Emotion Review 2013 Scherer 150 62

http://emr.sagepub.com/Emotion Review

http://emr.sagepub.com/content/5/2/150The online version of this article can be found at:

DOI: 10.1177/1754073912468166

2013 5: 150Emotion ReviewKlaus R. Scherer

EvidenceThe Nature and Dynamics of Relevance and Valence Appraisals: Theoretical Advances and Recent

Published by:

http://www.sagepublications.com

On behalf of:

International Society for Research on Emotion

can be found at:Emotion ReviewAdditional services and information for

http://emr.sagepub.com/cgi/alertsEmail Alerts:

http://emr.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

What is This?

- Mar 20, 2013Version of Record >>

by ancuta anca on October 25, 2014emr.sagepub.comDownloaded from by ancuta anca on October 25, 2014emr.sagepub.comDownloaded from

http://emr.sagepub.com/

http://emr.sagepub.com/content/5/2/150

http://www.sagepublications.com

http://www.isre.org

http://emr.sagepub.com/cgi/alerts

http://emr.sagepub.com/subscriptions

http://www.sagepub.com/journalsReprints.nav

http://www.sagepub.com/journalsPermissions.nav

http://emr.sagepub.com/content/5/2/150.full.pdf

http://online.sagepub.com/site/sphelp/vorhelp.xhtml



Emotion ReviewVol. 5, No. 2 (April 2013) 150 –162

© The Author(s) 2013ISSN 1754-0739DOI: 10.1177/1754073912468166er.sagepub.com

Moors, Ellsworth, Scherer, and Frijda (2013) have surveyed a number of central questions in appraisal theory that are open to debate (see also Ellsworth & Scherer, 2003; Roseman & Smith, 2001). For this contribution, I have selected three issues for closer scrutiny from the vantage point of my brand of appraisal theory, the component process model (CPM; Scherer, 1984, 2001, 2009a): (a) varieties of relevance detection, (b) varieties of valence appraisal, and (c) sequential-cumulative effects of appraisal results. I will not attempt to survey the literature at large, but will concentrate on theoretical development and recent findings by our group (Geneva Emotion Research Group). Before setting out, I define the technical terms used in this article, in particular relevance and valence, and review the essentials of the CPM.

Terminological ConsiderationsIn what follows, the terms relevance, pertinence, significance, and conduciveness will be copiously used. Even though the words within this set are often seen as synonyms, I will not use them interchangeably, but will use each with its own distinctive meaning: relevance in the sense of an event having significant

and demonstrable bearing on the well-being of the individual, pertinence as a rather concrete and strong or decisive type of relevance, significance as indicative of the importance or quan-tity of consequences. Conducive is defined as “tending to pro-mote or assist,” whereas obstructive signifies “to block or close up by an obstacle.”

The term valence was introduced by Lewin (1938/1951), who defined +valence as forces that attract and −valence as forces that repel people, in other words, differentiating good, positive qualities from bad, negative qualities, as defined by approach–avoidance behavior tendencies (see also Sacharin, Sander, & Scherer, 2012). I have argued that different types of valence (Scherer, 2010) need to be distinguished and will defend this view here.

Description of the CPMLike most appraisal theories, the CPM is based on the assumption that emotions are elicited and differentiated by the results of the individual’s evaluation of events according to a set of appraisal criteria or stimulus evaluation checks (SECs). Table 1 shows the set of SECs postulated by the CPM, organized into four major

The Nature and Dynamics of Relevance and Valence Appraisals: Theoretical Advances and Recent Evidence

Klaus R. SchererSwiss Center for Affective Sciences, University of Geneva, Switzerland

Abstract

Appraisal theories of emotion have had a strong impact on the development of theory and experimental research in the domain of the affective sciences. While there is generally a high degree of convergence between theorists in this tradition, some central issues are open to debate. In this contribution three issues have been chosen for discussion: (a) varieties of relevance detection, (b) varieties of valence appraisal, and (c) sequential-cumulative effects of appraisal results. In addressing these issues, new theoretical ideas are suggested and an update of recent research on the sequence of appraisal processes is provided. Special emphasis is placed on nonverbal signatures of appraisal processes.

Keywordsappraisal sequence, appraisal signatures, goal conduciveness, relevance detection, valence

Author note: This research was supported by an Advanced Grant of the European Research Council (ERC 230331 PROPERMO) to Klaus Scherer and the Swiss Center for Affective Sciences. The author acknowledges pertinent comments by Marc Mehu, Marcello Mortillaro, Vera Sacharin, Katja Schlegel, and Jacobien van Peer.Corresponding author: Klaus R. Scherer, Swiss Centre for Affective Sciences, University of Geneva, Rue des Battoirs 7, CH-1205 Geneva, Switzerland. Email: [email protected]

468166 EMR5210.1177/1754073912468166Emotion ReviewScherer Relevance and Valence Appraisals2013

by ancuta anca on October 25, 2014emr.sagepub.comDownloaded from


Scherer Relevance and Valence Appraisals 151

thematic groups. The CPM assumes that these SECs exert their effects on other emotion components in a sequential and cumula-tive fashion (see The Dynamics of Appraisal: Sequential Checking section). Leventhal and Scherer (1987) proposed that each check can be processed on different levels of information processing (sensorimotor/schematic/conceptual) but that earlier checks, mostly driven by intrinsic stimulus characteristics, tend to be processed more easily and more rapidly at lower levels, whereas checks involving external inference require processing at higher levels. I have further developed the notion of levels of processing and now postulate four such levels (entailing different neural struc-tures and circuits): (a) a low sensorimotor level with a pattern-matching mechanism that is largely genetically determined, using criteria consisting of appropriate templates; (b) a schematic level, based on memory traces from social learning processes and occurring in a fairly automatic, unconscious fashion; (c) an asso-ciation level, involving various cortical association areas, which may occur automatically and unconsciously or in a deliberate, conscious fashion; and (d) a conceptual level, involving proposi-tional knowledge and underlying cultural meaning systems, requiring consciousness and effortful calculations in prefrontal cortical areas (see Scherer, 2009a, p. 1314).

The synchronization of the multicomponential response pat-terning driven by the appraisal results is seen as constitutive for the emergence of a felt emotional experience that can, under certain circumstances, be categorized and labeled by an estab-lished emotion term in the language spoken in the respective culture (see Scherer, 2001, 2009a, for further detail).

Varieties of Relevance DetectionThe central criterion for emotion-antecedent appraisal is the per-ceived relevance of the eliciting stimulus as a condition sine qua

non for emotion elicitation. The current version of the CPM pro-poses that novelty, intrinsic pleasantness, and pertinence for goals or needs are the most important criteria for relevance appraisal (Scherer, 2001), selecting stimuli that require further, in-depth treatment. A series of studies in our laboratory (Aue & Scherer, 2008, 2011; Gentsch, Grandjean, & Scherer, 2012; van Peer, Grandjean, & Scherer, 2012; van Reekum et al., 2004) sug-gests that conceptual refinement is necessary in order to derive specific hypotheses and, in particular, to model potential interac-tion effects between different sources of relevance appraisal.

In describing the central issue of personal relevance and goal pertinence, appraisal theorists have mentioned different classes of motivational variables such as tastes, needs, goals, values, norms, self-esteem, and expectations. There has been little effort to differentially define these partially overlapping concepts. Such definitions are particularly important for an interdiscipli-nary area such as the affective sciences, given that differences in the use of the same concept by scholars in different disciplines (e.g., philosophers, psychologists, economists, and social scien-tists) tend to be the rule rather than the exception. Thus, one of the major aims for appraisal theory is to further elucidate the different relevance criteria, for example, determined by individ-ual needs, values, and aspirations on the one hand and by social expectations, norms, and conventions on the other.

As a first step toward further theoretical development, I sug-gest exploring the varieties of relevance detection in greater detail and discussing potential research approaches. Traditionally, relevance detection has been defined in the CPM as an evalua-tion of whether a stimulus deserves further processing because of its bearing on our well-being—as determined by the results of the checks for novelty, intrinsic pleasantness, and goal relevance (see Table 1). These three subchecks are predicted to occur in this order because of (a) the relative efficiency as a filter for

Table 1. Stimulus evaluation checks, organized in four groups, illustrated with typical features describing the event or the effects on the person

Stimulus evaluation checks Event or behavior/person

Relevance Novelty Event is sudden, familiar, unpredictable Intrinsic pleasantness Event is in itself un/pleasant for the person Goal/need pertinence Event is important and relevant for person’s goals or needsImplications/consequences Causal attribution Event was caused by the person’s own/somebody else’s behavior/chance; caused un/intentionally Outcome probability Consequences of the event are predictable Discrepancy from expectation Event confirmed/is inconsistent with expectations Goal/need conduciveness Consequences of the event are positive/negative for person Urgency Event required an immediate responseCoping potential Control Person can control the consequences of the event Power Person has power over the consequences of the event Adjustment Person can live with the consequences of the eventNorm compatibility Internal standards Event incongruent with own standards and self-ideals External standards Event violated laws or socially accepted norms

Note: Adapted from Tables 1 and 4 in Scherer (2001).



152 Emotion Review Vol. 5 No. 2

attention, and (b) the respective ease and rapidity of the execu-tion of the checks. The first two checks are amenable to very rapid low-level processing, whereas the third one may require somewhat higher levels of processing and thus more time.

To highlight the fact that the novelty check focuses on the occurrence of an event (such as the appearance of an object, manifestation of a behavior, or other events) rather than on its quality (which is appraised by later checks), I suggest relabeling it novelty occurrence check. Three subchecks have been postu-lated for the novelty check—suddenness, familiarity, and pre-dictability. I surmise that the outcomes of the individual checks integrate into a novelty continuum ranging from very high (sud-den, unfamiliar, unpredictable) to very low (slow onset, famil-iar, highly predictable). I further suggest that the results of the subchecks are stored in a constantly updated appraisal register (see The Dynamics of Appraisal: Sequential Checking section of this article and Scherer, 2009a). The degree of novelty is likely to immediately determine the amount of attention devoted to an event and the depth of further processing during the subsequent checks.

The existence of a separate “intrinsic un/pleasantness” check has been frequently debated in the literature, one critique being that its outcome cannot be distinguished from the outcome of the goal conduciveness/obstructiveness check. This argument can be rejected based on both theoretical and empirical reasons. Theoretically the central argument is that the un/pleasantness of some stimuli is intrinsic to the stimulus, that is, that it does not depend on the motivational state of the appraiser. Empirically, the evidence reviewed in what follows shows (with replication in several studies) that the two checks concerned can be manip-ulated independently, in an orthogonal design, and that signifi-cant differences in the ensuing response patterns are found. The central argument in making the distinction is the focus on the intrinsic quality of the stimulus, largely independent of the motivational state of the appraiser or contextual factor. While I have focused only on intrinsic un/pleasantness in the past, I think that there are several classes of intrinsically relevant stimuli, such as evolutionarily prepared threat-related (snakes, spiders, anger expressions), reproduction-related (sex), taste-related (sweetness), or nurturance-related (baby faces; Brosch, Sander, & Scherer, 2007) stimuli, as well as strongly condi-tioned or highly overlearned stimulus classes from personal learning history. I therefore propose changing the label of this check to intrinsic relevance check, highlighting the central fea-ture of the appraisal process: the readiness to respond, relatively independently of motivational state or contextual factors, to par-ticular classes of stimuli with given characteristics. This con-ceptualization provides a direct link to the large literature on fear stimuli in which the notion of biological preparedness has loomed large (see Öhman & Mineka, 2001). Clayton and Myers (2009) have suggested the interesting concepts of biophilia and biophobia, postulating the human need to adapt to the environ-ment by approaching positive and avoiding negative aspects, which is likely to have created an intrinsic, genetically-based predisposition to associate with and depend on the natural envi-ronment. The renaming of this check to intrinsic relevance does

not imply any change in my earlier conceptualization of intrin-sic un/pleasantness. It mainly extends the class of stimulus char-acteristics for which the appraisal is mostly driven by their intrinsic quality rather than motivational or contextual factors. It also does not mean that these stimuli are assumed to be exclu-sively treated on the lowest level of processing; rather, they may also be processed on higher levels. For example, the appraisal of the solution for a mathematical problem obviously requires con-ceptual processing, but the result can be determined by intrinsic epistemic beauty and elegance.

Whereas novelty relevance is detected by expected occur-rence and intrinsic relevance by specific types of stimuli, rele-vance criteria are much more complex for the third type of relevance check, which concerns what one might call motiva-tional relevance classes. These do not consist of specific classes of stimulus types. Rather, the defining criterion is that the respec-tive need, goal, or value of the appraiser on which the stimulus has a direct bearing must be salient or of high priority at the occurrence of the event. For example, food cues have been shown to be less relevant and attention-grabbing in a state of satiety (see Sacharin et al., 2012, for a review). Thus, for basic needs (food, sex, social contact), relevance appraisal may depend on the current motivational state. For goals in a means–end action plan, for example, executing a complex task, motivational relevance may depend on context, distance from goal attainment, or probability of success as evaluated in the appraisal of implica-tions. In the past, I have generally used the term “goal relevance” to cover this large array of motivational states. However, apart from the fact that the notion of a goal has a restrictive meaning for many people (laymen and scientists alike), it has a somewhat static connotation. Frijda (1986) has pointed out that relevance needs to be assessed with respect to current concerns, that is, needs, goals, values, norms, self-esteem, expectations, and so forth, that are particularly pertinent at the time. This provides a much more dynamic view of the underlying process, and it underlines that a very high level of relevance is required. In con-sequence, I suggest changing the label of this relevance check from goal relevance to concern pertinence. In a further step, one might want to further develop the distinctions between these motivational classes, with continuously operative needs being mostly biologically based and thus universal, values being a high-level construct of desirable qualities and achievements and thus in large part shared by groups and cultures, goals (with a definite end state) being more concrete objectives toward which action is directed and thus likely to show a large variability over individuals and time, and norms being strictly constrained socio-cultural codices for required behavior.

I now turn to promising research approaches in the domain of relevance appraisal, concentrating in large part on the work conducted in our group. Differences in relevance detection for the three checks described before can be examined experimen-tally, both individually and in combination. The experimental manipulations would be determined by the relevance class: types of stimuli for novelty and intrinsic relevance, manipula-tion of motivational need state and task goals for concern perti-nence, and systematic participant selection for values and social




norm relevance. The central prediction is that relevant stimulus events are, at an early stage, more rapidly identified and draw more attention, and, at a later stage, processed more intensively or deeply and enter more readily into memory. The CPM assumes sequential, cumulative processing and so the early detected intrinsic, stimulus-bound relevance value of each class of stimuli will, in the ensuing steps of the appraisal process, interact with the results of further checks, as well as with con-text factors. One of the aims of this research should be to deter-mine the relative importance of these factors and the nature of the underlying dynamic functions.

In addition to the systematic manipulation of specific appraisal checks (Ellsworth & Smith, 1988; Roseman & Evdokas, 2004; Smith, 1989; van Reekum et al., 2004), there are several established task paradigms, such as memory and selective attention tasks. The contributors to a review of appraisal research (Scherer, Schorr, and Johnstone, 2001) pointed out a number of nonverbal techniques for the objec-tive behavioral assessment of appraisal processes. The most frequently used methods consist of psychophysiological measurement (Pecchinenda, 2001), including the assessment of facial expression (Kaiser & Wehrle, 2001) via electromyo-graphy (EMG). Thus, Aue, Flykt, and Scherer (2007) manip-ulated intrinsic pleasantness and goal conduciveness and measured a set of psychophysiological variables in a picture-viewing task, including cardiovascular measures and facial EMG. The results showed facial EMG differences for m. zygomaticus and m. corrugator innervations for both checks, whereas heart rate differentiated only the levels of the rele-vance condition, suggesting efferent response differences between intrinsic pleasantness and goal conduciveness.

The development of techniques such as electroencephalo-graphic (EEG) and functional magnetic resonance imaging (fMRI) assessment has been very useful for examining brain activity responses underlying attention deployment to stimuli with differ-ent kinds of relevance for the organism (Dan Glauser & Scherer, 2008; Grandjean & Scherer, 2008; Kalisch, Wiech, Critchley, & Dolan, 2006). Much of this work has focused on negative— particularly fear-relevant—stimuli (see Öhman & Mineka, 2001). In contrast, appraisal theories of emotion posit a more general mechanism, predicting attention capture by stimuli that are intrin-sically relevant for the organism or for momentary needs or goals, regardless of the direction of valence. Our group has recorded event-related brain potentials (ERPs) from 20 subjects performing a dot-probe task in which the cues were fear-inducing and nurturance-inducing stimuli (i.e., anger faces and baby faces; see also Brosch et al., 2007). Highly similar validity modulation was found for the Positive Component 1 (P1) time-locked to target onset, indicating early attentional capture by both positive and negative emotional stimuli. Topographic segmentation analysis and source localization indicate that the same amplification process was involved whether attention orienting was triggered by negative, fear-relevant stimuli or by positive, nurturance-relevant stimuli. These results confirm that biological relevance, and not exclusively fear, produces an automatic spatial orienting toward the location of a stimulus (Brosch, Sander, Pourtois, & Scherer, 2008).

One of the major shortcomings in this research domain is the lack of a valid verbal self-report measurement of relevance. We have used self-report of affective impact (“being touched by,” “had a strong effect on me”) with considerable success as a measure of appraised relevance. Scherer, Dan, and Flykt (2006) asked two groups of participants to rate their evaluation of 59 pictures from the International Affective Picture System (IAPS) on a profile of nine appraisal criteria (including impact due to relevance) and on the classic dimensions of affective meaning (valence, arousal, potency). The ratings on the dimensions of valence, arousal, and potency correlate differentially with spe-cific appraisal ratings. However, contrary to what one may expect, the reported impact of a photo did not relate linearly to reported arousal (r = .15, n.s.). Rather, the results suggest a cur-vilinear relationship between reported impact and judged arousal, in that very low arousal photos (such as a mother with her baby) and highly arousing photos (such as a starved child) can have a strong affective impact (see Murphy, Hill, Ramponi, Calder, & Barnard, 2010, for a confirmation of the utility of using impact scales).

Finally, it has to be stressed that individual differences in relevance appraisal might or might not, depending on the emo-tions concerned, significantly mediate the processes involved, and it is thus highly desirable that more attention is paid to this fact. Apart from testing hypotheses suggested by the literature, it might be of interest to examine a general “sensitivity for rel-evance,” some individuals potentially being more readily “touched” by different situations. One can expect that the pres-ence of appraisal biases in individuals, which may be rein-forced by cultural value systems, will systematically influence relevance detection, sometimes to the point of contributing to the etiology of emotional disturbances (see Scherer & Brosch, 2009, for an overview).

Varieties of Valence AppraisalIn the affective sciences, valence is generally considered as one homogeneous, continuous dimension (e.g., positive vs. nega-tive; pleasurable vs. painful; more of this vs. less of that). In contrast, the CPM proposes two different types of valence appraisal—intrinsic un/pleasantness and goal conduciveness/obstructiveness. Our group has produced some preliminary evidence that shows the utility of this distinction. Aue and Scherer (2011) examined the somatovisceral response profiles for these two types of valenced appraisal results. Participants viewed unpleasant and pleasant pictures (intrinsic pleasant-ness) and performed either goal conducive (i.e., decreasing the size of unpleasant pictures, increasing the size of pleasant pic-tures) or goal obstructive (i.e., increasing the size of unpleasant pictures, decreasing the size of pleasant pictures) arm move-ments. The data suggest that the two appraisals have somewhat similar, but not identical, response patterns (see also Aue & Scherer, 2008). Specifically, main effects showed that intrinsic pleasantness and goal conduciveness were simultaneously reflected in zygomaticus activity, and their effects pointed in the same direction. Interaction effects showed that the effect of




intrinsic pleasantness was unmistakably present for both levels of goal conduciveness in both zygomaticus and corrugator activity. However, goal conduciveness effects were restricted to pleasant images only. Whereas facial EMG demonstrated more robust effects of intrinsic pleasantness than goal condu-civeness, mean skin conductance, forehead temperature, and finger temperature were influenced by the goal conduciveness manipulation but not by the intrinsic pleasantness manipula-tion. Obstructive events were characterized by stronger physi-ological mobilization, more specifically, by less habituation in mean skin conductance over the course of the experiment (con-sistent with findings reported in van Reekum et al., 2004). These results emphasize the importance of distinguishing between intrinsic pleasantness and goal conduciveness and suggest that (a) the efferent effects of the two appraisals com-bine multiplicatively, and (b) the predictability of goal condu-civeness may influence the impact of the respective appraisals on somatovisceral responding.

Recently, I have suggested that it is both theoretically fruitful and empirically sound to hypothesize at least six types of quali-tatively different valences (Scherer, 2010). This proposal, based on the assumption that for each type of valence there are spe-cific appraisal objectives and possibly also different appraisal mechanisms, is illustrated in Table 2. Concretely, it is assumed that each of the SEC checks has a built-in valence aspect giving rise to a specific type of valence: familiarity, pleasure, content-ment, power, self-worthiness, and moral worthiness. The claim is that, although all of these have something in common, namely, the fundamental implications of approach and avoidance ten-dencies used by Lewin (1951) as the basis for introducing the notion of valence, they are also differentiated by the special quality of valence that is due to the underlying appraisal crite-rion and its efferent results. In addition, different types of valence could be the result of appraisal at different levels of processing (as described in the brief description of the CPM in the first part of this article). The different levels continuously interact, producing top–down and bottom–up effects (Scherer, 2009a). It seems plausible that intrinsic relevance, for example the pleasantness of sensory stimuli such as smell, taste, or visual beauty, or other intrinsically relevant stimuli such as snakes and baby faces, is processed at a lower level than goal conducive-ness, which would seem to require at least associative and sometimes conceptual processing. Sacharin et al. (2012) further develop and justify the argument and provide suggestions for

further research. In particular, they discuss how different types of valence (microvalences) relate to one-dimensional valence (macrovalence), arguing that different microvalences, as gener-ated by different appraisals, can be integrated into a macrova-lence, as constituted by a single, homogeneous valence dimension, which can serve as a “common currency” in decision making.

Indirect support for the notion of different types of valence is provided by the results of a large-scale investigation of the semantics of emotion words. Following a proposal to define the meaning of emotion words by differential profiles of features representing all components of emotion (Scherer, 2005), Fontaine, Scherer, and Soriano (in press) have conducted a massive cross-cultural study—the GRID study—in over 30 different countries with 25 different languages. Native speak-ers were asked to indicate the probability that a coherent profile of specific component features (e.g., specific appraisals, bodily reactions, facial and vocal expressions, action tendencies, and feelings) would be present when a specific word was used to describe the emotion that a person was experiencing. Similarly, the use of a specific emotion word to describe an emotional episode was expected to conjure up the corresponding profile of characteristic component changes in the mind of the receiver. The term GRID refers to the matrix formed by component fea-tures and emotion words. The resulting intercultural data set strongly confirms a preliminary analysis in a small number of Western countries (Fontaine, Scherer, Roesch, & Ellsworth, 2007) showing that the affective space spanned by 24 major emotion terms requires, at a minimum, four dimensions to allow reasonable discrimination of the terms—valence, power/control, arousal, and unpredictability (or novelty), in this order of importance.

In the present context, it is of interest to discuss the results of a factor analysis of the 31 features representing the appraisal component (representing the set of SECs proposed by the CPM, as well as a number of features proposed by other appraisal theories). Two factors were extracted on the basis of a scree test and were stable across four language groups. The factor loadings, shown in Table 3, suggest that the first factor is a clear Valence factor, whereas the second factor can be called Novelty. The loadings on the first factor illustrate the different types of valence shown in Table 2: intrinsic pleasantness, con-firmed expectations and goal conduciveness, coping potential (in the sense of adjustment), and compatibility with normative

Table 2. Varieties of valence as a function of stimulus evaluation check outcome

Stimulus evaluation checks Evaluation outcomes/feeling dimensions Type of valence

Novelty Unknown–familiar FamiliarityIntrinsic pleasantness/beauty Pleasant–unpleasant PleasureGoal/need conduciveness Satisfied–disappointed ContentmentCoping potential Strong–weak PowerCompatibility self-standards Achieved–failed Self-worthinessCompatibility norms/values Virtuous–wicked Moral worthiness

Note: Adapted from Scherer (2010).




or moral standards. Interestingly, the novelty check loads on both factors—the familiarity/expectedness pole of novelty loads on the first factor, valence, and the suddenness and unpre-dictability pole of novelty loads on a separate factor that is orthogonal to valence (see Scherer & Fontaine, in press a, for further discussion).

One might argue that the fact that the large majority of appraisal items load on a general valence factor is difficult to reconcile with the notion that specific appraisal outcome pro-files or configurations allow differentiation of emotions. However, it often has been observed that different facets load on a general superfactor (Gignac, 2008). For example, in the domain of psychometric intelligence research a superfactor “g,” a general ability factor underlying various types of competen-cies, has been identified. Our analysis shows that the emotion domain also has its superfactor “g”—“general valence”—underlying various types of valenced appraisal criteria. Thus, the different SEC outcomes load together on one factor in the two-factorial solution shown in Table 2 because they all have,

like small “g” for cognitive ability, a component of valence. Thus, different domain-specific valences as part of the major SECs contribute to an overall macrovalence factor.

In a next step, we partialed out the superfactor “valence” to explore the factorial structure of the remaining variance (using the residual scores of the appraisal features) and obtained the factor structure shown in Table 4. The six factors that emerge can be labeled as follows: Novelty/Chance Cause, Coping Ability, Expected/Familiar, Goal Relevance, Norm Violation, and Self vs. Other Cause (see Scherer & Fontaine, in press a, for details). It should be noted that these factor loadings do not rep-resent a pure logical structure of the appraisal system, but are affected by ecological correlations (e.g., if events are unpredict-able, they are likely to be inconsistent with expectations, and unexpected and unpredictable events often occur by chance and require an urgent reaction; see Scherer & Fontaine, in press a). These results provide indirect evidence (given that we study “frozen” structures of lexical meaning) for the plausibility of assuming qualitatively different types of microvalence that, in

Table 3. Factor loadings of the appraisal features in the data set of the intercultural GRID study

Valence appraisal Novelty appraisal

Consequences positive for person 0.97 0.04In itself pleasant for the person 0.96 0.07Consequences positive for somebody else 0.93 0.07Important and relevant for person’s goals 0.88 0.09In itself pleasant for somebody else 0.88 0.11Important and relevant for goals of somebody else 0.84 0.07Confirmed expectations 0.81 −0.26Consequences able to live with 0.79 −0.11Centre of attention 0.73 0.12Familiar 0.70 −0.28Consequences predictable 0.53 −0.48Caused by the person’s own behavior 0.47 −0.36Inconsistent with expectations −0.55 0.48In itself unpleasant for somebody else −0.72 −0.09Consequences negative for somebody else −0.76 −0.12In danger −0.81 0.26Violated laws or socially accepted norms −0.83 0.01Irrevocable loss −0.86 0.05Incongruent with own standards and ideals −0.91 0.02Treated unjustly −0.92 0.05In itself unpleasant for the person −0.93 0.02Consequences negative for person −0.94 0.01Unpredictable −0.04 0.84Suddenly 0.15 0.76Caused by chance 0.42 0.64Required an immediate response −0.07 0.56Caused by a supernatural power 0.44 0.49Caused by somebody else’s behavior −0.09 0.34Caused intentionally 0.19 −0.23Consequences avoidable or modifiable −0.35 −0.37Enough resources to avoid or modify consequences −0.03 −0.41

Note: Based on a two-factorial varimax rotated structure; reproduced from Scherer and Fontaine (in press a).




combination, can generate a multitude of different emotions (possible mechanisms are discussed in Sacharin et al., 2012).

To summarize: I suggest that the outcomes of most appraisal checks are not neutral but valenced and that the type of valence is different in each case (having different feeling qualities and generating different response patterns). Because of the common valence component, these different outcomes (microvalences) tend to be integrated into a macrovalence dimension, which cor-responds to the classic pleasantness or evaluation dimension in dimensional theories of emotion. This corresponds to a projec-tion from higher dimensional to lower dimensional space. The empirical emergence of a superfactor “general valence” in the GRID study (with emotion words as elements) suggests that for most prototypical emotions the different appraisal outcomes are valenced in the same direction—due to structural constraints or ecological correlations. However, there may be many cases where the microvalence outcomes of different appraisal checks do not match, for example when my boss insists on my drinking a glass of spectacular vintage Bordeaux wine on a day on which I am strictly forbidden to drink any alcohol for medical reasons (intrinsic pleasantness of the wine, goal obstructiveness in terms of health, and low power and control to refuse). Not surpris-ingly, such ambivalent emotions are rarely studied; possibly they are rare events. However, their occurrence warrants research attention on the role of appraisal-check specific microvalences and their integration into overall feeling.

The GRID study also provides further indirect evidence for the claim that the appraisal results drive the differentiation of emotions. We tested the extent to which the 24 emotion terms under investigation can be classified on the basis of profiles of

appraisal features alone, using multiple discriminant analysis (MDA). In the GRID instrument, we use a total of 142 features to represent all of the emotion components (appraisal, bodily reactions, expression, action tendencies, and feeling). In an overall MDA, these 142 features allow us to classify the 24 emotions with a cross-validated hit rate of 82.1%. If we use only the 31 features representing the appraisal component, we reach a cross-validated hit rate of 70.7%, only about 10% less than for the combined discriminative power of all 142 component fea-tures. Successively adding other components in a series of MDAs shows that adding action tendencies adds about 5% to the hit rate, consequently adding bodily reactions and expres-sion adds another 5%, and finally, adding feelings adds the remaining 2% (see Scherer & Fontaine, in press b).1 The finding that the appraisal feature profiles alone allow us to explain the lion’s share of the variance in the semantic differentiation of the emotion terms is highly consistent with the claim of appraisal theories (especially the CPM) that the appraisal results drive the changes in the other components and produce a level of syn-chronization or coherence that is constitutive for the occurrence of an emotion episode and for the meaning of the respective emotion word.

The Dynamics of Appraisal: Sequential CheckingAs mentioned at the outset, the CPM has a dynamic architecture based on the assumption that four major groups of SECs have sequential-cumulative effects on all other emotion components.

Table 4. Factor loadings of the theoretically predicted appraisal features after partialling out the superfactor outcome valence (“g”) in the data set of the intercultural GRID study

Novelty/ chance cause

Coping ability

Expected/ familiar

Goal relevance

Norm violation

Self vs. other cause

Suddenly 0.87 −0.10 0.02 0.01 −0.04 0.02Unpredictable 0.85 −0.17 −0.12 −0.01 0.02 0.13Caused by chance 0.75 −0.08 −0.05 0.15 0.08 0.03Inconsistent with expectations 0.64 0.07 −0.11 −0.10 0.11 0.32Required an immediate response 0.63 0.19 −0.25 0.16 0.09 −0.11Enough resources to avoid or modify consequences −0.05 0.85 0.12 −0.12 0.06 −0.16Consequences avoidable or modifiable −0.05 0.79 0.17 −0.02 0.25 −0.16Consequences able to live with −0.01 0.71 0.14 0.31 −0.09 0.20Caused intentionally 0.05 0.14 0.73 −0.19 0.22 0.13Confirmed expectations −0.28 0.00 0.67 0.38 −0.05 −0.15Familiar −0.17 0.15 0.63 0.16 −0.26 0.04Consequences predictable −0.23 0.38 0.62 0.11 −0.09 −0.20Important and relevant for person’s goals 0.15 0.00 0.06 0.83 −0.04 −0.14Important and relevant for goals of somebody else 0.04 0.07 0.09 0.81 −0.01 0.25Violated laws or socially accepted norms 0.09 0.01 −0.01 −0.10 0.85 −0.24Incongruent with own standards and ideals 0.07 0.17 −0.06 0.05 0.81 0.24Caused by somebody else’s behavior 0.33 −0.02 0.33 0.15 0.05 0.69Caused by the person’s own behavior −0.02 0.26 0.39 0.02 0.09 −0.67

Note: Based on a six-factorial varimax rotated structure; reproduced from Scherer and Fontaine (in press a).




As this feature of the CPM is often misinterpreted, I will pro-vide a somewhat more extensive description here, followed by the empirical evidence to date.

From the very inception of the CPM (Scherer, 1982, 1984), I have postulated that the SECs exert their effects—for both logi-cal and economical reasons—in a sequential fashion. It seems plausible to assume that if the results of an SEC provide essential information to judge a criterion in another SEC, the former must yield a result before the latter can be processed. Logically, I can determine my potential to cope with the consequences of an event only if I know what they are. And it would be extremely uneconomical to invest precious resources before I am reasona-bly certain about the outcome of any actions I may want to pre-pare for. Apart from adducing logical and economical reasons, it can be argued that the microgenetic unfolding of the emotion-antecedent appraisal processes parallels both phylogenetic and ontogenetic development in the differentiation of emotional states. The earlier SECs, particularly the novelty and the intrinsic pleasantness check, seem to be present in most animals as well as in newborn humans. Thus it seems plausible that these low-level processing mechanisms are hard-wired detection capacities and occur very rapidly after the occurrence of a new stimulus. More complex evaluation mechanisms are successively devel-oped at more advanced levels of phylogenetic and ontogenetic development, with natural selection operating in the direction of more sophisticated information-processing ability in phylo-genesis, and with maturation and learning increasing the indi-vidual’s cognitive capacity in ontogenesis (see Scherer, 1984, pp. 313–314; Scherer, Zentner, & Stern, 2004).

As shown in Figure 1, all SECs are expected to be processed simultaneously, starting with relevance detection. However, the essential criterion for the sequence assumption is the point in time at which a particular check achieves preliminary closure, that is, yields a reasonably definitive result, one that warrants efferent

commands to response modalities (the diamond shapes in Figure 1). The darker coloration around the diamond shapes suggests more intense processing as preliminary closure of a check is achieved. Resource-intensive information processing should occur for only those stimuli that have been appraised as highly 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 appraisal and preparation of behavio-ral 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. Clearly, the causes and implica-tions of the event need to be established before the organism’s cop-ing potential can be conclusively determined, as the latter must be evaluated for a specific situational demand. The check for norma-tive significance is predicted to occur last, in part because it is likely to require a higher level of processing than some of the other SECs because it may involve the evaluation of complex rules.

Figure 1 illustrates that once the processing of a given SEC has produced a sufficiently stable result to justify a specific reaction (closure), there are efferent effects on all other compo-nents. One essential effect is exerted on the appraisal compo-nent itself by providing the necessary information for the subsequent checks, as described before. In addition, there are efferent effects on the somatovisceral, action tendency, and motor expression components. The effects of the different SEC groups are sequential and cumulative, as the efference of each subsequent check will modulate and interact with (or some-times replace) the effects of earlier checks. In addition, the set of curved arrows shown in the right margin of Figure 1 illus-trates (without implying precise specification) the existence of multiple direct interactions between the component subsys-tems, including the possibility that the cognitive appraisal pro-cesses can be affected by changes in the other components.

Figure 1. Sequential-cumulative efferent effects of appraisal results on different components of emotion (slightly modified after Table 5.2 in Scherer, 2001).




Figure 1 shows only one cycle of what is in reality a constant recursive process in which the results of each cycle are fed into the next cycle. The CPM predicts particular efferent change pat-terns for the results of each SEC. It is important to emphasize the ubiquity of interaction effects between SECs in the temporal sequence—each SEC result can modify the nature of the effer-ent changes produced by subsequent checks. In other words, the patterning of the component states is specific to the unique eval-uation “history” of the respective stimulus. For example, an unpleasant odor that was expected will yield a componential patterning that is different from an unexpected one because the changes produced by the preceding step are different. In other words, the SECs and the effects of their results on the other subsystems are not independent of each other; rather, each pre-ceding SEC result and the change produced by it “sets the scene” for the effects of the following SEC result.

Looking at it in another way, specific “patterns” of compo-nent states (e.g., those that seem to characterize emotions such as anger or fear) can only occur if there is a corresponding pattern of specific SEC results in which each SEC adds a par-ticular modification or “added value” in a complex sequential interaction. The result of this patterning of the appraisal results and the continuous efference to the other components is a form of synchronization of the states of the organismic subsystems concerned, constituting a higher degree of coherence of these systems, in the interest of adapting to the new situation created by the occurrence of the emotion-eliciting event (see Scherer,

2001, for more detail). The relatively homogeneous synchro-nization pattern that characterizes a specific type of emotion episode can be described as an attractor basin in the language of nonlinear systems theory (see Scherer, 2000, 2009b). As described in greater detail elsewhere (Scherer, 2001, 2009a), the appraisal-driven synchronization of the components may lead to emergent consciousness on the level of the feeling component. These feelings represent all of the accumulated changes in the components (and the consolidated results of the appraisal process) and thus have very specific qualia. In some cases, for example, when a person wants to engage in social sharing of an emotion episode, the feeling qualia are catego-rized and labeled with a word or an expression (sometimes a metaphor). The emotion episode will continue until massive changes in the appraisal (e.g., a problem is resolved or a new event intervenes) lead to a desynchronization or new synchro-nization pattern.

In response to critics of the sequence assumption I have insisted that it is entirely consistent with the notion of parallel processing (see Scherer, 2001, for a detailed argument). The reason is that I assume a type of appraisal register (rapidly accessible storage space for elementary information such as intermediate calculation results), the contents of which are con-tinuously updated in each recursive loop (see Figure 2). In par-allel processing, all checks are always performed and separate registers for each SEC hold temporary values that consist of the best available estimates of the respective criterion. An efferent

Figure 2. Sketch of the potential architecture of the appraisal process as part of a general information-processing system, separately driving peripheral support systems and alternative action tendencies (slightly modified after Figure 5.3 in Scherer, 2001).Note: NES: neuro-endocrine system; ANS: autonomic nervous system; SNS: somatic nervous system; Nov: novelty; Plea: pleasantness; Rel: relevance; Con: conduciveness; Urg: urgency; Ctr: control; Adj: adjustment; InSt: internal standards; ExSt: external standards.




response will only occur when the estimate is considered as relatively stabilized, justifying the investment of energy. In other words, while all checks may be processed in parallel on a very superficial level, deep processing only occurs when an appropriate register value is set in a prior check that requires more extensive investment of resources for further evaluation. As long as no change occurs (shallow processing), the register setting remains unchanged. Thus, the economy argument for the utility of sequential processing which I have advanced before has three bases: (a) costly deep processing is only initiated when required by previous results, (b) efferent reactions in other com-ponents are only elicited when there is sufficient justification, and (c) the efferent response pattern can be continuously adapted in terms of its specificity according to the cumulative-sequential information.

What is the empirical evidence for the prediction of sequential-cumulative appraisal effects? Scherer (1999) demonstrated that the recognition of emotions in scenarios that provide informa-tion corresponding to the SECs is faster and more accurate if the information is given in the predicted order. Although interest-ing, this provides only indirect evidence for the sequence assumption. More direct testing is required, if possible based on signatures for appraisal checks in the form of indicators in brain activity, physiological changes, or specific expression patterns. In recent years, a number of different studies have addressed these issues.

Reaction Time

Flykt, Dan, and Scherer (2009) reported an attempt to estimate the time window of the intrinsic pleasantness check using a dual-task probe paradigm. In three experiments, participants viewed negative and positive pictures. Their other task was speeded response on a probe superimposed on the pictures with different stimulus onset asynchronies (SOAs). Longer probe-reaction times were observed for negative than for positive pictures. This effect appeared at SOA 300 ms or 350 ms, suggesting that the intrinsic pleasantness appraisal check yields a differential behav-ioral outcome around 300 ms after stimulus onset, and seems to continue unless attention to picture content is inhibited.

Facial Expression Indicators

Lanctôt and Hess (2007) empirically tested the CPM hypothesis that the 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-Man type video game was used to manipulate goal conduciveness. Facial EMG was used to meas-ure facial reactions to each evaluation. As predicted, facial reac-tions 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. More recently, Krumhuber and Scherer (2011) studied affect bursts consisting of spontaneous, short emotional expressions in which

facial, vocal, and gestural components are highly synchronized. This study investigated the facial correlates of affect bursts that expressed five different emotions: anger, fear, sadness, joy, and relief, as enacted by professional actors. Detailed analysis of 59 facial actions with the Facial Action Coding System revealed a reasonable degree of emotion differentiation for individual action units. Interestingly, the expression of facial actions peaked in a cumulative-sequential fashion with significant dif-ferences in their sequential appearance between emotions. Although this study provided only indirect evidence for the sequential-cumulative hypothesis, it does highlight the potential of facial expression patterns to index underlying appraisals (as suggested by Scherer, 1992).

Physiological Indicators

Aue et al. (2007) tested the sequence assumption with a memory task in which participants were presented with pictures display-ing biological and cultural threat stimuli or neutral stimuli (stim-ulus relevance manipulation) with superimposed symbols signaling monetary gains or losses (goal conduciveness manipu-lation). Results for heart rate and facial EMG showed differential efferent effects of the respective appraisal outcomes and pro-vided direct evidence for sequential processing. As predicted, muscle activity over the brow and cheek regions marking the process of relevance appraisal occurred significantly earlier than that of goal conduciveness appraisal. Heart rate, in contrast, was influenced by the stimulus-relevance manipulation only.

Delplanque et al. (2008) tested whether an odor 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 odors (sample and target odors). Within a pair, the sample and target were either identical or different in order to assess participants’ novelty detection; unpleasant and pleasant target odors were contrasted to examine participants’ appraisal of intrinsic pleasantness. Dependent measures included facial expressions (using EMG) and physiological reactions (using electrocardiogram and elec-trodermal activity). 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 sequential effects of appraisal checks on odor-elicited emotional reactions.

Brain Activity

Grandjean and Scherer (2008) systematically manipulated nov-elty, goal relevance, intrinsic pleasantness, and goal conducive-ness SECs in visual stimuli to test the sequence hypothesis in two experiments with EEG recordings. Topographical analyses of the ERPs revealed a specific electrical map related to novelty (90 ms after the onset of the stimulus) preceding another topo-graphical map related to task goal 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 analyzed the global field power, 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 condu-civeness checks (see Grandjean & Scherer, 2008, Figure 4). The frequency analyses revealed late effects in the gamma band, indicating an effect of goal conduciveness 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, possibly on a conscious, controlled, and propositional level (as predicted by Leventhal & Scherer, 1987).

Van Peer et al. (2012) replicated and extended these results using a similar experimental design and EEG analysis proce-dure. Novelty and intrinsic pleasantness appraisals were manip-ulated by using a three-stimulus oddball paradigm with visual affect-inducing stimuli (IAPS pictures). Different measures of EEG brain activity were computed to identify markers for these specific appraisal results. In addition to replicating the predicted sequence, the data suggest that the effects of intrinsic pleasant-ness are modulated by novelty. This can be taken as an indica-tion that different neural mechanisms underlie the processing of the respective checks.

Gentsch et al. (2012) extended this approach by adding the coping potential check, predicted to occur after the goal condu-civeness check. In the context of a monetary gambling task, EEG recordings were used to measure brain activity accompa-nying the evaluation of feedback related to goal conduciveness (win vs. loss of money vs. break even) and coping potential (presence vs. absence of option to change the outcome). In each trial, information for both appraisal checks was simultaneously presented. Confirming the sequence hypothesis, feedback-locked ERPs showed early effects of the goal conduciveness (~230–300 ms), followed by a Coping Potential Goal Conduciveness interaction in the consecutive time interval (~350–600 ms).

In sum, these results show that signatures or markers of appraisal in peripheral components exist and can be exploited to test the sequence hypothesis generated by the CPM. Overall, there is now a rather impressive amount of empirical evidence for the predicted sequential processing of novelty, intrinsic pleasantness, goal conduciveness, and coping potential.

ConclusionI have reviewed recent theoretical ideas generated in our group, some of which are currently informing our experimental work, for example in the direction of testing the effects of different types of valence appraisal. I have also reviewed the current state of

evidence on the dynamic nature of the appraisal-driven emotion process, especially with respect to the sequential-cumulative hypothesis proposed in the CPM. Needless to say, I feel greatly encouraged by the accumulating support for the model, especially from neuroscience and psychophysiological research. Discussing the results of past research and designing new studies reveals that there is an urgent need to theoretically develop the role of interac-tions between appraisal checks and their effects on the cumulative sequence. For example, it is quite obvious (and we have unpub-lished empirical evidence) that the coping potential check plays a different role in the appraisal process depending on whether the event is goal conducive or obstructive. In the latter case, the ques-tion of whether I have enough power to deal with the conse-quences becomes a primary concern, probably leading to greater attention and more in-depth processing, and giving greater weight to the result of the check in the overall integration. Conversely, if an event is pleasant and/or if I have almost reached my goal, a coping potential check might seem much less important or even superfluous. Furthermore, particular SECs might be of greater importance for certain appraisal situations, requiring that they are weighted accordingly in the integration of the appraisal results. These questions need urgent attention as they have important implications for the computational modeling of emotion (Scherer, 1993, 2010; Scherer & Meuleman, 2012).

Answers to these questions require a much more detailed understanding of the exact mechanisms underlying appraisal, as well as the brain structures and circuits involved. Clearly, more theoretical work, as well as close contact with theory and research in cognitive psychology and the neurosciences, will be required to flesh out these preliminary ideas—future work that, given the complexity of both the theoretical framework and the methodological tools required, should preferably be taken on by interdisciplinary teams of researchers.

Note1 Of course, this result depends on the order in which the different compo-

nents are entered into the discriminant analysis because in the case of a high level of common variance, as is the case here, the first component entered will automatically explain a very high amount of the variance. However, as in regression analysis, the strongest component is always entered first, and this is indeed appraisal in our case, with over 70% of the variance explained on the basis of the component features alone (in comparison, action tendencies explain 67.7%, bodily reactions 62.4%, facial expression 62.7%, vocal expression 53.2%, and feeling 52.5%). Quite independently of the statistics, it seems difficult to imagine another sequence, given that it is hard to imagine other major sources of differentiation in the response components than appraisal.

ReferencesAue, T., Flykt, A., & Scherer, K. R. (2007). First evidence for differential

and sequential efferent effects of goal relevance and goal conducive-ness appraisal. Biological Psychology, 74, 347–357.

Aue, T., & Scherer, K. R. (2008). Appraisal-driven somatovisceral response patterning: Effects of intrinsic pleasantness and goal conduciveness. Biological Psychology, 79, 158–164.




Aue, T., & Scherer, K. R. (2011). Effects of intrinsic pleasantness and goal conduciveness appraisals on somatovisceral responding: Somewhat similar, but not identical. Biological Psychology, 86, 65–73.

Brosch, T., Sander, D., Pourtois, G., & Scherer, K. R. (2008). Beyond fear: Rapid spatial orienting towards emotional positive stimuli. Psychologi-cal Science, 14, 362–370.

Brosch, T., Sander, D., & Scherer, K. R. (2007). That baby caught my eye . . . Attention capture by infant faces. Emotion, 7, 685–689.

Clayton, S., & Myers, G. (2009). Conservation psychology. Hoboken, NJ: Wiley-Blackwell.

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, 224–231.

Delplanque, S., Grandjean, D., Chrea, C., Aymard, L., Cayeux, I., Le Calve, B., . . . Sander, D. (2008). Emotional processing of odors: Evidence for a nonlinear relation between pleasantness and familiarity evaluations. Chemical Senses, 33, 469–479.

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. 572–595). New York, NY: Oxford University Press.

Ellsworth, P. C., & Smith, C. A. (1988). Shades of joy: Patterns of appraisal differentiating pleasant emotions. Cognition & Emotion, 2, 301–331.

Flykt, A., Dan, E. S., & Scherer, K. R. (2009). Using a probe detection task to assess the timing of intrinsic pleasantness appraisals. Swiss Journal of Psychology, 68, 161–171.

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, 1050–1057.

Fontaine, J. R. J., Scherer, K. R., & Soriano, C. (in press). Components of emotional meaning: A sourcebook. Oxford, UK: Oxford University Press.

Frijda, N. H. (1986). The emotions. London, UK: Cambridge University Press.

Gentsch, K., Grandjean, D., & Scherer, K. R. (2012). Temporal dynam-ics of event-related potentials related to goal conduciveness and power appraisals. Manuscript submitted for publication.

Gignac, G. E. (2008). Higher-order models versus direct hierarchical models: g as superordinate or breadth factor? Psychology Science Quarterly, 50, 21–43.

Grandjean, D., & Scherer, K. R. (2008). Unpacking the cognitive architec-ture of emotion processes. Emotion, 8(3), 341–351.

Kaiser, S., & Wehrle, T. (2001). Facial expressions as indicators of appraisal processes. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotions: Theories, methods, research (pp. 285–300). New York, NY: Oxford University Press.

Kalisch, R., Wiech, K., Critchley, H. D., & Dolan, R. J. (2006). Levels of appraisal: A medial prefrontal role in cognitive appraisal of emotional material. Neuroimage, 30, 1458–1466.

Krumhuber, E. G., & Scherer, K. R. (2011). Affect bursts: Dynamic patterns of facial expression. Emotion, 11, 825–841.

Lanctôt, N., & Hess, U. (2007). The timing of appraisals. Emotion, 7, 207–212.

Leventhal, H., & Scherer, K. R. (1987). The relationship of emotion to cog-nition: A functional approach to a semantic controversy. Cognition & Emotion, 1, 3–28.

Lewin, K. (1951). Field theory in social science: Selected theoretical papers (D. Cartwright, Ed.). Westport, CT: Greenwood Press. (Original work published 1938)

Moors, A., Ellsworth, P., Scherer, K. R., & Frijda, N. H. (2013). Appraisal theories of emotion: State of the art and future development. Emotion Review, 5, 119–124.

Murphy, F. C., Hill, E. L., Ramponi, C., Calder, A. J., & Barnard, P. J. (2010). Paying attention to emotional images with impact. Emotion, 10, 605–614.

Öhman, A., & Mineka, S. (2001). Fear, phobias and preparedness: Toward an evolved module of fear and fear learning. Psychological Review, 108, 483–522.

Pecchinenda, A. (2001). The psychophysiology of appraisals. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emo-tion: Theory, methods, research (pp. 301–315). New York, NY: Oxford University Press.

Roseman, I. J., & Evdokas, A. (2004). Appraisals cause experienced emotions: Experimental evidence. Cognition & Emotion, 18, 1–28.

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 (pp. 3–34). New York, NY: Oxford University Press.

Sacharin, V., Sander, D., & Scherer, K. R. (2012). Levels of valence. Manuscript submitted for publication.

Scherer, K. R. (1982). Emotion as a process: Function, origin, and regulation. Social Science Information, 21, 555–570.

Scherer, K. R. (1984). On the nature and function of emotion: A component process approach. In K. R. Scherer & P. Ekman (Eds.), Approaches to emotion (pp. 293–317). Hillsdale, NJ: Erlbaum.

Scherer, K. R. (1992). What does facial expression express? In K. Strongman (Ed.), International review of studies on emotion (Vol. 2, pp. 139–165). Chichester, UK: Wiley.

Scherer, K. R. (1993). Studying the emotion-antecedent appraisal process: An expert system approach. Cognition & Emotion, 7, 325–355.

Scherer, K. R. (1999). On the sequential nature of appraisal processes: Indirect evidence from a recognition task. Cognition & Emotion, 13, 763–793.

Scherer, K. R. (2000). Emotions as episodes of subsystem synchronization driven by nonlinear appraisal processes. In M. D. Lewis & I. Granic (Eds.), Emotion, development, and self-organization: Dynamic systems approaches to emotional development (pp. 70–99). New York, NY: Cambridge University Press.

Scherer, K. R. (2001). Appraisal considered as a process of multi-level sequential checking. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.), Appraisal processes in emotion: Theory, methods, research (pp. 92–120). New York, NY: Oxford University Press.

Scherer, K. R. (2005). What are emotions? And how can they be measured? Social Science Information, 44, 693–727.

Scherer, K. R. (2009a). The dynamic architecture of emotion: Evi-dence for the component process model. Cognition & Emotion, 23, 1307–1351.

Scherer, K. R. (2009b). Emotions are emergent processes: They require a dynamic computational architecture. Philosophical Transactions of the Royal Society, Series B, 364, 3459–3474.

Scherer, K. R. (2010). The component process model: A blueprint for a comprehensive computational model of emotion. In K. R. Scherer, T. Bänziger, & E. B. Roesch (Eds.), Blueprint for affective computing: A sourcebook (pp. 47–70). Oxford, UK: Oxford University Press.

Scherer, K. R., & Brosch, T. (2009). Culture-specific appraisal biases con-tribute to emotion dispositions. European Journal of Personality, 23, 265–288.

Scherer, K. R., Dan, E., & Flykt, A. (2006). What determines a feeling’s position in three-dimensional affect space? A case for appraisal. Cognition & Emotion, 20, 92–113.

Scherer, K. R., & Fontaine, J. R. J. (in press a). Driving emotion: The appraisal component. In J. R. J. Fontaine, K. R. Scherer, & C. Soriano (Eds.), Components of emotional meaning: A sourcebook. Oxford, UK: Oxford University Press.

Scherer, K. R., & Fontaine, J. R. J. (in press b). Componential mean-ing structure of emotion terms: Integration across components. In J. R. J. Fontaine, K. R. Scherer, & C. Soriano (Eds.), Components of emotional meaning: A sourcebook. Oxford, UK: Oxford University Press.




Scherer, K. R., & Meuleman, B. (2012). Human emotion experiences can be predicted on theoretical grounds: Evidence for verbal labeling. Manuscript submitted for publication.

Scherer, K. R., Schorr, A., & Johnstone, T. (Eds.). (2001). Appraisal pro-cesses in emotion: Theory, methods, research. New York, NY: Oxford University Press.

Scherer, K. R., Zentner, M. R., & Stern, D. (2004). Beyond surprise: The puzzle of infants’ expressive reactions to expectancy violation. Emotion, 4, 389–402.

Smith, C. A. (1989). Dimensions of appraisal and physiological response in emotion. Journal of Personality and Social Psychology, 56, 339–353.

van Peer, J. M., Grandjean, D., & Scherer, K. R. (2012). Sequential unfold-ing of novelty and pleasantness appraisals of affective pictures: Evidence from EEG. Manuscript submitted for publication.

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 & Emotion, 18, 663–688.



Emotion Review 2013 Scherer 150 62

Documents