Naming a Feeling: At What Time and Under What Intensity?

Naming a feeling does not necessarily make it lighter. Selecting a word may clarify an experience, but it may also keep attention on that experience. A short instruction such as naming an emotion to feel better can therefore skip two questions worth investigating: when does the naming happen, and how intense is the initial negative situation? A laboratory study of affect labeling shows why the same activity need not produce a change in the same direction under different conditions.

Einat Levy-Gigi and Simone Shamay-Tsoory’s 2022 research contained two experiments. The first varied the timing of labeling; the second varied the intensity of the negative material. Participants were not asked to tell their life stories or devise solutions to a personal problem. In a computer task, they selected whichever of two labels best described their feeling. The measured outcome was reported distress during the task, not treatment success. This narrow frame specifies what was compared rather than making the evidence unimportant. [1]

Naming was not the same task as reinterpreting

Affect labeling describes putting a felt emotion into words. It need not require replacing an unpleasant interpretation with a favourable account of what is happening. That distinction matters: reappraisal can involve constructing another perspective on an event, whereas labeling can preserve the negative emotion’s name. A person identifies what an event evokes without necessarily being asked to transform its meaning. The researchers’ interest was not simply a new title for positive thinking.

The study used a choice between two supplied emotional labels, rather than unrestricted selection of any word. That helped create a comparable task, but it also means the activity was not identical to keeping a diary, having a long conversation or sharing an emotion with another person. The provision of the words was part of the procedure. Interpreting the laboratory evidence requires resisting the temptation to combine all these activities into one process merely because each involves language.

Timing changed in the first experiment

In Experiment 1, 63 participants worked with 60 stimuli selected as highly negative and emotionally arousing. Each stimulus appeared twice: once in a labeling trial and once in a control condition based on viewing without emotional labeling. Presentation order was counterbalanced. Responses under labeling and control could therefore be compared for the same material. Timing was a separate condition assigned between people, however; each person did not undergo all three timing arrangements.

Twenty participants selected labels while viewing the stimulus, twenty-one selected them immediately afterwards and twenty-two were assigned to the condition with a ten-second delay after presentation. Assignment was random. The corresponding intervals were also used for colour-frame choices in the control task. This detail helped avoid a simple comparison in which only one activity involved additional waiting. The intention was to distinguish labeling itself from the timing at which a task response occurred. [1]

After the choice, participants rated distress on a nine-point scale, with higher values indicating more distress. How accurately a selected label described their feeling was assessed separately. Finding a word appropriate and reporting less distress were consequently not collapsed into one measure. A label that seems to fit is not necessarily a label that provides relief. The second experiment would make that distinction especially relevant.

In the first experiment’s overall analysis, reported distress was lower under labeling than under control. The analysis did not find a significant difference in labeling’s effect across the three timing conditions. This does not establish that time is unimportant in every emotional process. The tested contrast involved simultaneous choice, choice immediately afterwards and a ten-second delay in this particular task. Failure to find a significant difference is not proof of equivalence among all possible delays or everyday situations.

Intensity changed the direction in the second experiment

Experiment 2 recruited 82 people. Two were excluded because of technical problems and one withdrew before finishing, leaving 79. Labeling was now simultaneous with the stimulus, while low and high negative intensity were compared. Participants encountered both labeling and control tasks and both intensity levels. The central comparison therefore concerned responses across conditions within the same people, rather than merely observing separate groups exposed to one intensity each.

Low and high intensity referred to properties of the experimental material, including its arousal and pleasantness ratings. High negative intensity was not defined solely by the emotional word a participant happened to choose. The material groups differed in both arousal and negativity. The experiment thus does not supply a single numerical cutoff separating mild from intense emotion in every real-life event. Its conditions were defined by the stimulus groups used in the procedure. [1]

Under high intensity, labeling was associated with lower reported distress than the control task. Under low intensity, the direction reversed: more distress was reported after labeling. The authors tested these contrasts statistically. Summarizing the study only as evidence that naming feelings helped under intense conditions would hide the other half of the result. Its distinctive contribution was a pattern with different directions depending on the condition, rather than a universally reassuring slogan about one activity.

A proposed explanation was not a demonstrated cause

The authors discuss the possibility that negative emotional labels made related meanings and feelings more accessible. With a relatively mild initial experience, that process might direct attention towards negativity and raise distress. This is an explanatory proposal; the relevant semantic network was not directly measured and demonstrated to cause the outcome. Turning the low-intensity finding into a rule that anyone discussing a feeling necessarily worsens it would go beyond the task format and the period actually measured.

The experiments used a computer task and self-reported outcomes. A lower rating does not by itself show that every physiological response decreased or that the person remained better off days later. Participants were also selected using exclusion criteria concerning psychiatric diagnoses. These were not trials of treatment in a clinical population. Interpreting the timing experiment as proof that waiting ten seconds before an argument is pointless would likewise combine different behaviours: waiting and labeling here were structured responses to experimental material, not a test of resolving interpersonal conflict.

An evidence-based conclusion is to evaluate emotional naming alongside the task used, the starting condition and the outcome measured. In these experiments, the intensity comparison revealed a more distinctive difference than the short timing contrasts did. Recognizing an emotion and reducing it are not identical aims, so naming need not be valued solely by whether it brings relief every time. The research has not tested all those alternative aims either. Its most careful contribution is that selecting a word participates in an emotional experience, and its influence should not automatically be assumed to run in only one direction.

Sources

[1] Levy-Gigi, E., and Shamay-Tsoory, S. (2022). Affect labeling: The role of timing and intensity. PLOS ONE, 17(12), e0279303. Methods of both experiments, sample details, self-report measures, findings and explanatory limits directly examined. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0279303

The source is published under a Creative Commons Attribution licence. Independently written explanation, not a reproduction or translation of the source prose. This article does not provide treatment recommendations. Source access and factual review: October 14, 2026.