EQUORA Institute · Standing summary · Evidence quality
Mental Fast Food™ · 19 · Research byproduct
Anxiety takes up room
Interference topic · A true mechanism, the wrong subject
What if what was measured describes a single morning, and we read it onto a whole generation?
01 · The saying you swallowed
“Today's young people are emotionally more insecure than any generation before them, and that is why they cannot learn.”
A real observation. It contains two claims, and the two were measured in entirely different places.
The first on a single person, in a single hour, against a control condition. A student sits down with a task, and how many items they hold in mind while solving it can be measured. When something else occupies their attention, fewer. This part comes from the laboratory, and the measurement can be repeated.
The second on populations, across three decades, with a questionnaire whose completion and interpretation shifted over that same period. Fifteen-year-olds in forty-one countries, every four years, half a million respondents. This part works at epidemiological scale, and it answers quite different questions.
We call both halves of the sentence science. The next four sections look at what each of them can actually carry.
02 · Occupied room
The same head, less free space
Working memory has a limited capacity, and anxiety draws on that limited pool. Eysenck and colleagues are more precise than that: in loading experiments, anxiety impaired main-task performance when the secondary task loaded the central executive, and did not when it loaded the phonological loop. Worry does not disturb in general; it ties up one specific resource.
The demo below imitates that same loading design. You will do two rounds: in the first you simply memorise six numbers, in the second you do the same while adding up eight digits in your head. The second round typically comes out lower.
Important: this simulates load, not anxiety. We ask for no real worry and set out to induce none. The experiments found that anxiety is one of the things that occupy this capacity — the demo shows what occupation means.
What one round does not prove
Your own result is one measurement on one person, so it proves nothing on its own. Some people do better on the second round because they have warmed up to the task by then. The effect shows up at group level, across many participants and under controlled conditions. The demo makes visible what the experiments measure, rather than repeating them.
03 · Which question does it answer?
Eight statements, three kinds of measurement
Each of these statements circulates in the public conversation on this topic. Put each into the bin where you think it was measured, then open the reveal. The question here is the unit of measurement: whether what we know concerns one person or a population.
04 · Before and after the pandemic
The same data, one survey wave apart
The eight-item HBSC symptom list has been unchanged since 1994, so the objection about shifting instruments does not apply here. The authors also ran the models without the 2022 wave, to see how much the data collected during the pandemic adds. The two panels below show those two runs: how many countries show which shape of trend.
Up to 2018
Before the pandemic
Up to 2022
Including the pandemic wave
Which reading does this support?
The rise is a pandemic effect.
The rise predates the pandemic, which accelerated it.
The pandemic started a rise in a group that had none.
A note on the country counts
The two columns of the published table rest on different numbers of countries: forty-one in the 2018 column and thirty-eight in the 2022 one, because Russia, Ukraine and the United States have no 2022 wave. The 2018 figures shown here exclude those three, so that both panels stand on the same base. Direction and magnitude are unaffected: for boys' somatic complaints, “no clear trend” falls from twenty-five to thirteen, and for psychological complaints from twenty to four.
The table itself received a corrigendum in October 2025, because the first two header rows were misaligned in the published version. The authors found it, corrected it, and the conclusions were unaffected.
05 · What frees up room?
Seven suggestions in the order you would put them
Click them in the order in which you think the strength of their evidence declines. The first one you click takes position 1. The reveal shows what was measured behind each.
What the ordering shows is a more general rule. An intervention is more reliable the fewer theories you have to believe for it to work. Sleep and load reduction hold even if the explanation behind them is wrong, because they reach the capacity directly. The ten-minute writing exercise stands only if the single mechanism it rests on is true — and that is precisely why it vanished when others tried to reproduce it.
Where this page stops
This page is about the learning situation, meaning the few hours in which someone sits down in front of a task. Persistent anxiety and deeper difficulty are a different question, and the capacity explanation says nothing to suggest that they are any less real or serious. Anyone living with that is best served by talking to a professional, and nothing read here replaces that.
TakeawayThe mechanism describes a morning, the trend describes a generation, and the distance between the two has not yet been measured by anyone.
The capacity is the same as it was. What has changed is how much of it is already spoken for.
Sources
The mechanism
Eysenck, M. W., Derakshan, N., Santos, R. & Calvo, M. G. (2007) — Anxiety and Cognitive Performance: Attentional Control Theory. Emotion 7(2), 336–353. Attentional control theory, an extension of the 1992 processing efficiency theory; anxiety degrades the efficiency of the central executive. doi:10.1037/1528-3542.7.2.336
Eysenck, M. W., Payne, S. & Derakshan, N. (2005) — Trait anxiety, visuo-spatial processing, and working memory. Cognition and Emotion 19, 1214–1228. The load design: anxiety impaired performance when the secondary task loaded the central executive, but not when it loaded the phonological loop doi:10.1080/02699930500260245
Eysenck, M. W. & Calvo, M. G. (1992) — Anxiety and performance: The processing efficiency theory. Cognition and Emotion 6, 409–434 doi:10.1080/02699939208409696
The population trend
Schrijvers, K. et al. (2024) — Three Decades of Adolescent Health: Unveiling Global Trends Across 41 Countries in Psychological and Somatic Complaints (1994–2022). International Journal of Public Health 69, 1607774. HBSC data from eight waves, 41 countries, 470,797 fifteen-year-olds; of the 164 cases examined, 82.3% show an increase, 14% no clear trend, and 3.7% a decrease. doi:10.3389/ijph.2024.1607774
Schrijvers, K. et al. (2025) — Corrigendum. International Journal of Public Health 70, 1609116. A correction to the first two header rows of Table 3. doi:10.3389/ijph.2025.1609116
Ottová-Jordan, V. et al. (2015) — Trends in multiple recurrent health complaints in 15-year-olds in 35 countries, 1994–2010. European Journal of Public Health 25(Suppl. 2), 24–27. The baseline against which, by 2018, the pattern had turned upward in considerably more countries doi:10.1093/eurpub/ckv015
Inchley, J. et al. (2023) — HBSC Study Protocol: Background, Methodology, Mandatory Questions and Optional Packages for the 2021/22 Survey. MRC/CSO Social and Public Health Sciences Unit, University of Glasgow
The interventions
Ramirez, G. & Beilock, S. L. (2011) — Writing About Testing Worries Boosts Exam Performance in the Classroom. Science 331(6014), 211–213. doi:10.1126/science.1199427
Myers, S. J., Davis, S. D. & Chan, J. C. K. (2021) — Does expressive writing or an instructional intervention reduce the impacts of test anxiety in a college classroom? Cognitive Research: Principles and Implications 6. Across four real exams neither intervention reduced test anxiety or improved performance; the paper also cites a well-powered 2018 double replication that ended with a slightly negative effect size. doi:10.1186/s41235-021-00309-x
Roediger, H. L. & Karpicke, J. D. (2006) — Test-Enhanced Learning: Taking Memory Tests Improves Long-Term Retention. Psychological Science 17(3), 249–255. The effect of retrieval practice, replicated many times since. doi:10.1111/j.1467-9280.2006.01693.x
Sweller, J. (1988) — Cognitive Load During Problem Solving: Effects on Learning. Cognitive Science 12(2), 257–285. The founding paper of cognitive load theory. doi:10.1207/s15516709cog1202_4
Ward, A. F., Duke, K., Gneezy, A. & Bos, M. W. (2017) — Brain Drain: The Mere Presence of One's Own Smartphone Reduces Available Cognitive Capacity. Journal of the Association for Consumer Research 2(2), 140–154. Two experiments; the effect grows with the degree of phone dependence and is undetectable at low dependence. doi:10.1086/691462
Ruiz Pardo, A. C. & Minda, J. P. (2022) — Reexamining the “brain drain” effect: A replication of Ward et al. (2017). Acta Psychologica 230, 103717. A pre-registered direct replication of the second 2017 experiment, with identical tasks and conditions; the effect failed to appear, together with the moderation by phone dependence. doi:10.1016/j.actpsy.2022.103717
Sisk, V. F., Burgoyne, A. P., Sun, J., Butler, J. L. & Macnamara, B. N. (2018) — To What Extent and Under Which Circumstances Are Growth Mind-Sets Important to Academic Achievement? Two Meta-Analyses. Psychological Science 29(4), 549–571. The first meta-analysis pooled 273 effect sizes from 365,915 students (r ~ 0.10), the second 43 intervention studies from some 57,000 students; in the intervention meta-analysis lower-achieving and lower-status students benefited more, while in the correlational meta-analysis the effect proved independent of both. doi:10.1177/0956797617739704
Yeager, D. S. et al. (2019) — A national experiment reveals where a growth mindset improves achievement. Nature 573(7774), 364–369. A small but real effect, mainly among lower-achieving students. doi:10.1038/s41586-019-1466-y
Macnamara, B. N. & Burgoyne, A. P. (2023) — Do Growth Mindset Interventions Impact Students’ Academic Achievement? A Systematic Review and Meta-Analysis With Recommendations for Best Practices. Psychological Bulletin 149(3–4). 63 studies, N = 97,672. A summary of the post-2018 literature: the apparent effect of growth-mindset interventions traces largely to design flaws, reporting gaps and bias; in the six studies following best practice the effect is not significant (d = 0.02). doi:10.1037/bul0000352
This page comes out of a research programme at the EQUORA Institute and records one state of it, in place of a closed institutional position. That state rests on the results available at the moment of publication; anything published later is included only where the page was updated, and the date shows this. AI takes part throughout the research as a thinking partner; responsibility for interpretation and publication remains human.