The argument for cognitivism from emotion regulation
Emotion theories have traditionally regarded emotion regulation as a temporally and ontologically separate process that either precedes or follows emotion generation. In the former case, regulation is about controlling the occurrence of emotions by deliberately seeking out or avoiding potentially emotion-evoking situations, whereas the latter type of regulation focuses on modulating the duration, intensity, and expression of already generated emotions by suppression, savoring or other means of response modification. An envy-prone individual who morally disapproves of this emotion for instance may avoid situations involving social comparison in order to ward off experiences of envy and purport to suppress its displays whenever he feels this emotion. In both cases, regulation is extrinsic to the process of emotion generation that determines the identity and constituents of the emerging emotion. Extrinsic regulation that precedes or follows emotion generation involves conscious planning and control but insofar as regulation is not capable of being an intrinsic part of emotion generation, the two processes appear to be distinct and only contingently associated with each other in some emotional responses of adult humans. Indeed, most theorists of emotion, cognitive and noncognitive alike, with the exception of Frijda (1986) and Parrott (2007), have until recently shared this assumption about the contingent, limited, and extrinsic role of regulation in human emotions.
In contrast, the theories I have discussed above, both process-oriented appraisal theories and psychological constructionism, emphasize the constitutive role of regulation in human emotions. Appraisal theories incorporate reappraisals into the process of emotional evaluation, whereas constructionists see regulation as an ongoing process that accompanies and influences the other ingredients of emotion. Both theories regard regulation as an intrinsic aspect of emotions and their generation. The connection between emotion generation and regulation emerges through a process of meaning-making that continues from the beginning of an emotion to its end. "The same processes may be considered to be generative when they occur at the beginning of a new emotional episode... but considered to be regulatory when they occur later in the episode", as James Gross and Lisa Feldman Barrett (2011, p. 14) remark in their useful review on the relation of emotion generative and regulative processes in different types of emotion theories:
In appraisal views, the emphasis is on making sense of one's external surroundings, and internal changes are assumed to result from this meaning analysis in a way that reflects it. In psychological construction views, the emphasis is on making meaning of internal body sensations, and this meaning then makes it possible to construct a unified awareness of both body and the world. This similarity in emphasis on meaning-making (albeit with a different focus) means that one key target of emotion regulation for both perspectives will be the meaning-making process (ibid.)
However, it is important to note that an essential involvement of regulation in emotions does not support a robust role of cognition in emotion without a few further conditions. The intrinsic regulation of emotions must be cognitive itself and it must be always present in ordinary post-infantile human emotions in one form or another. The process of meaning-making that continues during the entire emotion is obviously cognitive at the functional level of analysis, but we are here interested in its status at the algorithmic level of analysis. Emotion regulation comes in many forms some of which are implicit, instinctive, nonconscious, and effortless rather than explicit, learned, conscious, and deliberate, and the former kind of regulation does not seem to involve cognitive representations or processes. If this is the case, then cognition is only contingently involved in emotion regulation, quite similarly to its role in emotion elicitation. The challenge is then to determine whether or not intrinsic yet implicit, nonconscious, and effortless emotion regulation may qualify as cognitive. The first task, however, is to clarify the distinction between intrinsic and extrinsic emotion regulation.