James J. Gross (1998), Review of General Psychology 2(3):271–299 — DOI-10.1037/1089-2680.2.3.271
Key takeaways
- Emotion regulation is defined by its target, not its means: any process, automatic or controlled, conscious or not, by which a person changes which emotions they have, when, and how they are felt and shown. It covers increasing as well as decreasing, and positive as well as negative emotions. The definition makes no assumption that regulation is good or bad.
- The process model sorts regulation by where in emotion generation it acts: choosing the situation, changing it, choosing what to attend to in it, changing what it means, and modulating the response once it has started. The coarse cut, antecedent-focused versus response-focused, is the one the paper’s own cited experiments test: reappraisal lowered negative experience, while suppressing expression left disgust and sadness experience intact and raised sympathetic activation (pp. 284–285, citing Gross 1998 and Gross & Levenson 1997).
- Regulation and generation may not be separable. The paper names this as the field’s first open challenge (p. 286), citing Frijda’s view that regulation is “part and parcel of emotion”. It keeps the distinction but asks for “high” thresholds before inferring regulation from an absent response.
Standing in the record
Filed on 2026-10-03 at the owner’s request, in the batch on emotion theory and
psychopathology. It is the record’s statement of what emotion regulation is.
It ties emotion-and-affect to psychopathology-and-treatment, since the paper
puts dysregulation at the centre of mental disorder and calls interventions on
regulation “the staple of psychotherapy” (p. 280).
Its definition of emotion is borrowed, not argued for: a “consensual” model in which an evaluated situation calls forth response tendencies (Figure 1). Its footnote 1 calls that model a distillation of Arnold, Ekman, Izard, Lazarus, Levenson, Leventhal, Plutchik, Scherer and Tomkins. So the paper sits on the appraisal side of the debate this batch files. Lazarus’s Emotion and Adaptation (LIT-550) and Frijda’s The Emotions (LIT-546) are both cited here as sources of that model, and both are filed Deferred. Its regulation scheme inherits the model’s stimulus–evaluation–response order. Barrett’s constructionist account (LIT-542) argues that this “stimulus→response” picture is wrong for emotion generally. Its single stated consequence for regulation is a hypothesis, not a rejection of the five families: reappraisal and other regulation processes “are accomplished with predictions that categorize sensory inputs” (Barrett, Conclusions). Frijda, Ridderinkhof and Rietveld (LIT-540) push Gross’s own open challenge further. They deny “the exclusivity of emotion regulation processes” and hold that much regulation is itself one emotion’s action readiness checking another’s.
No instruction for machine-learning practice, and no anthology topic holds it.