The Science of Focus — Attention in the Age of Distraction

Attention is the filter that decides which fraction of the world reaching your senses gets processed and which gets ignored. Cognitive scientists have spent decades showing that this filter is not one thing but several systems, each with limits that no productivity trick removes. Some popular claims about distraction hold up well under testing. Others, including a few repeated in most articles on the topic, have failed replication. This piece separates the two and explains what the evidence says about training focus.

Educational content, not medical advice. Talk to your doctor before changing any medication, treatment, or routine. Information here reflects current research but every person's situation differs.

Key takeaways

• Attention is at least three systems (alerting, orienting, and executive control), each tied to different brain regions and chemical messengers.
• Switching between demanding tasks has real costs in time, errors, and stress, although the famous "23 minutes to refocus" figure is routinely misdescribed.
• Several well-known claims, including the "brain drain" effect of a nearby phone and willpower as a depletable fuel, have not held up in large replications.
• Small effects from sleep, exercise, and mindfulness are better supported than any single app or technique.
• Changing the environment is usually more reliable than relying on effort in the moment.

What attention is, and why researchers split it into parts

Everyday speech treats "focus" as a single dial. Laboratory work suggests something closer to a control room with several operators. One keeps you awake and ready. Another points your senses toward a location or object. A third resolves conflicts when two goals compete, such as reading a report while a colleague's conversation drifts across the room.

Psychologists usually describe four overlapping ideas. Selective attention is choosing one input and suppressing others, the skill you use to follow one voice in a crowded restaurant. Sustained attention is holding that choice over time, which most people mean when they say "focus." Divided attention is splitting resources across two streams, which works for a practiced, automatic activity but breaks down when both tasks need deliberate thought. Executive attention is top-down control, including the ability to resist a tempting distraction and return to the goal.

These are not separate organs. They draw on shared circuits, mainly in the frontal and parietal lobes (the front and upper-rear regions of the cerebral cortex), plus deeper structures that regulate arousal. The useful point for readers is that "bad focus" can have several different causes. A person who cannot start a task, a person who loses the thread after five minutes, and a person who is pulled away by every sound are showing different failures.

Why the spotlight analogy is useful, and where it breaks

The classic picture of attention is a spotlight: whatever it lights up gets detailed processing, and everything outside is dim. The metaphor captures the narrowness of conscious focus, and it explains why a person staring at a screen can fail to hear a question from across the table.

It is incomplete, though. A spotlight is passive, while attention is constantly steered by goals and by surprises. A better image is a stage crew with a director. The director (executive control) decides what the show needs, the crew (orienting) moves the light, and a house manager (alerting) makes sure the theater is open and staffed in the first place. When any of those three falters, the show looks bad for different reasons.

The three attention networks inside your head

The most influential framework in the field comes from Michael Posner and Steven Petersen, who proposed in 1990 that attention relies on three anatomically distinct networks. Their 2012 review in the Annual Review of Neuroscience, titled "The Attention System of the Human Brain: 20 Years After," updated the model with two decades of imaging work. The table below summarizes it as described in that review.

Network	What it does	Main regions	Chemical messenger linked to it
Alerting	Achieves and maintains readiness and vigilance	Brainstem arousal systems, plus right-hemisphere frontal and parietal regions	Norepinephrine, released from the locus coeruleus
Orienting	Selects information by location or sense	Dorsal system (frontal eye fields, intraparietal sulcus and superior parietal lobe) and ventral system (temporoparietal junction, ventral frontal cortex)	Acetylcholine
Executive control	Monitors conflict, maintains task goals, switches between tasks	Cingulo-opercular network (anterior cingulate, anterior insula) and a frontoparietal network	No single messenger established; dopamine and serotonin gene variants show conflicting associations

Source: Petersen and Posner, 2012, PMC3413263.

Why the executive network matters most for deep work

The 2012 review split executive control into two cooperating systems. The cingulo-opercular network provides steady maintenance of a task across minutes. The frontoparietal network handles starting a task, switching, and making rapid adjustments within a trial. Put simply, one system is the steady hand and the other is the quick correction.

This split fits ordinary experience. Many people can sit through a long task once started but struggle with the first five minutes, while others begin easily and drift. The same label, "poor focus," covers both. It also explains why different interventions help different people.

Neuromodulators in plain terms

Neuromodulators are chemicals that shift how a whole circuit behaves, rather than carrying one specific message the way a typical neurotransmitter does. Norepinephrine acts like a volume knob on general alertness. Acetylcholine seems to sharpen how strongly a selected location or stimulus is processed. The review notes that drugs affecting acetylcholine change orienting but not alerting, a pattern that suggests the networks are chemically distinguishable.

For readers curious about the wider chemistry, how neurotransmitters work explains the messaging system these modulators sit on top of. The role of dopamine in motivation is a separate story, covered in dopamine and motivation, and it is often oversimplified in popular writing on attention.

Matching a lapse to a system

The framework gives a vocabulary for everyday failures. The mapping below is an interpretive aid built on the model described above, not a clinical test, and real lapses usually involve more than one system.

What it feels like	Which system the model points to	What the research suggests about the lever
Drowsy, slow to react, "foggy"	Alerting	Sleep and arousal matter most; the network is tied to norepinephrine
Eyes on the screen but pulled toward every sound or alert	Orienting	Reduce competing cues so the spotlight has fewer reasons to move
Cannot get started, or lose the goal mid-task	Executive control (frontoparietal)	Cueing and clear next steps lower switching cost, per task-switching studies
Can start but cannot stay on a long task	Executive control (cingulo-opercular)	Fewer mid-task interruptions; breaks at natural stopping points

One reason this table is useful is that it explains why advice conflicts. A person whose real problem is poor sleep gains nothing from a notification blocker, and a person surrounded by alerts gains little from an extra hour in bed. Neither piece of advice is wrong. It is aimed at a different system.

What your brain does when you "multitask"

For tasks that demand deliberate thought, the brain does not run two streams in parallel. It alternates, paying a cost each time it changes rule sets. Researchers call this task switching, and the cost shows up as slower responses and more errors immediately after a switch.

A well-cited set of experiments by Joshua Rubinstein, David Meyer, and Jeffrey Evans, published in 2001 in the Journal of Experimental Psychology: Human Perception and Performance, tested participants who switched between tasks. Switching costs increased when the rules were more complex and shrank when a cue announced the upcoming task. The American Psychological Association's coverage of the work stressed the same pattern: time lost rises with the complexity and unfamiliarity of the tasks (APA Monitor, October 2001).

Two analogies help. Switching tasks is like closing one program and opening another on an old computer: the machine has to unload one set of instructions and load the next, and the delay grows if the new program is large. A second picture is a chef moving between a sauce that needs stirring and a dough that needs kneading. Each move is quick, but the chef never does both at once, and a burned sauce is the price of a long detour.

The practical conclusion is narrow but robust. Interleaving two demanding tasks reduces speed and accuracy compared with doing them one after another. The size of the penalty depends on task difficulty, practice, and cueing, so no single percentage applies to everyone.

The "23 minutes" figure, corrected

Few statistics in productivity writing are repeated as often as the claim that it takes 23 minutes to refocus after an interruption. It comes from field research by Gloria Mark and colleagues at the University of California, Irvine. In an interview with Gallup, Mark described that interrupted work was resumed the same day in 81.9 percent of cases, and that it took on average 23 minutes and 15 seconds to return to the original task (Gallup Business Journal).

Three details are routinely dropped. First, the interval covers returning to the original work, and in the study that often included handling about two other tasks in between. It does not measure how long it takes for full concentration to come back. Second, the underlying workplace observation involved a small number of information workers at one company. Third, a summary of the earlier conference paper reports a slightly different same-day average of 25 minutes and 26 seconds, which suggests the figure is better treated as "tens of minutes in a real office" than as a precise constant.

Interruptions finish work faster, at a price

One counterintuitive result comes from a 2008 paper by Gloria Mark, Daniela Gudith, and Ulrich Klocke, presented at the ACM CHI conference. Participants wrote emails in a simulated office, with a "supervisor" interrupting at intervals. Interrupted people completed their tasks in less time with no drop in quality. They paid for it in stress, frustration, time pressure, and effort (Mark, Gudith, and Klocke, CHI 2008).

This matters because it shows why interruptions feel productive. People compensate by working faster, and the output looks fine. The cost lands on the person, not on the document. Personality also mattered: openness to experience and a low need for personal structure predicted smaller disruption costs.

Attention residue

A related idea comes from Sophie Leroy, a management researcher at the University of Washington Bothell. In a 2009 paper in Organizational Behavior and Human Decision Processes, she described attention residue: when people switch away from an unfinished task, part of their thinking stays with it, which impairs performance on the next task. The idea has an intuitive basis, like a song that keeps playing in your head while you try to read.

Residue appears to be greater when the first task was left unfinished. This is one reason why researchers and writers often suggest stopping a session at a defined point rather than mid-sentence, and why a quick note about "where I was" before switching is a common coping habit. The research supports the mechanism; evidence for any particular note-taking ritual is thinner.

Heavy media multitaskers and the replication problem

In 2009, Eyal Ophir, Clifford Nass, and Anthony Wagner reported in PNAS that people who habitually used several media streams at once performed worse on a task-switching test, which the authors attributed to reduced ability to filter out irrelevant information. The finding was widely reported as evidence that chronic multitaskers have worse attention.

The result has aged unevenly. A 2017 paper by Wiradhany and Nieuwenstein noted that subsequent studies did not all show an association between media multitasking and distractibility, casting doubt on the original finding's reliability. The sample in the original study was also small, with 19 heavy and 22 light media multitaskers. The direction of causation is unclear too: people who are easily pulled toward new information might simply choose more media streams.

Supertaskers exist, but rarely

Not everyone slows down. In a 2010 study in Psychonomic Bulletin & Review, Jason Watson and David Strayer tested 200 participants who drove in a simulator while doing a demanding memory task. About 2.5 percent, which is five people, showed no decline on either task. The authors called them supertaskers. The result suggests real individual variation, but it also shows how rare the pattern is, and nobody can reliably tell from self-assessment whether they belong to that group.

What the multitasking evidence supports

Claim	Evidence	Verdict
Switching between demanding tasks costs time and accuracy	Laboratory task-switching studies, including Rubinstein, Meyer, and Evans (2001)	Well supported; size varies by task
It takes 23 minutes to refocus after any interruption	Field observation of resumption time, small sample	Misdescribed; measures return to work, not refocusing
Interruptions raise stress even when output holds up	Mark, Gudith, and Klocke (2008)	Supported in a simulated office
Chronic media multitasking damages attention	Ophir, Nass, and Wagner (2009), with later mixed results	Unsettled
Some people are immune	Watson and Strayer (2010)	Real but rare, around 2.5% in one study

For a lighter-weight companion read, the myth of multitasking walks through everyday examples, and how working memory limits thinking explains the capacity bottleneck that makes switching expensive.

Mind wandering, the quiet competitor

Interruptions come from outside. Mind wandering comes from inside, and it is more common than most people assume. In a 2010 paper in Science, Matthew Killingsworth and Daniel Gilbert used an iPhone app that asked about 2,250 adults at random moments whether their minds were wandering. Mind wandering occurred in 46.9 percent of the samples, and in at least 30 percent of samples during every activity except one. People also reported being less happy when their minds were wandering, and time-lag analyses suggested wandering tended to precede the unhappiness rather than follow it (Killingsworth and Gilbert, 2010).

Two cautions apply. The 46.9 percent figure describes sampled moments, not a share of waking hours, although press coverage often phrases it the second way. And mind wandering is not purely harmful: it is linked to planning and creative problem solving, and its brain basis overlaps with the circuitry described in the default mode network. The goal of focus training is not to eliminate wandering but to notice it sooner.

What attention-grabbing design does, and what is not proven

Many apps are built to maximize time spent. Features such as unpredictable rewards, endless feeds, and notifications are designed to bring users back. The psychological principles behind them are well established in older research on reward schedules, and readers who want the mechanism can look at infinite scroll psychology and how social media rewires the brain.

The causal claims are weaker than the design intent. Evidence that heavy app use permanently changes attention capacity is limited, and much of the data is correlational. Heavy users may differ from light users before they pick up a phone. What is measurable is narrower: notifications and easy access interrupt tasks, and interruptions cost time and add stress, as the studies above show.

The "brain drain" claim and why it deserves caution

Perhaps the most repeated claim in this area is that the mere presence of a smartphone on your desk, even switched off, reduces cognitive capacity. It comes from a 2017 study by Adrian Ward and colleagues in the Journal of the Association for Consumer Research, which reported that participants whose phones were in another room outperformed those whose phones were on the desk.

Later attempts to reproduce the effect have mostly come back empty. A pre-registered direct replication reported in Acta Psychologica in 2022 found no difference between location conditions on either task. A separate partial replication and extension by Hartmann and colleagues reported no evidence for the effect, while still advising against keeping phones on the desk. A 2025 conference paper described the original finding as seemingly real but noted that most replication attempts produced null results (FORRT Replication Atlas).

The honest summary is that the effect is not established. There are still sensible reasons to keep a phone out of sight, since a buzzing screen interrupts, but "its mere presence drains your brain" should not be stated as settled science.

Open offices

The workplace version of the distraction debate is the open-plan office. The best-known field study is by Ethan Bernstein and Stephen Turban, published in 2018 in Philosophical Transactions of the Royal Society B. Employees at two Fortune 500 companies who moved from cubicles to open plans wore sociometric badges that recorded face-to-face interaction. Interaction fell by about 70 percent in both cases, and electronic messaging rose. The authors read this as employees withdrawing from officemates (Bernstein and Turban, 2018).

Note what the study measured. It tracked collaboration patterns, not focus or output directly. It is evidence that open offices change behavior in ways opposite to what advocates promised, not a direct measurement of concentration. Individual preferences also vary widely.

How to read a focus study

Attention research produces appealing headlines, so it helps to check a few things before believing one.

Sample size. The Ophir study compared 19 heavy and 22 light media multitaskers, and the mindfulness trial randomized 48 students. Small samples are noisy, and effects found in them often shrink when larger groups are tested. That is exactly what happened to the phone-presence effect.

Laboratory versus real life. A task-switching experiment isolates one variable, which is useful for science and poor for predicting your Tuesday. Field studies, such as Mark's office observations, are more realistic but cannot control everything. The two kinds of evidence complement each other, and agreement between them is more persuasive than either alone.

Correlation or experiment. Surveys linking heavy device use to poor attention cannot say which came first. People who struggle to concentrate may reach for their phones more, not the other way around. Randomized designs, where researchers assign the condition, answer the causal question better, which is why the replications of the phone and depletion effects carried so much weight.

Effect size. A result can be statistically reliable and still small. The Chang meta-analysis of single exercise sessions is the clearest example: the overall effect was small, even though many studies pointed the same way. A small effect for a free, safe habit can still be worth having, but it should not be sold as transformation.

Who funds the claim. Products that promise to restore focus have an interest in the headline. This is not a reason to dismiss them, only a reason to look for independent trials and to notice whether a company's own studies measured something beyond its own test.

Applying these checks to the claims above explains the verdicts in the tables: the task-switching cost passes most of them, the 23-minute figure fails on description, and the brain-drain claim fails on replication.

Willpower, depletion, and what replicated

A common idea holds that self-control is a limited fuel that runs down with use, a theory called ego depletion. It shaped advice such as "do your hardest work first, before willpower runs out." The idea was tested in a 2016 registered replication report in Perspectives on Psychological Science led by Martin Hagger. Twenty-three laboratories with a combined 2,141 participants ran a standardized protocol, and the pooled effect was d = 0.04, with a confidence interval from −0.07 to 0.15 that includes zero (Hagger et al., 2016).

A commentary by Jianning Dang in Frontiers in Psychology has argued that the depleting task used may not have been strong enough (Dang, 2016), so the debate is not closed. Still, the practical lesson is modest. Feeling mentally tired is real, and so is the benefit of structuring your day around when you tend to work best. The specific mechanism of a depleting fuel tank has weak support. Readers interested in the related research can start with the science of willpower.

What the evidence supports for training attention

Evidence-based options are less dramatic than the products that promise to "rewire" your brain. The table sorts common strategies by what the research actually shows.

Strategy	Best available evidence	Realistic expectation
Mindfulness practice	Mrazek et al. (2013): 48 students, two-week course, small randomized trial	Possible gains in working memory and reading scores; small sample
Aerobic or other exercise	Chang et al. (2012) meta-analysis: small average benefit for one session	Small, short-lived boost
Adequate sleep	Large body of sleep-restriction research	Strong, consistent effect on attention
Reducing interruptions	Mark's field and lab work	Less stress and less switching
Pomodoro-style timed blocks	Popular method; the interval lengths are conventions rather than trial-derived thresholds	Useful if it helps you start; not a proven formula

Mindfulness

In the 2013 study by Michael Mrazek and colleagues, published in Psychological Science, 48 college students were randomly assigned to a mindfulness class or a nutrition class. Each met for 45 minutes, four times a week, for two weeks. The mindfulness group improved on the verbal section of the GRE (a graduate admissions test) and on a working-memory test, and mind-wandered less during testing, while the nutrition group showed none of these changes. The authors estimated a gain equivalent to about 16 percentile points on the GRE verbal section (UC Santa Barbara news release).

Because the sample was small and the participants were students, the result is a promising signal rather than a settled effect. Also, the mind-wandering explanation held for the participants who were prone to distraction at the start. A fuller account of the brain research is in what happens to your brain when you meditate.

Exercise

Chang, Labban, Gapin, and Etnier published a meta-analysis in Brain Research in 2012 pooling 79 studies of acute exercise (a single session) and cognition. The average effect was small, with an overall effect size of g = 0.097. Moderators such as duration, intensity, and fitness mattered. A 2025 meta-review of 30 meta-analyses by Chang and colleagues in Psychological Bulletin reported a larger, small-to-medium pooled effect of 0.33, strongest when cognition was tested right after exercise.

In practice this means a walk before a hard task may help a little, and the benefit appears to fade. Evidence for long-term habits is covered in the science of exercise.

Sleep

Attention is one of the functions most sensitive to short sleep, and the alerting network in particular is about arousal and readiness. Sleep is not a tactic to add to a list but a baseline everything else rests on. Two companion pieces go further: the science of sleep covers how much people need, and what happens to the body without sleep for 24 to 72 hours covers how performance degrades.

Learning and attention

Attention is also the gateway to memory. Information that you do not attend to rarely gets encoded, which is why divided attention in a lecture or a meeting leaves little behind. For the learning side of the story, see how learning actually works, and for a practical lens on removing digital triggers, digital minimalism and how to do a digital detox.

Environment beats effort

A recurring theme across this research is that situations steer behavior more reliably than intentions do. That is consistent with how attention works. Executive control is a limited resource that is best spent on the task itself, not on resisting a buzzing phone. Design changes that remove the decision are cheaper than daily acts of restraint.

Common environmental changes include silencing nonessential notifications, closing unrelated tabs, working in a space associated with one kind of task, and batching email into set windows. The studies cited here support the underlying mechanisms (fewer switches mean lower switching costs and less interruption stress). They do not prove that any one arrangement is best, and individuals differ in what they tolerate.

An illustrative day

Consider a writer drafting a report. In version one, the phone buzzes with messages, a chat window pings, and email is open in the next tab. Each alert is a small orienting event that pulls the spotlight away, and the executive network has to drag it back. In version two, notifications are muted for 90 minutes, the draft is the only open window, and the first pass is done before messages are read. Nothing about the writer's brain changed. The number of switches did. Per the work of Mark and colleagues, fewer interruptions should reduce stress even where output is similar.

This is an illustration, not a measured case, and a 90-minute block is a convention. Some people sustain longer stretches and others shorter, and the right length depends on the task and the person.

When attention problems are something else

Not all difficulty with focus comes from habits. ADHD (attention-deficit/hyperactivity disorder) is a neurodevelopmental condition involving persistent differences in attention regulation and, in many people, impulsivity or hyperactivity. The Centers for Disease Control and Prevention reports that an estimated 7 million U.S. children, or 11.7 percent, had a current ADHD diagnosis in 2024, based on a national parent survey (CDC). The CDC also notes that estimates vary by source and measure. For the biology behind the diagnosis, see ADHD neuroscience.

The distinction matters. A person with ADHD and a person overwhelmed by notifications may report similar symptoms, but the first reflects a lifelong pattern that a clinician can evaluate, and the second reflects an environment. Anxiety, depression, thyroid problems, sleep disorders, and medication side effects can also impair concentration. For that reason, persistent or distressing attention difficulty is a question for a qualified professional, not a self-diagnosis based on an article. Broader background on mood and cognition is in the science of mental health.

Gut, mood, and attention

Research on the gut-brain axis (the two-way signaling between the digestive system and the brain) has grown, and it is relevant to mood and stress. Evidence specifically linking gut bacteria to attention in healthy adults is far thinner, so claims of that kind should be treated as early. The overview in the gut-brain connection separates the established from the speculative.

Common misconceptions

"Multitasking is a skill you can train away." The cost of switching between demanding tasks appears to be a feature of the system, not a habit, though practice reduces it for familiar tasks. A few rare individuals show no decline, but self-assessment does not identify them.

"Everyone's attention span has collapsed to eight seconds." This statistic circulates widely, but this article does not rely on it, and a single number for "attention span" is not meaningful because attention depends on the task, interest, and fatigue. The 47-second figure from Gloria Mark's research is real but narrower than it sounds: it is the average time people in her logging studies spent on one screen before switching, not a measure of how long anyone can concentrate.

"My phone only distracts me when it buzzes." Notifications interrupt, which is documented. Whether a silent phone on the desk adds a cognitive cost is contested, as the failed replications show.

"More willpower is the answer." The depletion model has failed a large replication, and beyond that, relying on effort in the moment is a weaker strategy than removing the cue.

"Open offices make teams more collaborative." In the Bernstein and Turban study, face-to-face interaction dropped and messaging rose.

"Young people are digital natives who multitask better." This article found no evidence in the sources reviewed to support that claim, and the task-switching research does not show an exemption by age.

Frequently asked questions

Q: Is attention one ability or several? A: Several. The Posner and Petersen model describes alerting, orienting, and executive control networks with different brain regions and chemical modulators. Everyday words such as focus or concentration blend them together, which is why a single fix rarely solves every problem.

Q: Does multitasking make you less productive? A: For demanding tasks, usually. Task-switching studies show costs in speed and accuracy that grow with task complexity. The costs shrink for practiced, automatic activities. Interrupted people in one study finished work faster but reported more stress and frustration.

Q: How long does it take to refocus after an interruption? A: The widely quoted 23 minutes and 15 seconds measures how long it took workers to resume the original task in field observation, and often included other work in between. It is not a measured time for regaining concentration, and results varied across reports.

Q: Does having my phone nearby reduce my intelligence? A: This was claimed in a 2017 study, but pre-registered replications have mostly failed to find the effect. A phone can still interrupt through notifications. Evidence that a silent, unused phone drains cognition is not established.

Q: Can attention be trained? A: Some evidence suggests so, with modest effects. A small randomized trial of a two-week mindfulness course found gains in working memory and reading test scores, and meta-analyses find small cognitive boosts after exercise. Sleep has the strongest and most consistent effect on attention.

Q: How is ADHD different from ordinary distraction? A: ADHD is a clinically diagnosed neurodevelopmental condition with a persistent pattern, whereas distraction from notifications and habits is situational. Symptoms can overlap, and other conditions can mimic both, so evaluation belongs with a qualified professional.

Q: Is mind wandering bad? A: Not always. One study found minds wandered in 46.9 percent of sampled moments and that people reported lower happiness during wandering. Mind wandering is also associated with planning and creativity, so the aim of attention training is awareness rather than elimination.

The takeaway

Attention is a set of cooperating systems, each with limits, which is why no single trick fixes "poor focus." The best-supported findings are modest and practical: switching between hard tasks costs time and adds stress, interruptions are expensive even when output looks fine, and sleep, exercise, and mindfulness help a little each. Several famous claims, including the 23-minute rule as usually told, the brain-drain effect of a nearby phone, and willpower as a depletable fuel, are weaker than their reputations. The most useful reading of the evidence is that you should spend your limited executive control on the work itself and let the environment handle the temptations.