NOTE-429: The Emerging Field of Emotion Regulation: An Integrative Review

Contribution

The paper makes emotion regulation a field with one definition and one map. Before it, the same phenomena were spread across defense, coping, mood repair, self-control and developmental “self-soothing”. It (a) fixes a definition, (b) separates regulation from its neighbours, and (c) gives a process model that classifies regulatory acts by the stage of emotion generation they act on. It then surveys what seven subfields of psychology contribute, and closes with three open problems.

Key insight

If emotion is a sequence (situation, attention, meaning, response tendencies, response), then regulation can be classified by which link of the sequence it intervenes on, and interventions early in the sequence should behave differently from those late in it. That is the whole model: “emotion regulatory acts may be seen as having their primary impact at different points in the emotion generative process” (p. 281). The antecedent/response split is the version of it that experiments had tested.

Assumptions

• Emotion as response tendencies (pp. 272–273), after James: “adaptive behavioral and physiological response tendencies that are called forth directly by evolutionarily significant situations”, which “may be modulated”. Figure 1 gives the “consensual process model”: emotional cues, then evaluation, then response tendencies (behavioural, experiential, physiological), then modulation, then responses. Footnote 1 calls it “a distillation of major points of convergence” across appraisal and basic-emotion theorists, Lazarus (1991a) and Frijda’s citations among them.
• Emotion versus mood: emotions are shorter, have objects, and carry specific behavioural tendencies; moods are diffuse and “bias cognition more than they bias action” (p. 273). The paper regulates emotion, not mood.
• The nervous system as “multiple, partially independent information processing subsystems” with bidirectional limbic–cortical links (p. 275). This is what allows regulation to act on one response channel and not another.
• Regulation in self only (p. 275). Influencing others’ emotions is excluded, reversing Gross’s earlier usage.

Key results

The paper reports no data. Its results are definitions, distinctions and a classification, and it cites the empirical findings below.

• Definition (p. 275). “Emotion regulation refers to the processes by which individuals influence which emotions they have, when they have them, and how they experience and express these emotions.” Five glosses follow: it covers increase, maintenance and decrease of negative and positive emotion; regulation may differ by emotion; it is regulation of self; it runs on a continuum from conscious and effortful to unconscious and automatic; and “I make no a priori assumptions as to whether emotion regulation is good or bad”.
• Neighbouring constructs (p. 276, Figure 2). Affect regulation is the superordinate term, over coping, emotion regulation, mood regulation and defense. Coping differs in its focus on decreasing negative experience and in taking the whole episode as its unit. Defenses are usually unconscious, aimed at anxiety, and studied as stable individual differences.
• The five families (pp. 282–285, Figure 4):
  • situation selection: approaching or avoiding people, places or objects. Avoidance can trade short-term relief for long-term cost, as when a shy person avoids social situations “at the cost of longer term social isolation” (p. 283);
  • situation modification: problem-focused coping in Lazarus and Folkman’s sense, where the boundary with selection is unclear;
  • attentional deployment: distraction, concentration, and rumination. Rumination on depressive affect “leads to longer and more severe depressive symptoms”, and worry to “long-lasting anxiety” (p. 284);
  • cognitive change: choosing among a situation’s possible meanings, reappraisal included. Reappraisal of negative films lowered negative experience. For disgust films physiology did not always fall, which the paper explains by how little cognitive processing such images need (p. 285);
  • response modulation: acting directly on physiology, experience or expression, through drugs, exercise, relaxation, alcohol or food, and suppression of expression. Suppression lowered reported experience of some emotions (pride, amusement) but not others (disgust, sadness), and raised sympathetic responding (p. 285). The first four are antecedent-focused and the last response-focused (p. 282).
• Seven stated limits of the model (pp. 282–283): situations, aspects and meanings resist definition; modulation is richer than up and down; control may be asymmetric across channels; “situations” include internal ones; responses feed back into situations, giving iterated S-O-R cycles; regulation runs in parallel at several points; and the model is silent on mechanisms, relations among strategies, relative efficacy, and the interdependence of channels.
• Clinical reach (p. 280). “Emotion dysregulation is implicated in over half of the DSM-IV Axis I disorders and in all of the Axis II disorders” (citing APA 1994, Gross & Levenson 1997, Thoits 1985). Therapies are sorted by the same families: changing the situation (Lewinsohn et al.), its construal (Ellis), or the response (Deffenbacher, Novaco).
• Three open challenges (pp. 285–287):
  • whether regulation can be distinguished from generation. “Some theorists argue that emotion regulation is best seen as part and parcel of emotion (Frijda, 1986)”. Inferring regulation needs a counterfactual unregulated response, so “thresholds for inferring emotion regulation should be high”;
  • what people’s regulatory goals are. They are not simply hedonic, since people also up-regulate negative emotion, as bill collectors do with anger;
  • how emotion regulation relates to other self-regulation, such as delay of gratification and appetite.

Claims

id	claim	strength	support
C1	Emotion regulation is distinct from coping, mood regulation and defense, all under affect regulation	definitional	pp. 274–276, Figure 2
C2	Regulatory acts can be classified by the stage of emotion generation they act on, in five families	conceptual; offered as “working distinctions”	pp. 281–283, Figure 4
C3	Antecedent-focused and response-focused regulation have different consequences: reappraisal lowers negative experience, suppression of expression does not reliably lower it and raises sympathetic activation	moderate, from the cited experiments (not reported here)	p. 285; Gross 1998, Gross & Levenson 1993, 1997
C4	Rumination prolongs depressive symptoms and worry prolongs anxiety	moderate, cited	p. 284; Nolen-Hoeksema 1993, Borkovec et al. 1995
C5	Emotion dysregulation figures in over half of DSM-IV Axis I disorders and in all Axis II disorders	weak as stated: asserted with three citations and no count	p. 280
C6	Regulation should be inferred only against a well-established unregulated baseline, since otherwise it explains any result	argued	p. 286
C7	Well-being is most likely when antecedents are regulated toward pursuits of enduring value, emotions are attended to in a differentiated way, and modulation is flexible	speculative (“This perspective suggests”)	pp. 287–288

Concepts

• emotion: a short-lived, multicomponent response tendency (behavioural, experiential, autonomic, neuroendocrine), called forth by an evaluated situation and open to modulation (pp. 272–273).
• affect: the superordinate category for valenced states: emotions, emotion episodes, moods, dispositions and traits, following Scherer (p. 273).
• emotion episode: the extended scene, with its protagonists, instigator and sequence of responses (p. 273).
• antecedent-focused / response-focused regulation: before versus after response tendencies are generated (p. 282).
• reappraisal: “cognitively transforming the situation so as to alter its emotional impact” (p. 284).
• emotion dynamics: the latency, rise time, magnitude, duration and offset of responses, which regulation changes (p. 275, after Thompson).

Connections

• Lazarus, Emotion and Adaptation (LIT-550), Deferred. Cited as “Lazarus, 1991a”: the “adaptational encounters” of the emotion episode, the object-directedness of emotion, and a source of Figure 1’s consensual model. Problem-focused versus emotion-focused coping, and the coping definition quoted on p. 274, come from Lazarus and Folkman (1984). The paper also cites Lazarus (1991b), the American Psychologist summary, for emotions as the “wisdom of the ages” (p. 287).
• Frijda, The Emotions (LIT-546), Deferred. Cited for emotions preparing rapid motor response, for the “third approach” of conceptual analysis of regulatory processes (p. 281), for cognitive change as modification of appraisal steps (p. 284), and as the representative of the view that regulation is “part and parcel of emotion” (p. 286). Frijda (1988), “The laws of emotion”, is cited for flexible modulation (p. 288).
• Frijda, Ridderinkhof & Rietveld (LIT-540), read. Sixteen years later they take the side of Gross’s first open challenge that Gross did not. On their account an emotion’s own action readiness, triggered by appraising the consequences of acting on another, does much of the regulating, so “there is no clear distinction between processes of emotion and processes of emotion regulation”. That is the position Gross attributes to Frijda (1986) and sets aside.
• Barrett (LIT-542), read. Gross thanks Lisa Feldman Barrett for comments and cites her “in press” work on valence and arousal focus and on whether verbal labelling of emotion is needed for some regulation (p. 277). Her 2017 account rejects the stimulus→response framing that Figure 1 encodes. She keeps reappraisal as a phenomenon and recasts it as modification of predictions (her Figure 4C).
• Damasio’s somatic markers (LIT-384). Gross cites Damasio (1994), the book, for the prefrontal cortex suffusing “cortical information processing with emotional meaning” (p. 276), and for emotions as information about goal progress (p. 287). Same research programme as LIT-384, different document.

Bearing on the record

• The reading supports a THEORY the record does not yet hold, proposed to the owner in this batch’s report: how an emotion is regulated matters as well as whether, because antecedent-focused strategies such as reappraisal change experience at lower physiological cost than suppression. This paper states the distinction and cites the evidence. A THEORY should be sourced from the experiments and the later meta-analyses, not from this review alone, so it is Proposed at most on this source.
• It is a natural anchor for the batch’s psychopathology-and-treatment works on transdiagnostic processes. Rumination, worry, avoidance and suppression all appear here as regulation strategies with costs.
• No instruction for machine-learning practice. Nothing here belongs in the anthology.
• Its first open challenge is taken up by THEORY-056, which takes the side Gross set aside.

Limitations

• No data. Every empirical claim is cited, and the experimental base for the antecedent/response contrast in 1998 was a handful of laboratory studies, mostly with films.
• The definition of emotion is assumed, not defended. The model needs emotions with a “true”, unregulated course against which regulation can be measured. The paper itself admits that course is hard to know (p. 286).
• The families overlap. The paper concedes permeable boundaries at several points: modifying a situation can create a new one, and distraction can amount to selecting an internal situation.
• The clinical figure (C5) is unsourced as a count. “Over half” of Axis I disorders is not derived in the paper.

Open questions

• Can regulation be measured independently of generation? The paper’s own test is a reliable unregulated baseline. Constructionist and action-readiness accounts both deny that the baseline exists in the needed sense.
• Do the families differ in mechanism, or only in timing? The model is “silent” on mechanism (p. 283).
• What are people’s regulatory goals, and are they hedonic?

Corrections

• The PDF’s text layer reads the masthead as “Vol. 2, No. 5”. Crossref and the printed citation give No. 3. That is likely a text-layer misread; it was not checked against the page image.
• Author’s lab page. The Stanford Psychophysiology Laboratory’s publications page (spl.stanford.edu/papers, 1998 section) links this article to DOI 10.1207/s15327965pli0904_8, which is Gross’s Psychological Inquiry commentary, and links that commentary to this article’s DOI 10.1037/1089-2680.2.3.271. The two DOIs are swapped there. The DOI on the LIT is Crossref’s.