Reframing a bad situation in a more positive light is among the most widely repeated pieces of psychological advice in circulation. A study published in Psychological Science in 2013 found that skill at doing it was associated with fewer depressive symptoms in some people and more depressive symptoms in others. The thing that divided them was not how good they were at it. It was whether the situation causing them trouble was one they could act on.

The study, by Allison Troy, Amanda Shallcross and Iris Mauss, recruited 170 adults from the community who had recently been through a stressful life event. Participants watched a neutral film clip and then three sad ones, and were asked to reappraise what they saw, meaning to deliberately reinterpret it in a less upsetting way. That gave the researchers a measure of how well each person could do it on request. Their real-world stressors were separately classified as controllable, such as trouble with performance at work, or uncontrollable, such as a spouse’s illness.

We are writers, not clinicians. What follows is a reading of the research, not advice about how anyone should handle a difficult stretch.

What the study actually measured

Among participants whose main stressor was outside their control, greater reappraisal ability was associated with fewer depressive symptoms. Among those whose stressor was something they could address, the association ran the other way. Higher reappraisal ability went with more depressive symptoms, not fewer.

The authors’ reading is that talking yourself into a calmer view of a problem you could otherwise fix may reduce the pressure to fix it.

That is an interpretation of a correlational pattern, not a demonstrated mechanism. The design can show that the association reverses across contexts. It cannot show that reappraisal caused anyone’s symptoms to rise or fall.

The paper also carries a 2016 corrigendum, which is the sort of thing that rarely survives into secondary coverage. Depression scores for seven of the 170 participants, about 4 percent of the sample, had been exported on the wrong scale. The authors reran the analyses and reported that “the substantive results, their interpretation, and the conclusion remain completely unchanged.” One borderline result moved from p = .07 to p = .04. The three-way interactions the argument rests on did not change.

This is one study, not settled consensus

This is one study, not settled consensus. An experiment with 170 people recruited from a single community cannot establish a general rule about how anyone should handle stress, and the result would want replicating in other populations before it carried much weight on its own.

What makes it worth attention is that it lines up with an argument that had been building in coping research for years: that the field has been asking the wrong question.

The fallacy of uniform efficacy

In a 2013 review in Perspectives on Psychological Science, George Bonanno and Charles Burton gave that wrong question a name. They called it the fallacy of uniform efficacy: the assumption that particular coping strategies are consistently good or consistently bad, independent of the situation a person is in.

Their case against it rests partly on evidence that refuses to line up. Problem-focused coping is generally assumed to beat emotion-focused coping, and they point out there is also evidence for the reverse. Suppressing the expression of an emotion is widely treated as the unhealthy option, and in the meta-analytic evidence they summarize, covering hundreds of experimental comparisons, differences between strategy types were modest and suppression often proved effective. We were not able to open that meta-analysis ourselves, so this is their characterization of it rather than our reading of the original.

A second problem is more awkward for the whole idea of a coping style. Bonanno and Burton report that the within-person consistency of coping strategy use across situations is, in their phrase, surprisingly modest. People do not reliably do the same thing twice.

Three steps, and the trouble with measuring them

What they propose instead is a sequence with three parts. Context sensitivity is the ability to “perceive impinging demands and opportunities from the situational context.” Repertoire is “the ability to utilize a wide range of regulatory strategies that might accommodate divergent contextual demands.” Feedback responsiveness is “the ability to monitor and use feedback about the efficacy of a chosen regulatory strategy over time so as to adjust or correct behavior when needed.”

Read the situation, have more than one option, and notice when the option you picked is not working.

Todd Kashdan and Jonathan Rottenberg, working separately, arrived somewhere similar in a 2010 review in Clinical Psychology Review, describing psychological flexibility as the capacity to “recognize and adapt to various situational demands” and to shift approach when the current one is causing problems. They were also direct about the difficulty. Flexibility, they wrote, is a slippery construct to define, and “dynamic constructs require dynamic approaches. Shortcuts using static measurement approaches are perilous.” A questionnaire asking people how flexible they are is exactly such a shortcut.

Where the framework runs into trouble

The sharpest criticism of this line of research comes from the people conducting it.

In a 2024 review in the Annual Review of Psychology, Bonanno and colleagues report that stable psychological health is the most common outcome after a disaster, averaging around two thirds of people across the studies they surveyed. Those figures are averages pooled across separate studies rather than a breakdown of any single sample, and not every study identifies every trajectory.

Then comes the part that complicates everything above it. The same review states that resilience scales “do not hold up to their promise when tested in longitudinal or prospective research.” Individual predictors carry small effect sizes. Factors that look protective turn out to have costs in some contexts, including social support, which can increase distress depending on the situation. A 2023 paper in Nature Reviews Psychology from an overlapping group of authors describes the same gap, noting the limited capacity of the known correlates to predict who will follow the resilient path.

So the field can describe adaptation after the fact with reasonable confidence. Predicting it in advance is a different matter, and by the reviewers’ own account, not going well.

What would settle it

Flexibility is a better-supported description of how people adapt than any ranked list of coping strategies. It is still a description rather than a validated prediction, and the distinction matters when the idea gets sold as a skill with known returns.

Settling it would take measures that track what people actually do across situations over time, rather than what they report about themselves on one afternoon, and prospective studies that commit to a prediction before the adversity arrives instead of accounting for the outcome afterward.

If a difficult period is affecting your daily functioning, a clinician is better placed to help than any research summary, this one included.

Bonanno and colleagues note that machine learning models have not yet illuminated the mechanisms either. Whether the sequence they describe can be taught, and whether teaching it changes any outcome, is a further question again. No intervention trial was examined for this article.