When Naming Gets in the Way of Reappraisal
The key test for affect labeling is not whether a person can feel less emotion after naming it. It is whether naming changes what happens when the person must immediately reinterpret the same emotional material. This chapter compares the direct four-condition experiment from 2024 with a two-study replication and extension published in 2026. The evidence supports a specific judgment: reappraisal worked better on its own than when it was preceded by the tested labeling operation, while the separate claim that labeling alone makes people worse remains much less secure.
The four-way test of sequence
The 2024 study made the sequencing question explicit by crossing two factors. One factor was affect labeling: participants either selected an emotion label that matched their response or selected nonsense syllables as a control. The other factor was reappraisal: participants either tried to reduce their negative emotion by reinterpreting the stimulus or focused on the emotion without reappraising it. This produced four conditions:
- affect labeling followed by reappraisal;
- no affect labeling followed by reappraisal;
- affect labeling without reappraisal; and
- no affect labeling without reappraisal.
The design matters because it prevents a headline from hiding the comparison. To ask whether naming interferes with reappraisal, the relevant contrast is not labeling versus no labeling in general. It is reappraisal after labeling versus reappraisal without labeling. The other two conditions help show what the operations do when separated, but they do not answer the sequencing question by themselves.
Participants were adults and university students who imagined negative emotional vignettes, rated the intensity of the emotion, and then completed the assigned task. The vignettes evoked fear, anger, or sadness. In the labeling conditions, the person selected an emotion word; in the no-labeling conditions, the person selected a nonsense sequence. In the reappraisal conditions, the instruction was to reinterpret the vignette in a more positive or calmer way. In the no-reappraisal conditions, participants focused on the emotions elicited by the vignette.
The behavioral result had two parts. First, reappraisal reduced negative emotion. That is the expected direction for an intentional strategy whose purpose is to change the response. Second, reappraisal was less effective when it followed affect labeling. The combined affect-labeling-plus-reappraisal condition produced a greater change in negative emotion than the no-labeling-plus-reappraisal condition in the direction indicating a smaller reduction. The reported comparison was statistically significant, with a Hedges’ g of 1.71 for the relevant post hoc contrast.
Affect labeling by itself did not show a clear independent benefit in the comparison between labeling without reappraisal and the no-labeling, no-reappraisal condition. This detail is essential. The study did not show that naming an emotion invariably intensifies negative feeling. It showed that adding a naming operation before an intentional reinterpretation changed the effectiveness of that reinterpretation. The interaction is the finding; the broad slogan about naming alone is not.
This is the value of crossing the factors instead of comparing two broad groups. A simple label-versus-no-label comparison would mix together people who were asked to reappraise and people who were not. A reappraisal-versus-no-reappraisal comparison would show that reappraisal can change the response but would not reveal whether naming altered that change.