does training people to see stress as helpful make them perform, feel and recover better?
“they have basically trained people to view stress as positive during the stressful situations. In the lab, people do better. They perform better.”
a guest, a stress scientist said on Huberman Lab, 2026-10-08, at 15:12. hear it in context → the speaker cited: studies by Wendy Mendes and others.
first, the part that holds
the core is real: randomized lab studies, including ones from the lab named in the claim, find that telling people their stress arousal is useful improves performance, and a 2024 meta-analysis of 35 randomized trials (5,638 people) found a statistically significant performance benefit.
what we found when we went and looked
we read the original trials and the pooled evidence. performance holds up, but it is small: the 2024 meta-analysis pooled d = 0.23 (95% ci 0.14 to 0.33), which fell to d = 0.14 after correcting for likely missing studies, and stress-mindset-only interventions came in at d = 0.18. the "feel more positive emotion" clause is contradicted by the named lab's own key experiment, which found no effect on self-reported positive or negative emotion. the evidence for "recover faster" is one cardiac measure in one study of a combined intervention. the headline studies are tiny: the well-known exam study analyzed 28 people, and the cardiovascular study had about 16 per group.
assertion by assertion
| assertion | verdict | what the record says |
|---|---|---|
| people trained to view stress as positive perform better in lab studies | confirmed | meta-analysis of 44 effect sizes from 35 randomized trials (n = 5,638): d = 0.23, 95% ci 0.14 to 0.33, p < 0.001. the authors call this a 'small' improvement. there was funnel asymmetry, and trim-and-fill lowered it to d = 0.14 (still significant). p-curve 'could neither support the presence nor absence of an evidential value.' direction confirmed; size is small. [1][3] |
| they feel more positive emotion | false | in the named lab's randomized stress-test experiment (49 analyzed, about 16 per group), 'analyses of task demands, subjective stress, and positive and negative emotions produced no significant effects, Fs < 1.' reappraisal changed perceived resources and cardiovascular profile, not reported emotion. a later multi-study trial of a combined growth-plus-stress mindset intervention reported better well-being and less anxiety, but that is a different intervention and a different outcome from in-the-moment positive emotion in the lab. [2][4] |
| they problem solve better | overstated | cognitive written tasks show a small pooled benefit (d = 0.20, smaller than public-performance tasks at d = 0.34). the exam study found a math benefit (practice math d = .82, actual math d = 1.03, n = 28) and no verbal benefit. the effect is narrow and inconsistent across task types. [1][3] |
| they recover faster | overstated | the only recovery evidence we found is one study (study 3, n = 160) in a combined growth-plus-stress mindset trial. it reported a pre-ejection period effect of 2.13 ms during recovery. that trial also found that the stress-mindset arm alone did not reliably reduce the threat marker (posterior probability 0.785). the named lab's 2012 experiment did not measure recovery. [2][4] |
| this comes from studies by the named lab and others | confirmed | the named researcher is a co-author on the 2010 exam study and the 2012 cardiovascular study. the wider literature comes from many groups. [2][3] |
both directions
for the speaker: randomized trials consistently point the same way on performance. the pooled effect survives correction for publication bias, and study quality did not moderate it. the named lab's exam study found the benefit carried over to real exam math scores 1-3 months later. a large multi-study trial (studies of 2,717, 755, 160, 200 and 118 students) found a combined growth-plus-stress mindset intervention improved cardiovascular stress responses, cortisol, anxiety and course pass rates (63% vs 47% in study 5). mixed interventions showed the largest pooled effect, d = 0.45.
against: the pooled effect is small: d = 0.23, or 0.14 after bias adjustment. heterogeneity is moderate (i² = 52.6%), and p-curve could not confirm evidential value. the foundational studies are very small: 28 analyzed in the exam study after 60 were recruited, about 16 per group in the cardiovascular study. the named lab's own experiment found no change in positive or negative emotion, which contradicts the emotion clause. the strongest field results come from a combined intervention. in that trial the stress-mindset-only arm did not reliably reduce the cardiovascular threat marker. 'recover faster' rests on a single cardiac measure in one study.
what this cannot say
most trials measure short-term effects on lab or classroom tasks in students. none of them test people with diagnosed anxiety disorders or anyone in psychiatric treatment, so they say nothing about clinical populations. the meta-analysis did not pool emotion or recovery outcomes. we read pmc full texts through a summarizing fetch tool and did not reread every table by hand. we did not have full access to the publisher's version of the 2010 exam study and relied on its pmc author manuscript. we may have missed individual trials that measured positive emotion or recovery directly.
sources, each one fetched and read
- Effectiveness of stress arousal reappraisal and stress-is-enhancing mindset interventions on task performance outcomes: a meta-analysis of randomized controlled trials. Scientific Reports, 2024. PMC10994935 What we read there: 44 effect sizes, 35 studies, n = 5,638; d = 0.23, 95% ci [0.14, 0.33]; i² = 52.6%; trim-and-fill adjusted d = 0.14 (p = 0.007); mixed d = 0.45, reappraisal-only d = 0.22, stress-is-enhancing-only d = 0.18; public performance d = 0.34 vs cognitive written d = 0.20 Fetched 2026-10-28 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC10994935/
- Mind over matter: reappraising arousal improves cardiovascular and cognitive responses to stress. Journal of Experimental Psychology: General, 2012;141(3):417-422. PMC3410434 What we read there: 49 analyzed across three conditions; 'Analyses of task demands, subjective stress, and positive and negative emotions produced no significant effects, Fs < 1'; reappraisal improved cardiac output and vascular resistance and reduced attentional bias; no performance or recovery outcome measured Fetched 2026-10-28 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC3410434/
- Turning the knots in your stomach into bows: reappraising arousal improves performance on the GRE. Journal of Experimental Social Psychology, 2010;46(1):208-212. PMC2790291 What we read there: 60 recruited, 28 with actual exam scores; practice math 738.57 vs 683.57, p = .047, d = .82; actual math 770.00 vs 705.71, p = .015, d = 1.03; no verbal difference Fetched 2026-10-28 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC2790291/
- A synergistic mindsets intervention protects adolescents from stress. Nature, 2022;607:512-520. PMC9258473 What we read there: six randomized double-blind experiments; the combined intervention improved cardiovascular reactivity, cortisol, well-being, anxiety and academic outcomes; recovery pre-ejection period effect 2.13 ms (study 3); 'neither of the single-mindset treatments reliably reduced TPR relative to the neutral control condition'; pass rate 63% vs 47% (study 5) Fetched 2026-10-28 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC9258473/
speakers are described by role, not named. one sentence is quoted under fair use, with attribution and a link to the original at the timestamp; we do not re-host audio or video. nothing on this page is a reason to start, stop or change a medication. if you have the file behind anything marked unverifiable, the research board is where people can answer. adam