do you feel? Interoception: The sense of the physiological condition of the body. Nature Reviews Neuroscience, 3, 655666. Dan Glauser, E. S., & Scherer, K. R. (2008). Neuronal processes involved in subjective feeling emergence: Oscillatory activity during an emotional monitoring task. Brain Topography, 20, 224231. Darwin, C. (1998). The expression of the emotions in man and animals (3rd ed.). London: HarperCollins. (Originally published 1872, London: John Murray) Delplanque, S., Grandjean, D., Chrea, C., Aymard, L., Cayeux, I., Margot, C., et al. (2009). Sequential unfolding of novelty and pleasantness appraisals of odors: Evidence from facial electromyography and autonomic reactions. Emotion, 9, 316328. Dennett, D. (1988). Quining qualia. In A. Marcel & E. Bisiach (Eds.), Consciousness in modern science (pp. 4277). Oxford, UK: Oxford University Press. Dennett, D. (2001). Are we explaining consciousness yet? Cognition, 79, 221237.

DYNAMIC ARCHITECTURE OF EMOTION

1347

Dunn, J., & Brown, J. (1994). Affect expression in the family, childrens understanding of emotions, and their interactions with others. Merrill-Palmer Quarterly, 40, 120127. Ekman, P. (1984). Expression and the nature of emotion. In K. R. Scherer & P. Ekman (Eds.), Approaches to emotion (pp. 319344). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Ekman, P. (1992). An argument for basic emotions. Cognition and Emotion, 6, 169200. Ellsworth, P. C. (1994). William James and emotion: Is a century of fame worth a century of misunderstanding? Psychological Review, 101, 222229. Ellsworth, P. C., & Scherer, K. R. (2003). Appraisal processes in emotion. In R. Davidson, K. R. Scherer, & H. H. Goldsmith (Eds.), Handbook of affective sciences (pp. 572595). New York: Oxford University Press. Fontaine, J. R. J., Scherer, K. R., Roesch, E. B., & Ellsworth, P. E. (2007). The world of emotions is not two-dimensional. Psychological Science, 18, 10501057. Frijda, N. H. (1986). The emotions. London: Cambridge University Press. Frijda, N. H. (1987). Emotion, cognitive structure, and action tendency. Cognition and Emotion, 1, 115143. Frijda, N. H. (2007a). What emotions might be? Comments on the comments. Social Science Information, 46, 433443. Frijda, N. H. (2007b). The laws of emotion. Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Gehm, Th., & Scherer, K. R. (1988). Factors determining the dimensions of subjective emotional space. In K. R. Scherer (Ed.), Facets of emotion: Recent research (pp. 99114). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Germans, M. K., & Kring, A. M. (2000). Hedonic decit in anhedonia: Support for the role of approach motivation. Personality and Individual Differences, 28, 659672. Grandjean, D., Sander, D., Pourtois, G., Schwartz, S., Seghier, M., Scherer, K. R., et al. (2005). The voices of wrath: Brain responses to angry prosody in meaningless speech. Nature Neuroscience, 8(2), 145146. Grandjean, D., Sander, D., & Scherer, K. R. (2008). Conscious emotional experience emerges as a function of multilevel, appraisal-driven response synchronization. Consciousness and Cognition, 17, 484495. Grandjean, D., & Scherer, K. R. (2008). Unpacking the cognitive architecture of emotion processes. Emotion, 8, 341351. Griner, L. A., & Smith, C. A. (2000). Contributions of motivational orientation to appraisal and emotion. Personality and Social Psychology Bulletin, 26, 727740. Guttman, L. (1954). A new approach to factor analysis: The radex. In P. F. Lazarsfeld (Ed.), Mathematical thinking in the social sciences (pp. 258348). Glencoe, IL: Free Press. Hamlin, J. K., Wynn, K., & Bloom, P. (2007). Social evaluation by preverbal infants. Nature, 450, 557559. Henkel, R. D. (2000). Synchronization, coherence-detection and three-dimensional vision. Bremen, Germany: University of Bremen, Institute of Theoretical Physics. (Available from: http// www.g-kolleg.uni-bremen.de/research/papers/coherence.pdf retrieved 10 August 2008) Hirschfeld, L. A. (2001). On a folk theory of society: Children, evolution, and mental representations of social groups. Personality and Social Psychology Review, 5, 107117. Hofstede, G. (1984). Cultures consequences: International differences in work related values. London: Sage. Izard, C. E. (1977). Human emotions. New York: Plenum Press. Izard, C. E. (1993). Four systems for emotion activation: Cognitive and noncognitive processes. Psychological Review, 100, 6890. James, W. (1890). The principles of psychology. New York: Holt. Johnstone, T., van Reekum, C. M., Banziger, T., Hird, K., Kirsner, K., & Scherer, K. R. (2007). The effects of gain versus loss and difculty on vocal physiology and acoustics. Psychophysiology, 44, 827837.

1348

SCHERER

Johnstone, T., van Reekum, C. M., Hird, K., Kirsner, K., & Scherer, K. R. (2005). Affective speech elicited with a computer game. Emotion, 5, 513518. Johnstone, T., van Reekum, C. M., & Scherer, K. R. (2001). Vocal correlates of appraisal processes. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 271284). New York: Oxford University Press. Kaiser, S., & Scherer, K. R. (1998). Models of normal emotions applied to facial and vocal expressions in clinical disorders. In W. F. Flack Jr. & J. D. Laird (Eds.), Emotions in psychopathology (pp. 8198). New York: Oxford University Press. Kaiser, S., & Wehrle, T. (2001). Facial expressions as indicators of appraisal processes. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 285300). New York: Oxford University Press. Kirk, R. (1996). Raw feeling: A philosophical account of the essence of consciousness. Oxford, UK: Oxford University Press. Kreibig, S., Gendolla, G., & Scherer, K. R. (2009). Psychophysiological effects of emotional responding to goal attainment. Manuscript submitted for publicaation. Kuppens, P. (2005). Interpersonal determinants of trait anger: Low agreeableness, perceived low social esteem, and the amplifying role of the importance attached to social relationships. Personality and Individual Differences, 38, 1323. Kuppens, P., & Van Mechelen, I. (2007). Interactional appraisal models for the anger appraisals of threatened self-esteem, other-blame, and frustration. Cognition and Emotion, 21, 5677. Kuppens, P., Van Mechelen, I., & Meulders, M. (2004). Every cloud has a silver lining: Interpersonal and individual differences determinants of anger-related behaviors. Personality and Social Psychology Bulletin, 30, 15501564. Kuppens, P., Van Mechelen, I., & Rijmen, F. (2008). Towards disentangling sources of individual differences in appraisal and anger. Journal of Personality, 76, 9691000. Kuppens, P., Van Mechelen, I., Smits, D. J. M., De Boeck, P., & Ceulemans, E. (2007). Individual differences in patterns of appraisal and anger experience. Cognition and Emotion, 21, 689713. Lacewing, M. (2007). Do unconscious emotions involve unconscious feelings? Philosophical Psychology, 20, 81104. Lanctot, N., & Hess, U. (2007). The timing of appraisals. Emotion, 7, 207212. Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford University Press. Leventhal, H., & Scherer, K. R. (1987). The relationship of emotion to cognition: A functional approach to a semantic controversy. Cognition and Emotion, 1, 328. Lewis, C. I. (1929). Mind and the world order. New York: C. Scribners Sons. Locke, E. A. (2009). Its time we brought introspection out of the closet. Perspectives on Psychological Science, 4(1), 2425. Mandler, G. (1975). Mind and emotion. New York: Wiley. Matthews, G., Derryberry, D., & Siegle, G. J. (2000). Personality and emotion: Cognitive science perspectives. In S. E. Hampton (Ed.), Advances in personality psychology (Vol. 1, pp. 199237). Hove, UK: Psychology Press. McClelland, D. (1985). Human motivation. New York: Cambridge University Press. Metzinger, T. (2004). Being no one: The self-model theory of subjectivity. Cambridge, MA: MIT Press. Mischel, W., & Shoda, Y. (1995). A cognitiveaffective system theory of personality: Reconceptualizing situations, dispositions, dynamics, and invariance in personality structure. Psychological Review, 102, 246268. Oatley, K., & Johnson-Laird, P. N. (1987). Towards a cognitive theory of emotions. Cognition and Emotion, 1, 2950. .Ohman, A. (1986). Face the beast and fear the face: Animal and social fears as prototypes for evolutionary analyses of emotion. Psychophysiology, 23, 123145.

DYNAMIC ARCHITECTURE OF EMOTION

1349

Ortony, A., & Turner, T. J. (1990). Whats basic about basic emotions? Psychological Review, 97, 315331. Pecchinenda, A., Roseman, I. J., & Smith, C. A. (2001). In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research. New York: Oxford University Press. Plutchik, R. (1980). Emotion: A psychobioevolutionary synthesis. New York: Harper & Row. Pourtois, G., Grandjean, D., Sander, D., & Vuilleumier, P. (2004). Electrophysiological correlates of rapid spatial orienting towards fearful faces. Cerebral Cortex, 14(6), 619633. Power, M., & Dalgleish, T. (1997). Cognition and emotion: From order to disorder. Hove, UK: Psychology Press. Reisenzein, R., & Weber, H. (2009). Personality and emotion. In P. J. Corr, & G. Matthews (Eds.), The Cambridge Handbook of Personality Psychology. Cambridge: Cambridge University Press. Revelle, W., & Scherer, K. R. (2009). Personality and emotion. In D. Sander & K. R. Scherer (Eds.), Oxford companion to emotion and the affective sciences (pp. 303305). Oxford, UK: Oxford University Press. Rosch, E. (1973). Natural categories. Cognitive Psychology, 4, 328350. Roseman, I. J., & Smith, C. A. (2001). Appraisal theory: Overview, assumptions, varieties, controversies. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 334). New York: Oxford University Press. Rotter, J. B. (1966). Generalized expectancies for internal vs. external control of reinforcement. Psychological Monographs: General and Applied, 80, 128. Russell, J. A. (2003). Core affect and the psychological construction of emotion. Psychological Review, 110, 145172. Russell, J. A. (2009). Emotion, core affect, and psychological construction. Cognition and Emotion, 23(7), 12591283. Sander, D., Grandjean, D., & Scherer, K. R. (2005). A systems approach to appraisal mechanisms in emotion. Neural Networks, 18, 317352. Schachter, S., & Singer, J. E. (1962). Cognitive, social, and physiological determinants of emotional state. Psychological Review, 69, 379399. Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health: Assessment and implications of generalized outcome expectancies. Health Psychology, 4, 219247. Scherer, K. R. (1984a). Emotion as a multicomponent process: A model and some cross-cultural data. In P. Shaver (Ed.), Review of personality and social psychology: Vol. 5. Emotions, relationships and health (pp. 3763). Beverly Hills, CA: Sage. Scherer, K. R. (1984b). On the nature and function of emotion: A component process approach. In K. R. Scherer & P. Ekman (Eds.), Approaches to emotion (pp. 293317). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Scherer, K. R. (1986). Vocal affect expression: A review and a model for future research. Psychological Bulletin, 99, 143165. Scherer, K. R. (1987). Toward a dynamic theory of emotion: The component process model of affective states. Geneva Studies in Emotion and Communication, 1, 198. Scherer, K. R. (1988). On the symbolic functions of vocal affect expression. Journal of Language and Social Psychology, 7, 79100. Scherer, K. R. (1992). What does facial expression express? In K. Strongman (Ed.), International review of studies on emotion (Vol. 2, pp. 139165). Chichester, UK: Wiley. Scherer, K. R. (1994). Toward a concept of modal emotions. In P. Ekman & R. J. Davidson (Eds.), The nature of emotion: Fundamental questions (pp. 2531). New York: Oxford University Press. Scherer, K. R. (1995). In defense of a nomothetic approach to studying emotionantecedent appraisal. Psychological Inquiry, 6, 241248.

1350

SCHERER

Scherer, K. R. (1997a). Proles of emotionantecedent appraisal: Testing theoretical predictions across cultures. Cognition and Emotion, 11, 113150. Scherer, K. R. (1997b). The role of culture in emotionantecedent appraisal. Journal of Personality and Social Psychology, 73, 902922. Scherer, K. R. (1999a). Appraisal theories. In T. Dalgleish & M. Power (Eds.), Handbook of cognition and emotion (pp. 637663). Chichester, UK: Wiley. Scherer, K. R. (1999b). On the sequential nature of appraisal processes: Indirect evidence from a recognition task. Cognition and Emotion, 13, 763793. Scherer, K. R. (2000a). Emotions as episodes of subsystem synchronization driven by nonlinear appraisal processes. In M. D. Lewis & I. Granic (Eds.), Emotion, development, and selforganization: Dynamic systems approaches to emotional development (pp. 7099). New York: Cambridge University Press. Scherer, K. R. (2000b). Emotion. In M. Hewstone & W. Stroebe (Eds.), Introduction to social psychology: A European perspective (3rd. ed., pp. 151191). Oxford, UK: Blackwell. Scherer, K. R. (2001). Appraisal considered as a process of multilevel sequential checking. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 92120). New York: Oxford University Press. Scherer, K. R. (2004). Feelings integrate the central representation of appraisal-driven response organization in emotion. In A. S. R. Manstead, N. H. Frijda, & A. H. Fischer (Eds.), Feelings and emotions: The Amsterdam symposium (pp. 136157). Cambridge, UK: Cambridge University Press. Scherer, K. R. (2005a). Unconscious processes in emotion: The bulk of the iceberg. In P. Niedenthal, L. Feldman-Barrett, & P. Winkielman (Eds.), The unconscious in emotion (pp. 312334). New York: Guilford Press. Scherer, K. R. (2005b). What are emotions? And how can they be measured? Social Science Information, 44, 693727. Scherer, K. R. (2009). Emotion theories and concepts (psychological perspectives). In D. Sander & K. R. Scherer (Eds.), Oxford companion to emotion and the affective sciences (pp. 145149). Oxford, UK: Oxford University Press. Scherer, K. R., & Brosch, T. (2009). Culture-specic appraisal biases contribute to emotion dispositions. European Journal of Personality, 23, 265288. Scherer, K. R., & Ceschi, G. (1997). Lost luggage emotion: A eld study of emotionantecedent appraisal. Motivation and Emotion, 21, 211235. Scherer, K. R., & Ceschi, G. (2000). Criteria for emotion recognition from verbal and nonverbal expression: Studying baggage loss in the airport. Personality and Social Psychology Bulletin, 26, 327339. Scherer, K. R., & Ellgring, H. (2007). Are facial expressions of emotion produced by categorical affect programs or dynamically driven by appraisal? Emotion, 7, 113130. Scherer, K. R., Schorr, A., & Johnstone, T. (Eds.).(2001). Appraisal processes in emotion: Theory, methods, research. New York: Oxford University Press. Scherer, K. R., Wranik, T., Sangsue, J., Tran, V., & Scherer, U. (2004a). Emotions in everyday life: Probability of occurrence, risk factors, appraisal and reaction pattern. Social Science Information, 43, 499570. Scherer, K. R., Zentner, M. R., & Stern, D. (2004b). Beyond surprise: The puzzle of infants expressive reactions to expectancy violation. Emotion, 4, 389402. Schorr, A. (2001). Appraisal*The evolution of an idea. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 2034). New York: Oxford University Press. Schwartz, S. (1992). Universals in the content and structure of values: Theoretical advances and empirical tests in 20 countries. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 25, pp. 116). New York: Academic Press.

DYNAMIC ARCHITECTURE OF EMOTION

1351

Schwartz, S. (2006). A theory of cultural value orientations: Explication and applications. Comparative Sociology, 5, 137182. Seligman, E. P., Abramson, L. Y., Semmel, A., & Baeyer, C. von. (1979). Depressive attributional style. Journal of Abnormal Psychology, 88, 242247. Shaver, P. R., Schwartz, J., Kirson, D., & OConnor, C. (1987). Emotion knowledge: Further exploration of a prototype approach. Journal of Personality and Social Psychology, 52, 10611086. Smith, C. A., & Kirby, L. (2001). In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 121138). New York: Oxford University Press. Smith, C. A., & Lazarus, R. S. (1993). Appraisal components, core relational themes, and the emotions. Cognition and Emotion, 7, 233269. Stoeber, J., & Otto, K. (2006). Positive conceptions of perfectionism: Approaches, evidence, challenges. Personality and Social Psychology Review, 10, 295319. Tallon-Baudry, C., Bertrand, O., Delpuech, C., & Pernier, J. (1996). Stimulus specicity of phase-locked and non-phase-locked 40 Hz visual responses in humans. Journal of Neuroscience, 16, 42404249. Tolman, E. C. (1935). Psychology versus immediate experience. Philosophy of Science, 2, 356380. Tomkins, S. S. (1984). Affect theory. In K. R. Scherer & P. Ekman (Eds.), Approaches to emotion (pp. 163196). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Van Hooff, J. A. R. A. M. (1972). A comparative approach to the phylogeny of laughter and smiling. In R. A. Hinde (Ed.), Non-verbal communication (pp. 209241). Cambridge, UK: Cambridge University Press. van Reekum, C., Banse, R., Johnstone, T., Etter, A., Wehrle, T., & Scherer, K. R. (2004). Psychophysiological responses to appraisal responses in a computer game. Cognition and Emotion, 18, 663688. van Reekum, C. M., & Scherer, K. R. (1997). Levels of processing for emotionantecedent appraisal. In G. Matthews (Ed.), Cognitive science perspectives on personality and emotion (pp. 259300). Amsterdam: Elsevier Science. Vuilleumier, P. (2005). How brains beware: Neural mechanisms of emotional attention. Trends in Cognitive Sciences, 9(12), 585594. Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 92, 548573. Wiens, S. (2005). Interoception in emotional experience. Current Opinion in Neurology, 18, 442447. Wranik, T., & Scherer, K. R. (2009). Why do I get angry? A componential appraisal approach. In M. Potegal, G. Stemmler, & C. Spielberger (Eds.), International handbook of anger: Biological, psychological, and social processes. New York: Springer.