Contribution
A review that reorganises the emotion literature by mechanism. Instead of asking what distinguishes anger from fear, it asks what each appraisal criterion does to each response component. Emotion labels become “one class of dependent variables” (p. 736), placed at the end of the process.
Key insight
The stable regularities are between appraisal criteria and responses, not between emotion categories and responses. Kreibig’s review of 134 studies found only heart rate and respiration rate stable across emotions (p. 729), but intrinsic valence reliably moves the zygomaticus and corrugator, and control reliably moves vocal energy.
Assumptions
- A process model shared “across many theories” (p. 738): appraisal, action tendencies, physiology, expression, feeling, labelling.
- That “it is the nature of the appraisal process that determines the nature of the remaining stages in the emotion process” (p. 723).
- Five criteria after Ellsworth and Scherer (2003): novelty, valence (intrinsic and extrinsic), agency, control or power, fairness.
Key results
From the studies reviewed:
- Valence to action (p. 725). Positive and negative stimuli prime approach and avoidance in compatibility tasks, though semantic overlap may explain this. Negative stimuli also prime repair: “conflict stimuli (i.e., stimuli that call for two competing actions, such as Stroop stimuli), which are incongruent with the goal to perform well on the task, instigate the tendency to adjust one’s behavior”.
- Intrinsic valence versus goal congruence (p. 730). In Van Reekum et al. (2004), “the manipulation of intrinsic valence had comparatively little impact on physiological responses”, while goal congruence moved heart, skin conductance and finger temperature. In Aue and Scherer, “Intrinsic valence had more robust effects on facial EMG (Aue & Scherer 2011), whereas only goal congruence significantly influenced mean skin conductance, forehead temperature, and finger temperature.” Goal incongruence mobilised more, with less habituation. “The efferent effects of the two appraisals combined multiplicatively.” In the voice (p. 734), goal congruence moved energy, f0 and duration, and intrinsic valence moved spectral energy distribution.
- Sequence (p. 732). Intrinsic valence acts on the face faster than goal congruence (Lanctôt and Hess 2007); goal congruence (about 600 ms) precedes prospective control (about 800 ms) (Gentsch et al. 2015).
- Control (pp. 726–728). High general control, including social power, goes with approach, activity and, depending on salient goals, antisocial or prosocial behaviour. Learned helplessness and self-efficacy research link low control to passivity. Goal congruence effects are “amplified” when control is high and “invariant” when low (p. 732).
- Fairness (p. 729). In an ultimatum game, “participants’ rejection rates of these offers were more influenced by goal incongruence than by fairness” (Bediou and Scherer 2014).
- Semantic sedimentation (sidebar, p. 737). Scherer and Fontaine (2018), 24 emotion terms in 28 languages, find “four appraisal factors (valence, novelty, prospective control, and other-agency)” and four feeling factors, including “strong versus weak”.
- Coherence and awareness (p. 737). An episode begins with “increased synchronization and coherence among the various components beyond a certain threshold”; “a person will become aware of an altered state after a second threshold of coherence has been reached.” The mechanism is a central representation that “serves to monitor and regulate the degree of system coherence.”
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | Appraisal criteria have more stable response correlates than emotion categories | moderate | review of experimental work; Kreibig 2010 for the categorical side |
| C2 | Intrinsic valence and goal congruence have partly different efferent signatures | moderate | a handful of studies (Van Reekum 2004; Aue & Scherer 2008, 2011; Johnstone 2005; Lanctôt & Hess 2007) |
| C3 | High control amplifies goal-congruence effects | weak | Gentsch et al. 2015 |
| C4 | Coherence above a threshold marks an episode; above a second, awareness | weak | proposed, not tested |
Concepts
- intrinsic valence — appraisal of a stimulus “as merely positive or negative”, independent of goals.
- goal congruence — appraisal of a stimulus as serving or obstructing a goal; “one form of extrinsic valence”.
- general, situational, retrospective and prospective control — control over stimuli in general, over the current one, over its having come about, and the expectation of turning it from negative to positive.
- component coherence — synchronisation among components, the proposed criterion both for an episode and for its awareness.
Connections
Updates the 2009 evidence review (LIT-1446, unread) and restates the architecture of Scherer (2009, LIT-1451, NOTE-1113). Frijda and Scherer’s (2009) four functions open it, the same source as the control precedence the record reads through THEORY-055. The valence split is the one argued in Shuman, Sander and Scherer (NOTE-1108). The awareness proposal restates Grandjean, Sander and Scherer (NOTE-1115).
Bearing on the record
Does the process description already contain “appraisal as negotiation among perspectives”? No. It contains most of the parts, and a reader looking for the record’s idea will find places it could go, but the review does not say it.
What is there:
- Several criteria, weighted. Theories “tend to assign different weights to the various appraisal criteria” (p. 722). The outcome is “the net effect” of a profile, in Grandjean et al.’s phrase.
- Recursion. Each component’s effect “is expected to feed back to the eliciting component, setting off another round of processing likely to affect the ultimate outcome” (p. 722).
- Conflict. It appears twice. In conflict monitoring, stimuli “call for two competing actions” (p. 725), and the response is to adjust. And in blends: emotions are mixed in “many cases” (p. 721).
- One criterion prevailing. Bediou and Scherer’s finding that rejection followed goal incongruence more than fairness (p. 729) is, read the record’s way, a case of one ordering prevailing over another in what the agent does.
- Dominance of stages. The four functions “each dominate a different stage of the process” (p. 721). This is sequence, not contest.
What is missing is the frame. In the review an appraisal criterion is a dimension on which one evaluator scores an event, and the score profile drives responses. Criteria do not have rankings of options that compete for authority over action. Nothing wins, persists or resumes. The interaction terms (multiplicative valence effects, control amplifying goal congruence) are statistical interactions of inputs, not the outcome of a contest. The closer precedents in the same school are Sander and Scherer’s “activation and inhibition thresholds” (NOTE-1113) and Shuman et al.’s micro-valence integration (NOTE-1108). CLAIM-440’s part 1 remains the record’s reading of the CPM, not the CPM’s self-description. The record should not cite this review as stating it.
What the review does give CLAIM-440 and QUESTION-268:
- Markers that tell the two valences apart. If the record concedes CLAIM-438 in the way NOTE-1108 suggests, with the potential account covering goal-conduciveness valence and intrinsic pleasantness a separate check, the physiology here is how one would tell which ordering is driving. Goal congruence shows up autonomically, with less habituation when it is negative. Intrinsic valence shows up mainly in facial EMG. These are human, individual markers, so they bear on the animal side of QUESTION-268, not on organizations.
- Control as a gain on the potential ordering. “Amplified goal congruence effects when control was high, in contrast to invariant goal congruence effects when control was low” (p. 732). If goal conduciveness is the ordering that prevails, control modulates how strongly it drives the components, which is what TERM-076 calls intensity. The control appraisal concerns the agent’s power over its situation. It is not the potential of one internal ordering over others, and it is the axis PAD calls “dominance” (see LIT-1437). The rename to potential keeps the two apart; this finding says how they interact.
On CLAIM-439. The two-threshold proposal (p. 737) puts awareness at a degree of the regime itself: a “second threshold of coherence”. That is nearer Veit’s identity (THEORY-101) than to a self-model condition. But the mechanism offered is a representation: a central integration that “serves to monitor and regulate the degree of system coherence.” So the review, like the 2009 paper (NOTE-1113), holds both a threshold view and a model view and does not choose. The model in question represents coherence, not which ordering wins. It also adds temporal chunking: feelings come as “temporal chunks with phenomenal unity” that “might occur when a certain level of component synchronization persists” (p. 737). Persistence is one of TERM-076’s marks of potential, so this is a point of contact with the record’s account. The review gives it as a hypothesis, not a finding. The Grandjean et al. reading (NOTE-1115) takes the comparison further.
Limitations
A review, restricted to experiments with manipulated appraisals; it excludes neuroscience and reappraisal “due to space limitations”. Paths D and E of its own model (feeling, labelling) are not reviewed for lack of a “coherent body of empirical literature” (p. 724), which is where both CLAIM-439 and QUESTION-059 would need evidence. The coherence proposal is untested.
Open questions
Whether the multiplicative combination of intrinsic valence and goal congruence is a weighted integration or a gating, which would place it in one of CLAIM-435’s two regimes. Whether the two coherence thresholds can be estimated per episode with the time-series methods it cites.