Deep work, in Cal Newport's phrase, means focusing without distraction on hard thinking. It comes from a popular book, not a study, but real research sits underneath it: the cost of switching tasks, the attention residue an unfinished task leaves, what interruptions do to stress, and how often minds wander. The famous 8-second attention span, shorter than a goldfish's, sits under none of them.
EdCitation publishes this guide. We read every study in full in September 2026 unless we say we read only the abstract, and we give each one's sample, method, finding and limit. EdCitation's Find sources is how we traced the goldfish claim, and the focus routine towards the end is our own design, drawn from these findings rather than tested as a package.
Phones, the Pomodoro technique and if-then plans are covered in how to focus and stop procrastinating and phone and social media while studying, so this guide leaves them out.
What is deep work, and is it backed by research?
Deep work is a term from Newport's 2016 book Deep Work (Grand Central Publishing), a popular argument, not a study. Its publisher's page defines deep work as "the ability to focus without distraction on cognitively demanding tasks" and describes the book as cultural criticism mixed with advice, organised around four rules: work deeply, embrace boredom, quit social media, and drain the shallows (Newport, 2016). The rest of this guide weighs its territory study by study.
Is multitasking possible? What task-switching studies found
Doing two demanding tasks at once is not really possible: what feels like multitasking is rapid switching, and every switch costs time.
The switch cost in the laboratory
Rubinstein et al. (2001) ran four experiments with 108 University of Michigan undergraduates, who sorted cards or solved arithmetic problems, repeating one rule or alternating between two. In the first experiment, alternating added almost a second to each response on average (975 ms). The cost grew when the rules were more complex and shrank when a cue said which task came next.
Monsell (2003) reviewed the task-switching research: responses are slower, and usually less accurate, just after a switch. Knowing the next task in advance reduces the cost but does not remove it, and even a block that mixes tasks stays slower than one task done throughout.
The limit: these are split-second costs on simple laboratory tasks. They show that switching is never free, not how many minutes a student loses to it.
Attention residue: the unfinished task that follows you
Attention residue is the part of your mind still thinking about the last task after you have moved to the next. Leroy (2009) studied it in Organizational Behavior and Human Decision Processes. According to the abstract, two experiments found that people struggled to shift attention away from an unfinished task and did worse on the next one. Finishing the first task was not enough on its own: finishing it under time pressure helped people let go of it and perform better on the next. We read the abstract only (the paper is paywalled), so we cannot give its samples or tasks.
Leroy and Glomb (2018) then tested a remedy across four studies. When people expected to return to an interrupted task under time pressure, residue built up and their work on the interrupting task suffered. A brief "ready-to-resume" plan, noting where they stopped and how they would restart, reduced the residue and protected performance. Again, we read the abstract only.
The limit: both papers used workplace-style tasks in the laboratory, and neither measured studying or marks.
What happens when you are interrupted?
Interrupted people often work faster, not slower, and pay for it in stress. Mark et al. (2008) tested 48 people, 81% of them German university students with an average age of 26, on a simulated office task: answering 12 emails as a human resources manager. In two of three conditions a "supervisor" interrupted every two minutes by phone or instant message.
| Condition | Minutes to finish (average) | Stress (1 to 20) |
|---|---|---|
| No interruptions | 22.77 | 6.92 |
| Interrupted, same topic | 20.31 | 9.46 |
| Interrupted, different topic | 20.60 | 9.13 |
Errors did not differ, emails were shorter, and interrupted people also rated frustration, time pressure and effort higher. The authors' own limits: a laboratory task of about 20 minutes per condition, and no way to say whether people would cope or tire over longer spells.
Where "23 minutes to refocus" comes from
The popular claim that it takes 23 minutes to refocus after an interruption appears in neither of the two papers by Mark and colleagues that we read. The nearest figure is from Mark et al. (2005), who shadowed 24 managers, analysts and software developers for over 700 hours in all. Of their work on a topic, 57% was interrupted. When interrupted work was resumed the same day (77.2% of the time), the average gap was 25 minutes 26 seconds (standard deviation 54 minutes), and people had worked on more than two other topics before returning.
That figure measures how long work sat untouched, not how long refocusing took, and it came from 24 workers at one outsourcing company, not from students.
Does your mind wander when you study?
Probably about half the time, if studying is like the rest of daily life. Killingsworth and Gilbert (2010) built an iPhone app that asked people at random moments what they were doing, how they felt, and whether they were thinking about something else. Among 2,250 adults (mean age 34, 73.9% in the United States), minds wandered in 46.9% of samples. People were less happy when their minds wandered to neutral or unpleasant topics, and no happier when they wandered to pleasant ones; a time-lag analysis suggested the wandering usually came before the low mood.
The limits: the sample was iPhone users of a happiness-tracking web app, the paper is a one-page report in Science, and it measured happiness, not learning. It does show that a wandering mind is normal.
What is a flow state, and can studying produce it?
Flow is the absorbed state in which a demanding task seems to run itself. Nakamura and Csikszentmihalyi (2001), in a chapter co-written by the psychologist who led the first studies, trace it to interviews with chess players, rock climbers, dancers and surgeons, later followed by pagers that sampled everyday experience at random. Two conditions come first: a challenge that stretches your skills without overwhelming them, and clear, near goals with immediate feedback. Too much challenge brings anxiety; too little brings boredom.
Studying can produce flow when those conditions hold. Peifer et al. (2022) reviewed 252 flow studies published from 2000 to 2016. Most of the studies on performance linked flow with doing better, and flow went with engagement in the classroom, but the reviewers warn that studies defined and measured flow in different ways, and that most designs cannot show whether flow causes good work or good work produces flow.
Do humans have an 8-second attention span?
No study we could find shows that humans have an 8-second attention span, or that goldfish have a 9-second one. The figures travelled widely in 2015, and their trail runs out at a statistics website.
Where the figures came from
Microsoft Canada (2015) published a marketing report on attention, built on an online survey of 2,000 Canadians and brain recordings (EEG) of 112. One slide gives the average human attention span as 12 seconds in 2000 and 8 seconds in 2013, and a goldfish's as 9 seconds, citing Statistic Brain rather than Microsoft's own research. We read a copy hosted on another site; we could not find it on Microsoft's.
An archived copy of the Statistic Brain page, from March 2015, gives the same three figures. It lists as sources the US National Library of Medicine's National Center for Biotechnology Information and the Associated Press, with a "research date" of 1 January 2014, but no study, author or link (Statistic Brain, n.d.).
What the people who checked found
Maybin (2017), for the BBC's statistics programme More or Less, asked both listed sources: neither could find any research behind the figures, and attempts to reach Statistic Brain came to nothing. An Open University attention researcher told the programme that attention depends on the task, making an average span close to meaningless, and a fish behaviour researcher said goldfish are a standard animal for studying memory. Thirkettle and Pike (2015), writing from the Open University, noted that the report gave no detail beyond the figures it borrowed.
A related claim, that students' attention collapses 10 to 15 minutes into a lecture, fares little better. Bradbury (2016) reviewed the literature and found that the source most often cited barely discusses attention, that many studies had flawed or subjective methods, and that the data do not support a fixed limit. We read the abstract only.
What we could not confirm: whether any study ever measured a 12-second or 8-second human attention span, and what Statistic Brain based its figures on.
What does the evidence say, claim by claim?
| Claim | What the evidence says | Strength | Source |
|---|---|---|---|
| Multitasking is really fast switching | Every switch costs time; preparing reduces the cost but does not remove it | Many laboratory experiments | Rubinstein et al. (2001); Monsell (2003) |
| An unfinished task drags on the next | Attention residue lowered performance on the next task | Two experiments, workplace-style tasks | Leroy (2009) |
| A note on where you stopped helps | A ready-to-resume plan reduced residue | Four studies, one team | Leroy and Glomb (2018) |
| Interruptions slow you down | Interrupted people finished faster but felt more stress | One laboratory experiment, 48 people | Mark et al. (2008) |
| It takes 23 minutes to refocus | Not found; interrupted work waited about 25 minutes before resuming | One observation study, 24 workers | Mark et al. (2005) |
| Minds wander a lot | In 46.9% of random samples | One large sampling study | Killingsworth and Gilbert (2010) |
| Flow improves performance | Linked in most studies; cause unclear | Review of 252 studies | Peifer et al. (2022) |
| Humans have an 8-second attention span | No study found behind it | None | Maybin (2017) |
| Checking a claim like this yourself | Search the claim in the published record | Our method | EdCitation's Find sources |
How can you concentrate for longer? A focus routine
You concentrate for longer by protecting one task from switches, interruptions and loose ends. We designed this routine ourselves from the research in this guide; nobody has trialled it in one piece, and the block length is our suggestion, not a finding.
- Choose one task and write its finish line. "Draft section 2 to 400 words" is a near goal with feedback, which is where flow starts.
- Close the loop on whatever you are leaving. Write two lines on where you stopped and what comes next, as in the ready-to-resume plan.
- Batch the small tasks. Put messages and admin into one block later, so you switch once rather than twenty times.
- Guard a block of 60 to 90 minutes. Tell people, and follow the phone settings in our phone guide.
- Set the difficulty just above comfortable. Bored: raise the goal. Anxious: break it smaller.
- When your mind wanders, note the thought and return. Wandering is normal.
- Park source hunting. Mark a gap "[source?]" and fill it in the break with Find sources, building the reference in Cite a source, so searching does not break the writing block.
- Stop with a note for next time, so tomorrow's session starts without residue.
If poor concentration persists or worries you, talk to your university's support services or a doctor. This guide is study technique, not medical advice.
How do you check a claim like the goldfish attention span?
Search the claim in the published record. EdCitation is the best tool for that job, because it looks sources up in the record and never writes an answer of its own, so it cannot hand you the kind of untraceable figure this guide has just followed.
Find sources, run on the goldfish claim
Find sources looks through about 300 million published works. You can search a topic or paste the claim a sentence needs to support, then narrow by kind of source, years or open access. When we ran it on 27 September 2026, "human attention span is shorter than a goldfish" returned 3,503,998 works from Crossref. First was Thirkettle and Pike's 2015 piece, "Is technology making your attention span shorter than a goldfish's?": a commentary on The Conversation, not a peer-reviewed study, so check the kind of source before you cite it.
"Attention span 8 seconds lectures" returned 22,077 works from OpenAlex, with Bradbury's 2016 review first, cited 490 times. Second was a 2020 record with the same title under another author and journal; we did not open it, but it shows why you check the author, journal and year before citing.
Cite a source and Verify references
Hand Cite a source an ISBN, a DOI, a PubMed ID, a web address or just a title, and back comes the reference in whichever of APA 7, Vancouver, MLA 9, Harvard, IEEE or Chicago 18 author-date you need, screened for retraction. Paste a finished list into Verify references and every entry is marked verified, doubtful or not found, with any retraction flagged. On this guide's 15 references it returned 14 verified, after catching one year we had wrong: the flow chapter's publisher dates it 2001, not 2002. It left the Microsoft report, a PDF, for us to check by hand. You need no account for any of the three, and they cost nothing; Pro and Max are set out on Pricing.
EdCitation never writes a word of your work. It finds the sources, cites them and checks them, so every sentence you hand in is your own and every claim in it can be traced.
Quick questions
What is attention residue?
Attention residue is thinking that stays on the previous task after you switch. Leroy (2009) found it lowered performance on the next task.
Is multitasking possible?
For demanding tasks, no. People switch rapidly between them, and experiments find each switch costs time, even when you know the next task in advance.
Is the 8-second attention span real?
No study has been found behind it. It came from Statistic Brain via a 2015 Microsoft Canada report, and the sources Statistic Brain listed told the BBC they had no record of it.
How long does it take to refocus after an interruption?
No study we read supports the popular 23 minutes. Mark et al. (2005) found interrupted work waited about 25 minutes before people returned to it, which is a different thing.
How can I check whether a study claim is real?
Type the claim into EdCitation's Find sources, open the studies that come back, and check each one before you cite it.
References
- Bradbury, N. A. (2016). Attention span during lectures: 8 seconds, 10 minutes, or more? Advances in Physiology Education, 40(4), 509–513. https://doi.org/10.1152/advan.00109.2016
- Killingsworth, M. A., & Gilbert, D. T. (2010). A wandering mind is an unhappy mind. Science, 330(6006), 932. https://doi.org/10.1126/science.1192439
- Leroy, S. (2009). Why is it so hard to do my work? The challenge of attention residue when switching between work tasks. Organizational Behavior and Human Decision Processes, 109(2), 168–181. https://doi.org/10.1016/j.obhdp.2009.04.002
- Leroy, S., & Glomb, T. M. (2018). Tasks interrupted: How anticipating time pressure on resumption of an interrupted task causes attention residue and low performance on interrupting tasks and how a "ready-to-resume" plan mitigates the effects. Organization Science, 29(3), 380–397. https://doi.org/10.1287/orsc.2017.1184
- Mark, G., Gonzalez, V. M., & Harris, J. (2005). No task left behind? Examining the nature of fragmented work. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 321–330). Association for Computing Machinery. https://doi.org/10.1145/1054972.1055017
- Mark, G., Gudith, D., & Klocke, U. (2008). The cost of interrupted work: More speed and stress. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 107–110). Association for Computing Machinery. https://doi.org/10.1145/1357054.1357072
- Maybin, S. (2017, March 10). Busting the attention span myth. BBC News. https://www.bbc.co.uk/news/health-38896790
- Microsoft Canada. (2015). Attention spans [Consumer insights report]. https://static1.squarespace.com/static/59bf2bf68fd4d28e59627113/t/5ab01bdff950b70546a08167/1521490918597/265348695-Microsoft-Attention-Spans-Research-Report.pdf
- Monsell, S. (2003). Task switching. Trends in Cognitive Sciences, 7(3), 134–140. https://doi.org/10.1016/S1364-6613(03)00028-7
- Nakamura, J., & Csikszentmihalyi, M. (2001). The concept of flow. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology (pp. 89–105). Oxford University Press. https://doi.org/10.1093/oso/9780195135336.003.0007
- Newport, C. (2016). Deep work: Rules for focused success in a distracted world. Grand Central Publishing. https://www.hachettebookgroup.com/titles/cal-newport/deep-work/9781455586691/
- Peifer, C., Wolters, G., Harmat, L., Heutte, J., Tan, J., Freire, T., Tavares, D., Fonte, C., Andersen, F. O., van den Hout, J., Šimleša, M., Pola, L., Ceja, L., & Triberti, S. (2022). A scoping review of flow research. Frontiers in Psychology, 13, Article 815665. https://doi.org/10.3389/fpsyg.2022.815665
- Rubinstein, J. S., Meyer, D. E., & Evans, J. E. (2001). Executive control of cognitive processes in task switching. Journal of Experimental Psychology: Human Perception and Performance, 27(4), 763–797. https://doi.org/10.1037/0096-1523.27.4.763
- Statistic Brain. (n.d.). Attention span statistics. Archived 4 March 2015 by the Internet Archive. http://web.archive.org/web/20150304024952/http://www.statisticbrain.com:80/attention-span-statistics/
- Thirkettle, M., & Pike, G. (2015, May 28). Is technology making your attention span shorter than a goldfish's? The Conversation. https://theconversation.com/is-technology-making-your-attention-span-shorter-than-a-goldfishs-42072