Comment: Appraisal Affords Flexibility to Emotion in More Ways Than One

The contributions to this special section provide a valuable and needed update of contemporary thinking about the role of appraisal in emotion, including several avenues for future research to further position appraisal in its different forms and processing levels at the heart of what elicits and constitutes emotion. In this commentary, I will focus on an issue that is central to the role of appraisal in emotion, namely on the ways in which it affords flexibility to emotional responding, and propose to extend appraisal accounts of emotions to incorporate flexibility not only in how people appraise events, but also in how appraisals lead to changes in other emotion components.

As stated in Moors, Ellsworth, Scherer, and Frijda (2013), the central tenet of appraisal theories of emotions that also distinguishes them from other emotion accounts is that “appraisal triggers and differentiates emotional episodes through synchronic changes in other components” (p. 120). Appraisal is not situated in the external world, but reflects a process in the individual of detecting and assessing the significance of this external world for the person’s well-being (Ellsworth, 2013; Moors, 2013; Moors et al., 2013; Roseman, 2013; Scherer, 2013). As such, appraisal is considered to act as the interface between the person and the situation or environment, affording flexibility to emotional responding by decoupling stimulus and response, inserting the appraisal process in between (see also, e.g., Scherer, 2000).

According to the views expressed in this special section, the flexibility afforded by appraisal is considered to exclusively reside in the appraisal process, or in the ways people appraise events as a function of the combination of the situation and of their own idiographic goals and concerns, and cultural and learning history. How appraisal is next tied to or even causally drives changes in the other components is considered to be relatively fixed. As stated in Moors et al. (2013), “Appraisal theories assume … a stable relationship between appraisals and emotions. In general, the same appraisals lead to the same emotion; different appraisals lead to different emotions” (p. 121).

To me, and I would argue according to a large amount of empirical evidence, this conception of flexibility is largely and literally one-sided. It assumes flexibility in how appraisal outcomes come to be, but assumes a relatively rigid response system in terms of how appraisal output next feeds into the other components. Nowhere in the accounts (with perhaps the exception of Roseman, 2013) is the possibility of flexibility taken into account in the ways that appraisals are associated with or cause changes in the other components or, more generally, in the ways in which emotion components interrelate. Rather, all accounts mention the occurrence of synchronicity in the components subsumed under emotions during an emotional episode, suggesting strong and characteristic interrelations between components following appraisal theoretical predictions. In my view, appraisal theories should be extended to incorporate flexibility not only in how appraisal output or results come to be, but also in how they relate to other components, and by extension, how components more generally interrelate as parts of emotions as continuous dynamic processes.

The versions of appraisal theory defended in this special section posit themselves against affect program theories (although Roseman, 2013, takes a more nuanced position). The opposition is fully justified in the sense of not positing a limited number of affect programs, but instead defending the existence of an infinite combination of appraisal outcomes, and resulting componential makeups of emotional states (Ellsworth, 2013; Scherer, 2013). But the versions of appraisal theory defended in this special section do not seem to contradict affect program theories in the sense that appraisal outputs lead to invariable consequences in other emotion components. It is assumed that given a certain configuration of appraisal results, a fixed pattern of other components ensues. In fact, several new research lines are being described in search of strong correlates of appraisals in terms of action tendencies, behavioral patterns, or physiological components. I think this search for such strong associations is misguided and underestimates the true flexibility that appraisal as the interface between stimulus and response affords us.

Indeed, the available empirical evidence does not seem to support the existence of strong associations between appraisal outputs and other components of emotions. Rather, the evidence seems to be in favor of flexible and dynamical componential consequences of appraisal outputs (see also Roseman, 2013, for an overview of factors playing a role in this respect).

In terms of appraisals and reported emotional experience (the feeling or experiential component of emotions), there is ample empirical evidence showing that the experience reported in association with particular appraisal outcomes or profiles of such outcomes can differ substantially across individuals and contexts (Ceulemans, Kuppens, & van Mechelen, 2012; Kuppens, van Mechelen, & Rijmen, 2008; Kuppens, van Mechelen, Smits, De Boeck, & Ceulemans, 2007; Nezlek, Vansteelandt, van Mechelen, & Kuppens, 2008; Silvia, Henson, & Templin, 2009; Tong, 2010a, 2010b; Tong & Tay, 2011; van Mechelen & Hennes, 2009). In other words, the same (set of) appraisal outcomes do not lead to the same reported emotional experience in all contexts or individuals. For instance, evidence shows that while one person’s anger may be highly contingent on the frustrating nature of an event, another person’s anger is more strongly associated with the appraisal of blaming someone for what has happened (Kuppens et al., 2008). As reviewed in Kuppens and Tong (2010), such individual differences have been acknowledged and described extensively outside of the appraisal literature as well, dating back to assumptions of early theories of personality (Eysenck, 1967; Gray, 1981; Strelau, 1987) to contemporary accounts of personality and emotion (Gross, Sutton, & Ketelaar, 1998; Meier & Robinson, 2004; Meier, Robinson, & Wilkowski, 2006; Robinson, 2007; Robinson, Ode, Moeller, & Goetz, 2007). In sum, the relationships between appraisals and emotional experience are not stable, and the same appraisals do not always lead to the same emotions. Depending on a person’s disposition, learning history, culture, etcetera, appraisals can differentially impact a person’s emotional experience. Yet most prominent appraisal theorists do not explicitly take this into account when detailing their view on the role of appraisal in emotion. For instance, it clearly runs against the statements in Moors et al. (2013) cited before and against the (admittedly cautiously worded) claim that all influence on emotion is through appraisal (Ellsworth, 2013, p. 126).

Also for behavioral and physiological components, variability and flexibility in their relationship with appraised meaning is the rule rather than the exception (Mauss & Robinson, 2009). Clearly, people do not always display the same corresponding emotional behavior when faced with a situation that is appraised in a certain way (e.g., Gross, John, & Richards, 2000; Reisenzein, Bordgen, Holtbernd, & Matz, 2006). This may be because of social constraints (e.g., display rules; Ekman & Friesen, 1975; Matsumoto, 1990), but can also result from individual or contextual variation in associations between appraisal outcomes and motivational tendencies. Different factors come into play to determine, for instance, whether a human or animal will fight, flee, freeze, tend, or befriend in the presence of threat (Taylor et al., 2000) or will fight, flee, socially share, reconcile, assert, or restrain oneself or itself when faced with an anger-eliciting situation (Kuppens, van Mechelen, & Meulders, 2004). Likewise, until now researchers have yet to identify strong correspondence between appraised meaning and physiological response patterns (Cacioppo, Berntson, Larsen, Poehlmann, & Ito, 2000; Mauss & Robinson, 2009). Instead, there seem to be large individual differences and contextual moderation in the physiological changes associated with appraised meaning (Lodewyck, Tuerlinckx, Kuppens, Allen, & Sheeber, 2012; Stemmler & Wacker, 2010), for a large part because physiology is also determined by many different factors outside of the appraisal or emotion realm.

In terms of synchronicity between emotion components, this notion has large theoretical appeal. Yet empirically, the prevailing evidence points to low rather than high coherence among different emotion components (Mauss & Robinson, 2009; Russell, 2003). While there may be modal or averaged typical componential patterns associated with certain emotional states across individuals, possibly as a consequence of the existence of attractor basins or “magnets” in the emotion system as argued by, for instance, Ellsworth (2013) and Scherer (2013), clearly this does not imply that associations between emotion components always and for everyone occur in a fixed manner. The search for synchronicity or response patterning during emotional episodes is a very difficult endeavor, however (it involves complex and possibly nonlinear interrelations between different components over different time scales). I do not want to preclude that with advances in data collection and mathematical modeling approaches researchers will be able to more successfully identify instances of synchronicity and its boundary conditions. Yet, as far as we now know, it does not seem to be that appraisal outputs lead to invariable downstream consequences in physiology, behavior, and feeling.

In sum, the complete componential process underlying emotions, involving both the ways in which appraisals are shaped by the interaction between the person and situation and the ways in which appraisals drive changes in other components, is characterized by flexibility. Such a view strongly resonates with contextualized approaches to personality (e.g., Mischel & Shoda, 1998) that describe personality as a set of “IF-THEN” relationships of the type of “IF certain feature of situation, THEN certain behavioral response.” Similar to the present context, such relationships are considered to be shaped by cognitive-affective processes mediating between objective events and behavioral output, with flexibility characterizing the accessibility and interrelations of these units (Kuppens, 2009).

Both from a theoretical and applied perspective, it makes sense to also assume flexibility in the interrelations between appraisals and emotion components or between components in general. It allows for culture, personal history, etcetera, to etch itself into the emotional componential dynamics of the individual, not only in how people appraise events, but also in how they further respond to appraisals. In other words, it allows flexibility in the ways in which people emotionally and behaviorally respond to frustration, opportunity, (un)fairness, risk, etcetera, as a function of their disposition and considerations of the moment. Such flexibility also helps us to understand the emotional dysfunctioning observed in certain forms of psychopathology and mood disorder, where relatively innocuous appraisals can produce disruptive emotional responses, or vice versa when otherwise meaningful appraisals are met with emotional indifference. These are topics that are widely studied in domains outside of emotion research and have direct applications to real-world problems.

The future task first of all lies in explicitly building this flexibility into theoretical accounts of appraisal by abandoning the notion that appraisal outcomes are invariably linked to other emotion components, and by explicitly inscribing flexibility in appraisal outcomes in theoretical models. Empirically, the challenge remains for emotion psychology to study the boundary conditions and exact moderators of how and when components interrelate, and infuse more applied domains with its results. Only in this way can appraisal theory realize its potential, namely to account for the large variability in emotional life.