The First Surprise: Labeling Can Quiet the Feeling
The early behavioral evidence on affect labeling asks a narrow but consequential question: what happens to a person’s reported emotion when they label an affective stimulus instead of merely watching it? The answer was that labeling often reduced immediate subjective intensity, even though people expected it to make negative experiences worse. Understanding that result requires keeping three things separate—the observed feeling, the comparison strategy, and the participant’s expectation—because together they show why affect labeling can count as regulation without being deliberately used as one.
What participants actually reported
The core experiment is easy to describe. Participants viewed emotionally evocative pictures, including negative images, and then reported how distressed they felt. In the labeling condition, they selected a word that described the emotional content of the image. In the watching condition, they looked at the same kind of material and let their response unfold without making an overt labeling response. The comparison was not between feeling and feeling nothing. It was between staying with an affective stimulus while categorizing its emotional content and staying with it without that categorization.
Across the first three studies of a four-study report, the people who actually performed the task reported less distress when labeling negative images than when watching them. In the first study, for example, the average distress rating across negative images was 6.25 in the labeling condition and 6.64 in the watching condition. The difference is modest in absolute rating terms, but it points in the same direction as earlier work showing dampened affect-related neural responses during labeling. The behavioral result matters because it extends the question beyond what the brain signal does: it suggests that participants themselves experienced a change in the emotional response.
The effect was not uniform across all kinds of negative stimuli. The report found a clearer difference for moderately negative images than for the most extremely aversive images in some comparisons. That pattern is plausible without being definitive. Very intense stimuli may leave less room for a brief categorization task to reduce the reported response, or the task may operate differently when the initial reaction is already near the top of the scale. The important point is not that labeling works only at one intensity. It is that the effect should not be imagined as a fixed amount of relief added to every emotional situation.
The fourth study changed the valence of the pictures. Instead of negative images and a distress rating, participants viewed positive images and reported how pleasurable they found them. Labeling reduced reported pleasure relative to watching. This result complicates the comforting phrase “labeling calms you down.” If the operation merely removed the unpleasant part of an experience, it should not necessarily reduce a pleasant response. The observed pattern is more consistent with attenuation: labeling may dampen the intensity of affect in either direction, moving a person toward a less charged state rather than reliably toward a happier one.
That distinction is easy to miss when most examples involve fear, sadness, or anger. Suppose a person labels an upsetting scene as frightening and then reports less distress. It is tempting to describe the result as relief. But if the same kind of labeling also makes an exciting or joyful scene feel less pleasurable, the more cautious interpretation is that the emotional response has been muted. Muting can be useful when the response is overwhelming. It can also be undesirable when intensity is part of a positive experience. The outcome has to be named before it can be evaluated.
The report also compared labeling with two more deliberate strategies. Reappraisal asked participants to generate a less distressing way of thinking about the image. Distraction asked them to attend to something unrelated to the image. Both strategies reduced self-reported distress in the negative-image studies. Reappraisal produced a larger reduction than labeling in the direct comparison, while labeling and distraction produced reductions of similar general magnitude in the relevant study. These results do not make the strategies equivalent. They show that different operations can attenuate the same self-reported outcome while reaching it through different routes.
Reappraisal has an explicit goal: change the emotional response by changing the meaning assigned to the event. Distraction has an explicit attentional goal: reduce contact with the emotional material by thinking about something else. Labeling, by contrast, keeps the person engaged with the emotion or the affective content of the stimulus. The fact that it can produce a comparable direction of change despite that continued engagement is the distinctive finding. A lower distress rating after labeling cannot be assumed to result from the same process as a lower rating after reappraisal or distraction.
The relationship becomes clearer when the task is written as a set of observable contrasts:
The diagram puts the data and the interpretation in the same frame, but it does not turn them into one causal chain. The direct observations are the differences in reported experience under the specified conditions. “Incidental attenuation” is the authors’ interpretation of the relationship between those observations and the participants’ expectations. The positive-image result supports the attenuation part of the model; it does not tell us whether the same operation would improve a later decision, a relationship, or a personally meaningful act of reappraisal.
The prediction gap changes what “regulation” means
The most unusual part of the early report was not only that labeling reduced distress. It was that people did not expect it to do so. Some participants completed the task and reported their experience. Others were asked to predict how distressing it would be to watch or label negative images. The predictors generally expected labeling to produce more distress than simply watching. Their expectation was almost the reverse of the experiencers’ report.
That gap cannot be dismissed as a difference between people who had different personalities. In later studies, some participants made predictions after they had already gone through the experience protocol. They had just reported less distress while labeling than while watching, yet when asked to imagine similar future trials, they still tended to expect labeling to be more distressing. The recent experience did not readily update the expectation. This makes the finding more than a mismatch between an uninformed guess and a measured result. It suggests that people’s intuitive theory of labeling can remain detached from the effect they have just experienced.
The authors use “incidental emotion regulation” to describe this pattern. “Incidental” does not mean that the person is unaware of the act. Participants know that they are choosing or producing a word. Nor does it mean that the act is effortless or entirely automatic. The narrower point is about purpose: people are not necessarily labeling in order to regulate emotion, yet the labeling operation has a regulatory consequence in the task. The consequence arrives as a by-product of the stated activity rather than as its explicit objective.
This is different from the way participants approach reappraisal and distraction. When asked to reappraise, people are told to generate a way of thinking that should decrease their emotional response. When asked to distract, they are told to think about something other than the image. Participants’ predictions about those strategies generally followed the intended direction: they expected them to reduce distress. Their expectations were not perfectly calibrated—the report notes that they could overestimate the effects of intentional strategies—but they recognized the basic purpose. Labeling was the unusual case in which the measured direction and the expected direction diverged.
The prediction gap is useful evidence, but it should not be inflated into proof of an unconscious switch. A prediction is a belief about a future experience, not a direct measurement of the mechanism that produces that experience. Participants may have assumed that attending closely to an upsetting image and naming its negative content would make the image more salient. They may have interpreted the instruction as an invitation to focus on what was wrong. They may also have lacked a familiar concept for a process whose benefit, if present, is subtle. The experiments establish the mismatch; they do not settle why the mismatch persists.
The comparisons also have limits that matter for the strength of the conclusion. Watching an image is a useful baseline, but it is not an absence of processing. Participants still attend, interpret, and respond. Labeling adds categorization and language; reappraisal adds a goal to change the response; distraction adds a competing object of attention. Therefore, a difference between conditions can reflect the intended operation together with its cognitive demands and expectations.
Self-report creates a related interpretive boundary. A distress rating captures the participant’s conscious assessment of the experience, which is exactly why the early work was valuable: it examined a channel that neural measures alone cannot provide. But the rating does not reveal whether the person changed the meaning of the image, became more able to generate alternatives, or would respond differently later. Nor does it distinguish every route by which a lower intensity might arise. The evidence supports a claim about subjective attenuation under the task, not a complete account of emotional regulation.
The target of labeling narrows the application still further. The participants were generally labeling the emotional content of a picture, not describing the full history of their own distress. The image supplied a controlled stimulus and, in several tasks, a constrained choice of words. That control makes the comparison interpretable. It also means that a person’s response to an image should not be treated as a direct model of naming a complicated feeling in a conversation, a journal, or therapy. Personal relevance, social feedback, and the possibility of several emotions at once may change the operation.
The report’s comparisons with reappraisal contain one more caution. Reappraisal instructions explicitly ask for a reduction in emotion, so a self-reported decrease may include both a genuine change and some pressure to comply with the task’s demand. Other evidence indicates that reappraisal affects emotional processing beyond a mere reporting instruction, but the early authors still treat demand as a factor that can make the size of a self-report comparison difficult to interpret. Labeling does not carry the same explicit instruction to feel better, which may partly explain why its measured effect is smaller. This is a reason to interpret relative magnitudes carefully, not a reason to discard the comparison.
Taken together, the early experiments support a precise but limited conclusion. Affect labeling can attenuate the immediate subjective response to negative and positive emotional images relative to watching them. People may not recognize that consequence in advance, even after experiencing it, which is why the process is described as incidental rather than intentionally regulatory. Labeling shares an outcome direction with reappraisal and distraction, but it is not the same operation, and the study does not show that it changes the meaning of an event or improves later flexibility.
That last distinction is the unresolved pressure point. If labeling reduces the immediate force of an emotion, it might seem natural to expect it to prepare the mind for reappraisal. But a quieter response is not necessarily a more malleable interpretation. The early report establishes the first half of the problem; it does not tell us whether adding a label before an intentional reinterpretation helps the second operation, leaves it unchanged, or gets in its way. A responsible reading keeps that question open rather than treating the first behavioral success as a universal rule.