How to focus better: what the science of attention really says

Most advice on how to focus better is a list of tips with no indication of which ones are supported by evidence and which are folklore. This page does the opposite. It explains what attention is, gives the measured baseline for how fragmented screen attention has become, then grades six families of tactics by the strength of the evidence behind them, with the effect sizes where the literature reports them.

What attention is: three systems, not one spotlight

Attention is not a single resource that you have more or less of. Since 1990 the dominant model has described at least three separable networks: an alerting network that keeps you ready to respond, an orienting network that selects what to attend to, and an executive network that resolves conflict between competing inputs and keeps a goal active A^[1]. Twenty years of imaging later the same authors kept the three-way split and added anatomical detail, including a second executive network and the different neuromodulators involved A^[2].

The practical consequence is that "focus" problems come in different kinds. Drifting off during a long document is a sustained-alerting problem. Being pulled towards a notification is an orienting problem. Holding a complex plan in mind while a colleague talks is an executive problem. The tactics below work on different networks, which is one reason no single trick works for everyone.

Two further findings frame everything else. Executive functions, the inhibition, working memory and flexibility that make deliberate focus possible, are measurably degraded by stress, low mood, loneliness, poor sleep and physical inactivity A^[3]. And the brain appears to allocate effortful control by weighing its expected payoff against its cost, which is why concentrating on something dull feels aversive even when you are rested B^[4]. Focal working memory, the amount you can hold in mind at once without rehearsal, is about four chunks, not the seven often quoted A^[5]. The interactive explainer How focus works in the brain animates these three ideas.

The baseline: 47 seconds and 275 interruptions

How fragmented is attention at work today? Two measured figures set the baseline, and both come with caveats.

Gloria Mark's group has logged screen activity of information workers since 2004. In a 2024 interview she summarised the trend: average attention to a single screen before switching fell from about two and a half minutes in 2004 to 47 seconds by 2020 B^[6]. The figure comes from the researcher describing her own field data, not from a single peer-reviewed table, so we grade it B.

Microsoft's 2025 Work Trend Index, built on a 31,000-person survey plus Microsoft 365 telemetry, estimates that employees are interrupted every two minutes by meetings, emails or pings, which adds up to about 275 interruptions over a workday including off-hours activity B^[7]. It is vendor research, though the sample is the general workforce rather than Microsoft's own customers.

What interruptions cost is better established than how often they happen. In a controlled simulation of office tasks, interrupted participants finished their work faster but reported significantly more stress, frustration, time pressure and effort B^[8]. Interruptions of under three seconds roughly doubled sequence errors in a procedural task, and interruptions of about four and a half seconds roughly tripled them B^[9]. Simply receiving a phone notification, without touching the phone, raised errors on a sustained-attention task about as much as active phone use had in earlier studies B^[10].

Recovery time after an interruption is best given as a range with its context: a few seconds to resume a well-practised lab task, up to the "almost half an hour" Mark reports from field observation of workers returning to an interrupted project ^[6]. The widely quoted "23 minutes and 15 seconds" sits in that second category: it comes from an interview, not from a published measurement, and we do not cite it as a study result.

Why willpower framing fails

Popular attention advice leans on the idea that focus is a tank of willpower that drains through the day. That idea rests on ego-depletion experiments from the late 1990s. Two multi-site preregistered replications have since tested the effect: 23 labs with 2,141 participants found d = 0.04 A^[11], and 36 labs with 3,531 participants, using a paradigm chosen by the theory's proponents, found d of roughly 0.06 A^[12]. Neither is distinguishable from no effect in practical terms. We cover the full story in decision fatigue after the replication crisis.

The better-supported account is that mental fatigue is a motivational signal about whether continuing is worth the cost; performance recovers when the task becomes more rewarding B^[4]. The practical difference matters: you do not protect focus by rationing it in the morning, you protect it by lowering the cost of the task in front of you (fewer competing inputs, clearer next step) and by not degrading the hardware (sleep, movement, mood).

Tactic grades: what works, by how much

The table ranks six families of tactics by the evidence behind them, with the effect sizes the literature reports. "Effect" refers to the outcome the studies actually measured, which is rarely "productivity".

Tactic	Evidence grade	What the studies measured	Size of effect	Load-bearing source
Protect sleep	A	Sustained attention, executive function after partial sleep restriction	Small-to-moderate deficits that accumulate night after night; the largest hits are on sustained attention A[13]	61-study meta-analysis
Batch communication	B	Daily stress when email was checked three times a day	Significantly lower stress in the limited-email week; output was not measured B[14]	124-person crossover trial
Remove interruption sources	B	Errors and stress under interruption	Errors doubled or tripled with 3–4 s interruptions; stress and effort higher even when speed was maintained B[8][9][10]	Lab experiments, n = 48 to 300
Take short breaks	A for wellbeing, C for performance	Vigour, fatigue, task performance after micro-breaks	Vigour d ≈ 0.36 and fatigue d ≈ 0.35; performance gains not reliable on demanding tasks A[15]	22-study meta-analysis
Exercise regularly	A	Cognition after exercise programmes	Small effect, g ≈ 0.2; larger for coordinative exercise and 45–60-minute sessions A[16]	80 RCTs
Meditate	B	Attention and other psychological outcomes	Medium average effect across outcomes (around r = 0.28), with attention among them; many underlying studies were small or low quality B[17]	163-study meta-analysis
Silence background music	A (for reading and memory)	Task performance with music playing	No overall benefit; small impairment for reading and memory, worse with lyrics A[18]	Meta-analysis of about 100 studies

Two additions that often appear in tip lists get a split verdict. Low-to-moderate caffeine (roughly 40–300 mg) reliably improves alertness, vigilance and reaction time, especially when you are tired, but does far less for memory or reasoning, and it costs sleep if taken late A^[19]. Nature exposure produces small benefits for working memory and cognitive flexibility across 42 studies B^[20]; the famous "20 % better memory after a walk in the park" comes from a 38-person experiment and should not be quoted as a general effect.

What attention training can and cannot do

Brain-training apps sell the idea that attention is a muscle. The evidence says the gains stay close to home. In a meta-analysis of 87 publications, working-memory training improved the trained tasks and produced small near-transfer gains, but when compared with active control groups it did not improve intelligence, verbal ability, decoding or arithmetic, the "far transfer" that would matter at work A^[21]. Diamond's review of executive functions reaches the same conclusion: they can be trained, but transfer is narrow and effortful, and the biggest everyday levers are the state variables, sleep, stress and exercise, rather than drills A^[3].

Meditation is the one training route with a moderate average effect, though the literature is heterogeneous and the attention-specific effects are not larger than its effects on anxiety or wellbeing B^[17]. If you enjoy it, it is a reasonable bet; it is not a shortcut around sleep.

What is myth

• "The average attention span is eight seconds, less than a goldfish." The figure traces to a Microsoft Canada marketing report with no primary data; there is no peer-reviewed measurement behind it A^[22]. Attention is task-dependent and cannot be summarised in one number for a species.
• "It takes 23 minutes to refocus." An interview figure, not a study result. The measured range is seconds in the lab to tens of minutes in field self-reports (see above).
• "Willpower depletes like a muscle." Preregistered replications found effects of d ≈ 0.04–0.06 ^[11][12].
• "Attention is a muscle you can grow with an app." Training transfers narrowly ^[21]; the anatomical papers cited for this claim are about networks, not training ^[1].
• "You can hold seven things in mind." Focal capacity is about four chunks when rehearsal is prevented ^[5].

Measuring your own attention honestly

Published averages describe populations, not you. Three measurements are cheap and informative:

1. Interruption log for two days. Note each interruption, its source and roughly how long it took you to get back into the task. You will get your own recovery range, which is more useful than any published number.
2. Longest uninterrupted block per day. Count only time with notifications off and no self-initiated switches. Most people are surprised how short it is.
3. A sleep check before blaming focus. If the preceding week averaged under seven hours a night, treat the attention problem as a sleep problem first; the restriction literature predicts exactly the deficit you are feeling, and it also predicts that you will have stopped noticing it ^[13].

A note on health: persistent trouble concentrating alongside low mood, exhaustion or sleep that never feels restorative is worth discussing with a clinician rather than treating as a productivity gap. This page describes research, not a diagnosis.

What to do next

Take the two or three tactics with A or B grades that fit your situation, for most people that means a sleep floor, one or two protected blocks with inputs closed, and communication batched into fixed windows, and run them for two weeks before adding anything else. The focus planner turns your meeting load, energy pattern and interruption sources into a week template with deep-work blocks placed where your own answers say they will survive. It uses the same evidence grades as this page.

Related reading in this cluster: the research behind context switching, how to improve deep work, what the evidence says about flow state, and how your chronotype shifts the best time to focus.

Frequently asked questions

How do I focus better at work when everything interrupts me?

Start with the sources you control: close email and chat during one or two protected blocks and check them in fixed windows. In a two-week crossover trial, people who checked email three times a day reported significantly less daily stress than when they checked freely ^[14]. Then log your own interruptions for two days to see which ones you can remove and which you can only shorten.

Is it true that attention spans are now eight seconds?

No. The figure comes from a marketing report with no primary data, and a critical review found no peer-reviewed support for it or for the related claim that attention collapses after ten minutes of a lecture ^[22]. The measured figure that does exist, about 47 seconds of attention to a single screen before switching, describes switching behaviour on screens, not the capacity of human attention ^[6].

Do brain-training apps improve concentration?

They improve the trained tasks and produce small gains on very similar tasks. Against active control groups, working-memory training did not improve intelligence, verbal ability or arithmetic in a meta-analysis of 87 publications ^[21]. Sleep, exercise and reducing stress have better evidence as everyday levers on executive function ^[3].

Does meditation help with focus?

A meta-analysis of 163 studies found a medium average effect of meditation across psychological outcomes, with attention among them, but many of the underlying studies were small or methodologically weak ^[17]. It is a reasonable practice if you enjoy it, with a B grade rather than a guarantee.

Does music help you concentrate?

On average, no. Across roughly 100 studies background music had no overall effect on task performance and slightly impaired reading and memory, with lyrics being worse than instrumental music ^[18]. If you use music to mask noise, instrumental and familiar is the lower-risk choice.

How much does one bad night of sleep affect focus?

Partial sleep restriction produces small-to-moderate deficits, largest for sustained attention and executive function, and the deficits accumulate across nights ^[13]. People also stop noticing the deficit after a few days, so a run of short nights is a stronger explanation for poor focus than most people assume.

Sources and evidence

1. The Attention System of the Human Brain — Posner, M. I.; Petersen, S. E. (1990), review A doi.org/10.1146/annurev.ne.13.030190.000325 (retrieved 2026-11-09)
2. The Attention System of the Human Brain: 20 Years After — Petersen, S. E.; Posner, M. I. (2012), review A doi.org/10.1146/annurev-neuro-062111-150525 (retrieved 2026-11-09)
3. Executive Functions — Diamond, A. (2013), review A doi.org/10.1146/annurev-psych-113011-143750 (retrieved 2026-11-09)
4. The Expected Value of Control: An Integrative Theory of Anterior Cingulate Cortex Function — Shenhav, A.; Botvinick, M. M.; Cohen, J. D. (2013), review B doi.org/10.1016/j.neuron.2013.07.007 (retrieved 2026-11-09)
5. The magical number 4 in short-term memory: A reconsideration of mental storage capacity — Cowan, N. (2001), review A doi.org/10.1017/S0140525X01003922 (retrieved 2026-11-09)
6. How to sharpen your attention and meet your goals (interview with Gloria Mark, UC Irvine) — University of California Newsroom (2024), report B www.universityofcalifornia.edu/news/how-sharpen-your-attention-and-mee (retrieved 2026-11-09)
7. 2025 Work Trend Index Annual Report: The year the Frontier Firm is born — Microsoft WorkLab (2025), report B www.microsoft.com/en-us/worklab/work-trend-index/2025-the-year-the-fro (retrieved 2026-11-09)
8. The cost of interrupted work: more speed and stress — Mark, G.; Gudith, D.; Klocke, U. (2008), conference paper B doi.org/10.1145/1357054.1357072 (retrieved 2026-11-09)
9. Momentary interruptions can derail the train of thought — Altmann, E. M.; Trafton, J. G.; Hambrick, D. Z. (2014), RCT B doi.org/10.1037/a0030986 (retrieved 2026-11-09)
10. The attentional cost of receiving a cell phone notification — Stothart, C.; Mitchum, A.; Yehnert, C. (2015), RCT B doi.org/10.1037/xhp0000100 (retrieved 2026-11-09)
11. A Multilab Preregistered Replication of the Ego-Depletion Effect — Hagger, M. S.; Chatzisarantis, N. L. D.; et al. (2016), RCT A doi.org/10.1177/1745691616652873 (retrieved 2026-11-09)
12. A Multisite Preregistered Paradigmatic Test of the Ego-Depletion Effect — Vohs, K. D.; Schmeichel, B. J.; Lohmann, S.; et al. (2021), RCT A doi.org/10.1177/0956797621989733 (retrieved 2026-11-09)
13. The neurocognitive consequences of sleep restriction: A meta-analytic review — Lowe, C. J.; Safati, A.; Hall, P. A. (2017), meta-analysis A doi.org/10.1016/j.neubiorev.2017.07.010 (retrieved 2026-11-09)
14. Checking email less frequently reduces stress — Kushlev, K.; Dunn, E. W. (2015), RCT B doi.org/10.1016/j.chb.2014.11.005 (retrieved 2026-11-09)
15. Give me a break! A systematic review and meta-analysis on the efficacy of micro-breaks for increasing well-being and performance — Albulescu, P.; Macsinga, I.; Rusu, A.; et al. (2022), meta-analysis A doi.org/10.1371/journal.pone.0272460 (retrieved 2026-11-09)
16. Systematic review and meta-analysis investigating moderators of long-term effects of exercise on cognition in healthy individuals — Ludyga, S.; Gerber, M.; Pühse, U.; Looser, V. N.; Kamijo, K. (2020), meta-analysis A doi.org/10.1038/s41562-020-0851-8 (retrieved 2026-11-09)
17. The psychological effects of meditation: A meta-analysis — Sedlmeier, P.; Eberth, J.; Schwarz, M.; et al. (2012), meta-analysis B doi.org/10.1037/a0028168 (retrieved 2026-11-09)
18. The impact of background music on adult listeners: A meta-analysis — Kämpfe, J.; Sedlmeier, P.; Renkewitz, F. (2011), meta-analysis A doi.org/10.1177/0305735610376261 (retrieved 2026-11-09)
19. A review of caffeine's effects on cognitive, physical and occupational performance — McLellan, T. M.; Caldwell, J. A.; Lieberman, H. R. (2016), review A doi.org/10.1016/j.neubiorev.2016.09.001 (retrieved 2026-11-09)
20. Attention Restoration Theory II: a systematic review to clarify attention processes affected by exposure to natural environments — Stevenson, M. P.; Schilhab, T.; Bentsen, P. (2018), systematic review B doi.org/10.1080/10937404.2018.1505571 (retrieved 2026-11-09)
21. Working Memory Training Does Not Improve Performance on Measures of Intelligence or Other Measures of "Far Transfer" — Melby-Lervåg, M.; Redick, T. S.; Hulme, C. (2016), meta-analysis A doi.org/10.1177/1745691616635612 (retrieved 2026-11-09)
22. Attention span during lectures: 8 seconds, 10 minutes, or more? — Bradbury, N. A. (2016), review A doi.org/10.1152/advan.00109.2016 (retrieved 2026-11-09)