Say you’re writing a report. You open a reference to check a detail, go back to the draft, then open your inbox without needing anything from it. That’s three screen changes, but they didn’t all do the same job.
The reference helped you carry on with the report. The inbox visit took you somewhere else. If you only count the changes, you miss the difference, and the difference is what helps you decide what to keep and what to change.
This module is for adults who work at a computer and can put off checks they choose to make without letting urgent duties slip.
What you’ll get from this module
In this module
By the end of this module, you’ll know which screen changes help your work and which checks you might change. You’ll choose a situation from your daily life and try an exercise on paper. You’ll compare the work you finish and the messages you need to answer, as well as the time between switches. Workplace observations and studies of going without email, attention training and feedback show what these choices rest on.
Ask Jing
Become a Jingius.
What changes when the screen changes?
Workplace observation: what a short interval measures.
Gloria Mark and colleagues’ 2016 study recorded the computer activity of forty volunteers at one large US high-tech workplace for about twelve working days. They looked at how long people stayed with one thing on screen before switching.
For each person, they took each day’s recorded computer time and divided it by the number of screen switches. Then they averaged those daily results across the person’s workdays. The average time people stayed with one screen was 47.0 seconds.
How the software recorded matters. It tracked the active window, the one the person was using. It stopped counting time in that window when the person changed windows, or when five minutes passed with no keyboard or mouse activity. So the result describes screen use, recorded under those rules, in that one workplace.
Why it matters: That average isn’t the longest time someone can pay attention. A window change tells you the screen changed. To understand the work, you also need to know what the person was doing. The number can describe a pattern without putting a limit on your attention.
Read the workplace study of online screen-focus duration.
Developer field study: different activities can serve one task.
André N. Meyer and colleagues’ 2017 study looked at records from twenty professional software developers at four companies, over eleven full workdays on average. They studied what the developers did and how they switched in ordinary work.
The researchers told an activity apart from the task it served. A developer could move from writing code to looking up documentation and still be on the same task. Those were changes in activity, not always a move away from the work that mattered.
Think back to the report. Opening a reference you need helps you finish it. Cutting out that switch would remove a useful part of the work, just because it happened on another screen.
Why it matters: This study sorted what developers did into activities. The previous study only timed active windows. The difference matters: before you judge a switch, ask what work it served.
Read the field study of developers’ activities and switches.
Workplace intervention: what happened without new email.
Some switches may still be worth changing. In 2012, Gloria J. Mark, Stephen Voida and Armand V. Cardello studied thirteen volunteers at one US scientific research organisation.
The plan was three ordinary days to start with, then five workdays with no new email coming in and no work email going out. Some of the data collection ran past that plan because of absences or scheduling problems.
The average time in one window rose from 75.5 seconds on the ordinary days to 131.9 seconds without email. People stayed in each window longer before switching while they had no email.
Everyone did their usual work with email first, then without new email. They had all volunteered, and their workload could have changed too. The longer stretches give you a reason to try changing how you check email where you can. They don’t settle whether other changes helped produce the result.
Why it matters: For you, the useful question is whether changing when you check email helps you get work done. A short pause is not the same as the study’s five workdays without new email. Judge it by what you get done and whether you still answer the messages you have to, as well as by the time between switches.
Read the workplace study of work without email.
Controlled training study: practice changed some task behavior.
Changing when you check email is one option. Practising attention is another. In 2012, David M. Levy and colleagues tested women who worked in human resources on several tasks in a stressful, made-up office.
The study compared three groups. One had mindfulness meditation training, one had relaxation training, and one waited and got meditation training later. The course lasted eight weeks, with two hours with an instructor each week, plus homework. Thirty-nine of the fifty-seven people chosen finished the study. People weren’t put into groups wholly by chance.
After meditation training, people switched between activities less often. The first meditation group also spent longer on each activity. The waiting group later spent longer too, but that rise wasn’t statistically significant: the evidence for it was less clear.
Overall, the training made people no faster or more accurate, beyond the gains that come from simply taking the test again.
Structured practice may be worth a look if you want to change how you move between activities. But switching less and finishing better work are different results. In this study, switching less did not come with any proven gain in speed or accuracy from the training.
Why it matters: These findings came after an eight-week course, in a made-up office. They don’t tell us that a quick breathing tip on an ordinary workday would do the same.
Read the study of mindfulness training and multitasking.
Randomized field experiment: feedback helped while people used it.
Maria Wirzberger and colleagues’ 2024 experiment studied university students and graduates doing computer work of their own choosing over twelve days. Both groups used software to name their tasks and the programs that went with them. One group also got feedback based on what they did.
Of 170 people put into groups by chance, 127 gave data through every stage of the study. The feedback group got a higher focus score than the group without feedback. The software worked out that score. It added points for focused behaviour and took points for distracted behaviour. It didn’t count finished work.
The researchers also asked whether feedback helped people’s scores rise faster over time. Their main test found no statistically significant difference between the groups, once they allowed for how many tests they ran.
Further analyses, which looked for other patterns in the data, found some hopeful changes. But it stayed unclear whether people improved for good without the software. And the study did not show that feedback helped people get further with their goals.
Why it matters: These results pull apart two useful questions. Does support help while you use it? Does it leave you with a skill that lasts without it? The first can matter even while the second is still open, as long as you judge the work as well as the score.
Read the field experiment on behavioral feedback and focus.
What this means for you
Start with the work you want to finish. Sort the screen changes that help you do it from the checking that could wait. The studies offer different options, measured in different ways. They don’t give you one ideal time between switches for your day.
Name the task before judging a switch
Say in ordinary words what finishing would look like. In the made-up report example, that could be finishing a section with its supporting details checked. Now you have a purpose to judge each move against.
Opening a reference you need still serves that purpose. Opening the inbox out of curiosity may take you away from it. Finishing the section and moving to another task is a third kind of change: you’re done.
In your notes, keep screen changes and task changes apart. For each switch you notice, mark why it happened. You finished, needed something for the same task, answered something from outside, chose to check, or don’t know.
For a check you chose yourself, ask whether it could wait without letting a duty slip. That’s what a check you choose means in this module.
Write “unknown” when you can’t tell why you switched. Even if nothing outside prompted you, you may not know what led you to check. You’re looking for something you could change, not marking every move as a failure.
Observe comparable sessions and keep missing information visible
Pick a session length that fits your life. Watch two sessions on different occasions. Keep the kind of task, the time of day and the messages you have to answer as alike as you can. When the conditions are alike, the differences are easier to read.
Use paper and an ordinary clock. Write down the start time. Then make a short note of each change you notice in the active window, document or browser tab. Keep the notes brief, because writing them can interrupt the work too.
Here, an interval is the time between two screen changes. You can count an interval as complete only when you saw both the switch that started it and the switch that ended it. If you started watching partway through, mark that interval as partial. If the session ends before the next switch, the last interval is partial too.
A planned break or the end of the session doesn’t count as a switch. Mark any switches you missed. Leave out of your sums any interval with a missing start or end. That keeps guessed times out of your summary.
If your clock only shows minutes, don’t report times in seconds. Compare sessions you watched in the same way. A paper log won’t catch the detail the research software did, so what you find isn’t a research benchmark for your attention.
Understand the summary before using it
For each session on its own, count the complete intervals. To find the mean, add up how long they lasted and divide by how many there are. To find the median, put them in order of length and take the middle one. If there’s an even number, take the average of the middle two.
Say your complete intervals are 30, 60 and 90 seconds. Together they make 180 seconds. Divide by three and the mean is 60 seconds. The middle value is also 60 seconds, so that is the median.
If you also saw only the last 20 seconds of an earlier interval, keep it marked partial and leave it out. You don’t know how long it really lasted. If there are no complete intervals, write “no complete intervals”, not zero.
Next to the times, note the checks that took you away from the task and could have waited. Note what useful work you finished, and any problems with messages you had to answer. The times describe the pattern. The work tells you what you got done. A longer interval isn’t progress if important work or messages suffered.
Change discretionary email checking while preserving obligations
Before you try a pause, make sure your email can wait, and agree how someone can reach you if it’s urgent. Never hold back a reply you have to give, or let an urgent duty slip, for this exercise. If answering email all the time is part of your job, choose another situation or stop here.
In two more sessions like the first ones, leave the email checks you choose to make until a point you pick in the task, or an agreed time. Keep up the messages you have to answer the whole time. Notice whether some other kind of checking takes email’s place. If it does, count that change when you judge whether the pause helped your work.
For each session, compare the times, the checks that took you away, the work you finished and any problems with messages. Keep the pause, change it or drop it, depending on how well you could work and keep up with your duties. Stop if your duties suffer.
If the sessions were very different, try again in more similar conditions before you judge the change. A small trial like this can’t prove the pause made the difference. You may still get useful work done even if the intervals stay the same.
Choose training or assistance for the outcome you want
Structured training is an option if you want to practise and have time for the course and the homework. The made-up office study gives a reason to look into it, but it leaves open what it would do for your everyday work. The study didn’t test whether a quick version would do the same.
Feedback is a different kind of help. It may support you while you work, even though it hasn’t been shown to leave you with a lasting skill. Judge what you get done, and how good it is, apart from the score on the screen.
You don’t need to buy a system to start understanding your own switching. The paper exercise below helps you look at the task and one change to your checking first.
How about your daily life?
Where do you move between things you need and checks that could wait? You might be writing a report, reviewing documents or studying at your computer. Choose a situation where it’s clear what finished looks like, and where people can still reach you if it’s urgent.
List the situations you’d like to understand better. One good idea is enough. Add more if they help you choose.
| A situation in my daily life | What I would like to understand or improve |
|---|---|
| ____________________ | ____________________ |
| ____________________ | ____________________ |
Which situation would be most useful to look at now?
My chosen situation: ____________________
You can change your choice. Take it into the exercise below. Choosing a situation doesn’t yet mean it’s right to put off your email.
Try a manageable checking change
Start with your chosen situation at the computer. If you came straight here, choose one task with a clear point where it’s finished, and email checks you choose to make that can wait without letting a duty slip.
Never hold back a reply you have to give, or let an urgent duty slip. Before a pause, agree how someone can reach you if it’s urgent. Keep up the messages you have to answer, and stop if your duties suffer. If you can’t meet these conditions, choose another situation or leave the pause untried.
- Define the work and the observation. Write down the task, what finished looks like and a session length that fits your life. Choose two sessions on different occasions, with the same kind of task, time of day and messages to answer. Use paper and an ordinary clock, and note whether it shows seconds or only minutes.
Task, completion, sessions and clock precision: ____________________ - Observe before changing checking. Write down the start time and each change you notice in the active window, document or browser tab. Note whether it’s for the same task or a new one. Mark why it happened: you finished, needed something for the same task, answered something from outside, chose a check that could wait, or don’t know.
Use the headings below and add rows as you need them. Keep the notes short and private. Mark the time you spent writing notes, and any switches you missed. Don’t make up times for them.
| Time | Window/content | Same task or new task | Reason | Complete/partial/missed | Useful output / communication issue |
|---|---|---|---|---|---|
| ______ | ______ | ______ | ______ | ______ | ______ |
- Summarize each session separately. Use only complete intervals, where you saw both the start and the end. Keep partial first or last intervals in the log, but leave them out of the sums, along with any interval missing its start or end. A planned break or the end of the session isn’t a switch you saw.
Count the complete intervals. Mean = total complete-interval time ÷ count. For the median, put the intervals in order and take the middle one, or the average of the middle two when there’s an even number. With no complete intervals, say so instead of writing zero. Keep to what your clock can really show.
Also note the checks that took you away and could have waited, the useful work you finished and any problems with messages. Don’t lump unlike sessions together, or compare yourself with a research benchmark.
Session summary: ____________________ - Try the checking pause in two further comparable sessions. Hold off the email checks you choose to make until the point you picked in the task, or the agreed time. Keep up the messages you have to answer and the agreed way to reach you if it’s urgent. Note the same things as before, and whether some other checking takes email’s place.
Endpoint or check time, urgent route and observations: ____________________ - Decide from the work and obligations. Compare the session summaries and the useful work you finished. Keep the pause, change it or drop it, depending on how well you could work and keep up with your duties. Stop if your duties suffer. Try again if the sessions were very different. A small trial like this can’t prove the pause made the difference. You may still get useful work done even if the intervals stay the same.
Decision, reason and any unresolved comparison: ____________________
Ask Jing
Become a Jingius.
Your next step
You’ve finished when you can say what your notes show, what you still don’t know, and whether the checking change is worth keeping, changing or dropping. Leaving the pause untried because your messages can’t wait is a clear result too.
A note on the evidence
The studies measured different things: time in active windows, kinds of activity, performance in a made-up office and a score from software. They show a lot of switching in the places studied, and some changes in behaviour. They don’t show that people’s ability to pay attention has got worse over time.
Everyone in the small email study used email first and then went without new email. The training study involved a long course. The feedback study lasted twelve days, and some people didn’t give data all the way through. Keep those differences in mind when you choose what to try, and judge your real work and the messages you have to answer.
How we prepared this module
This module links workplace observations with a study of going without email, a training study and a feedback experiment. A made-up report example explains switches that serve the task, and made-up times show how the sums work. The written exercise helps you look at similar sessions and one manageable change in your checking.
Sources
- Gloria Mark, Shamsi Iqbal, Mary Czerwinski, Paul Johns, and Akane Sano. (2016-05). Neurotics Can’t Focus: An in situ Study of Online Multitasking in the Workplace. ACM CHI 2016, pp. 1739–1744. Author-hosted camera-ready version; Method, PDF pp. 2–3; Table 1 and Results, PDF p. 3.
- André N. Meyer, Laura E. Barton, Gail C. Murphy, Thomas Zimmermann, and Thomas Fritz. (2017-12; online 2017-01-22). The Work Life of Developers: Activities, Switches and Perceived Productivity. IEEE Transactions on Software Engineering, 43(12), pp. 1178–1193. Published paper; §3.1, p. 1181; §4, p. 1182; §5.2, p. 1186.
- Gloria J. Mark, Stephen Voida, and Armand V. Cardello. (2012-05). “A Pace Not Dictated by Electrons”: An Empirical Study of Work Without Email. ACM CHI 2012, pp. 555–564. Author PDF; Method, PDF pp. 2–3; Multitasking and Table 2, PDF pp. 4–5; Limitations, PDF pp. 8–9.
- David M. Levy, Jacob O. Wobbrock, Alfred W. Kaszniak, and Marilyn Ostergren. (2012-05-28/30). The Effects of Mindfulness Meditation Training on Multitasking in a High-Stress Information Environment. Graphics Interface 2012, pp. 45–52; CHCCS/SCDHM. Author PDF; §§3.1–3.3, pp. 46–47; §§4.1.1–4.1.3, pp. 48–49; §§5.2 and 5.5, p. 50.
- Maria Wirzberger, Anastasia Lado, Mike Prentice, Ivan Oreshnikov, Jean-Claude Passy, Adrian Stock, and Falk Lieder. (2024-02-07). Optimal feedback improves behavioral focus during self-regulated computer-based work. Scientific Reports, 14, article 3124. Published article; Field experiment, pp. 7–9; Table 2, p. 10; Feedback-induced learning, p. 11; Limitations, p. 12.
