Process-oriented appraisal theories
Process-oriented appraisal theories maintain that dynamic, parallel, non-linear and holistic emotional processing is both elicited and driven by appraisals. Appraisal "is a process that detects and assesses the significance of the environment for well-being," as leading contemporary appraisal theorists Agnes Moors, Phoebe Ellsworth, Klaus Scherer, and Nico Frijda (2013, 120) summarize. Different appraisal theories have somewhat dissimilar conceptions about the components of the appraisal process but commonly suggested dimensions include novelty, intrinsic pleasantness, certainty or predictability, goal significance, agency, coping potential, and compatibility with social or personal standards (Ellsworth & Scherer, 2003). Some appraisal theories argue that the appraisal process follows a certain fixed logical order so that, for instance, the relevance and implications of a stimulus must be evaluated before proceeding to the dimensions of coping ability and normative significance. The sequential view of emotional appraisal can be reconciled with parallel and nonlinear processing by assuming that each "stimulus evaluation check" (Scherer's term for appraisal dimension) has a particular point in time at which it achieves a preliminary closure. The closure yields a result that warrants efferent commands to various response modalities - ANS physiology, action tendencies, motor expression, and feeling state - and allows the evaluation to proceed to the next phase even if the previous stages remain open to revision according to the results of subsequent appraisals. Other appraisal theories deny the sequential assumption as overly restrictive, suggesting that subordinate appraisals need not unfold in any particular order, especially in recurring appraisal cycles (e.g. Lazarus, 1991; Smith & Kirby, 2001). However, both versions of appraisal theories emphasize that emotional appraisal may occur at different levels of processing; fast, automatic, implicit, and schematic on the one hand, and slow, controlled, explicit, and conceptual on the other hand. "Appraisal may be a single process at the functional level of analysis, but it covers multiple processes (i.e., mechanisms) at the algorithmic level of analysis," as Moors (2012, 266) points out. A cognitive view about the nature of ordinary adult human emotions at the algorithmic level of analysis can be maintained if all emotions involve the latter kind of conceptual processing at some stage of the appraisal process even if the process may have been triggered by appraisals of the former, nonconceptual type.
What reasons are there to believe that cognitive, conceptual appraisals regularly rather than just contingently participate in the formation of human emotional responses in situations where the computation of personal meaning occurs online rather than by association to an existing appraisal pattern? The first reason focuses on the nature of certain appraisal dimensions. Appraisal theorists maintain that the early appraisals in the emotion generation process on such themes as novelty and intrinsic pleasantness may operate on a sensory basis without cognitive processing. Accordingly, those appraisals are also present in most animals, including newborn humans. These appraisals launch the initial physiological, behavioural and affective consequences whose feedback influences later stages of appraisal at which the level of processing is expected to move upwards. This is particularly the case with appraisals on the self and the normative significance of the emotion-eliciting event. These appraisals focus on such issues as coping ability, credit or blame, and the compatibility of the response with social and personal norms (Scherer, 2001, 2009, 2013). Scherer argues that "the normative significance of the event... is expected to be appraised last, as it requires comprehensive information about the event and comparison with high-level propositional representation" (Scherer, 2009, p. 1318). In this way, noncognitively elicited emotions turn into cognitive through the noncontingent involvement of cognitive processing at later stages of evaluation that bring the evaluation process into a closure.
The second reason for the involvement of cognitive appraisals in emotional processing emerges from the recursive character of emotional appraisal. Appraisal is not a one-shot affair that gives way to other aspects of emotional response after having initiated them but a process that evolves during the entire emotional response, updating and modifying the various response elements on the basis of both internal and external feedback. This means that initiating appraisals on the various stimulus evaluation themes give way to reappraisals of the same themes during the dynamically unfolding emotion process. Scherer maintains that the recursive "checking process repeats the sequence continuously, constantly updating the appraisal results that change rapidly with changing events and evaluation... until the monitoring subsystem signals termination of or adjustment to the stimulation that originally elicited the appraisal episode" (Scherer, 2009, p. 1317). Thus if the level of processing has not risen from schematic to conceptual towards the end of the first round of appraisals, it will do so during the recursive appraisal cycles. In this way, conceptual appraisals join the appraisal process during recursive reappraisals in determining the type and shape of the evolving emotional response, including its experienced evaluative content.
Appraisal theories have often been contested on methodological grounds. Even if there is conclusive evidence that people associate distinct emotions with certain appraisal themes in self-report studies, this evidence alone does not support an empirical connection between emotions and appraisals but only a conceptual one (e.g. Parkinson, 1997, 2001; Barrett, 2006a,b). In order to support the empirical claim, process-oriented appraisal theories have sought to find correlates of various dimensional appraisals in diverse domains of emotional responding, such as vocal and facial expression, ANS response, and action tendencies. Scherer (2001; 2009) has presented detailed predictions for specific effects of appraisal outcomes in all response modalities, some of which have been supported in empirical research. Thus, Aue and Scherer (2008, 2011) found out that the appraisals of intrinsic pleasantness and goal conduciveness had somewhat similar but still not identical somatovisceral response patterns, with similarities in facial activity and differences in mean skin conductance, forehead temperature, and finger temperature. Lanctot and Hess (2007) obtained similar results about the difference and sequential order of intrinsic pleasantness and goal conduciveness in facial EMG. Likewise, Kaiser and Wehrle (2001) detected stages of appraisal from filmed sequences of dynamically unfolding facial expressions. Finally, facial muscle activity over the brow and cheek regions marking the process of stimulus relevance appraisal occurred significantly earlier than that of goal conduciveness appraisal in the study of Aue et al. (2007). Some correlations between physiological responses and appraisals have also been established. Differences in appraised task difficulty and coping potential have been associated with dissimilar cardiovascular reactivity and electrodermal activity (Pecchinenda, 2001). However, these are only a small fraction of the hypothesized correlations between distinct appraisals and the various central nervous system, neuroendochrine system, autonomic nervous system, and somatic nervous system changes, which means that the empirical validation of the theory is still at an early stage. Nevertheless, some work has been done on finding the neurophysiological correlates of distinct appraisals as this issue bears heavily on the default-interventionist claims of the theory.
Tobias Brosch and David Sander (2013) propose a model of neural mechanisms that underlie the processing of distinct appraisal criteria on the basis of existing research in affective neuroscience. They suggest that the novelty of stimuli is detected subcortically and automatically in the perirhinal cortex, the hippocampus, and the amygdala. The amygdala also detects concern relevance, but whether or not automatically is a debated issue. Goal congruence is processed in a "conflict-control loop consisting of anterior cingulate cortex (ACC) and dorsolateral prefrontal cortex (DLPFC). This circuit subserves the monitoring of performance toward a goal, the detection of goal conflicts, and the adjustment of top-down cognitive control to resolve potential conflicts" (165). Here emotional processing enters the cortical sphere in being related to individual differences in higher-order belief systems. Even so, the detection of goal congruence may occur automatically. Different neural regions underlie attributions of internal and external agency "with temporoparietal junction (TPJ), precuneus, dorsomedial prefrontal cortex (DMPFC), presupplementary motor area involved in attributing external agency, and insulate and motor-specific regions are involved in attributing self-agency" (ibid.). Finally, an appraisal of norm or value compatibility is related to activation in the superior anterior temporal lobe for norms and in the medial prefrontal cortex and the dorsal striatum for values. With the last two appraisal themes, both automatic and controlled processes are possible. Brosch and Sander argue, with reference to Pessoa and Adolphs (2008), that emotional processing proceeds in the form of "multiple sweeps of activation with numerous feedforward and feedback loops that refine neural processing patterns and underlying computations with each iteration" (166). When cortical refinement of the amygdala-induced response reaches a stable state, the ventromedial prefrontal cortex (VMPFC) represents a conceptualization of the appraisal outcome. From the perspective of different levels of analysis, this evidence shows that "appraisal may or may not be a single process at the functional level, but it does not correspond to a single process at the algorithmic level, nor to a single process at the implementational level" (Moors, 2012, 267).
-  Brian Parkinson and Antony Manstead (1992) argue that it is possible to detect normative significance effortlessly on the basis of implicit interaction and feeling rules that associate with interpersonal roles and status differences. Accordingly, these authors contrast "contextual meaning apprehension" with cognitive appraisal. However, the dichotomy is exaggerated since adequate responsiveness to social interaction and feeling rules is nevertheless mediated by the processing of meaning, which qualifies as cognitive activity. I shall return to the argument on the contrast between implicit emotion regulation and cognitive appraisals below in the section on emotion regulation.