Micro-Environments: How Your Physical and Digital Space Shapes Cognitive Performance (With a Design Checklist)

TL;DR. Attention is usually treated as a matter of discipline. A large body of research says it is also a matter of settings: the light, sound, temperature, air, objects and screens within two metres of you set the defaults your brain works against. The defaults are measurable. Office workers with windows slept 46 minutes more per night during the workweek than windowless colleagues (Boubekri and colleagues, 2014). Raising open-plan noise from 39 to 51 decibels reduced word memory and raised fatigue (Jahncke and colleagues, 2011), and performance on verbal tasks begins to fall once irrelevant speech becomes even slightly intelligible, at a speech transmission index of 0.21 (Haapakangas and colleagues, 2020). Office task performance peaks at around 22 degrees Celsius and is 8.9 percent lower at 30 degrees (Seppanen, Fisk and Lei, 2006). In a blinded exposure study, cognitive scores were 61 percent higher on a low-pollutant day and 101 percent higher on days with doubled ventilation than in a conventional office (Allen and colleagues, 2016). Across 22 studies, the mere presence of a smartphone costs a small but real amount of cognitive capacity, g = -0.14 overall and -0.23 for memory (Bottger, Poschik and Zierer, 2023). A notification you never open still disrupts an attention task about as much as answering it would (Stothart, Mitchum and Yehnert, 2015). None of these effects is large on its own, but all of them are cheap to change, and most people have never changed any of them. The CEO move is to treat your workspace as infrastructure with specifications. The student move is to run the 20-item checklist below, change the three cheapest failures, and measure for two weeks.

This article is part of the attention hygiene series that starts with the notification audit. That piece covers the signals that reach you. This one covers the room they reach you in, physical and digital, and what the evidence says each setting is worth.

Why does the micro-environment matter more than willpower?

Because the environment acts on every hour and willpower acts on a few. A desk that faces a window, a door that closes, a phone in another room and a notification tray that is empty do their work without being remembered. The evidence for most of these factors comes from three kinds of study: laboratory exposures where one variable changes at a time, field experiments in real offices, and large observational data sets. Each has limits, and the ledger below says which is which, because a 24-person blinded exposure study and a 302-person twelve-month field study do not deserve the same confidence.

Two principles organise what follows. First, effects stack. No single factor changes performance by more than a few percent in a well-run office, but a workspace with four or five settings wrong is a different place to think than one with none wrong. Second, the cheapest fixes are often the best evidenced. Moving a phone costs nothing; the meta-analysis behind it pools 43 effects.

Light: windows, daylight and the 500 lux rule

The clearest field result is small but concrete. Mohamed Boubekri and colleagues compared 27 office workers in windowless spaces with 22 whose desks had windows, using wrist actigraphy, a sleep questionnaire and a quality-of-life survey. Workers with windows slept an average of 46 minutes more per night during the workweek, and they scored better on vitality and on role limitations due to physical problems. It was a pilot study of 49 people, so the number is a pointer, not a law, but the direction matches the physiology: daytime light sets the circadian clock, and the clock sets sleep.

Mariana Figueiro and colleagues measured 109 workers in five buildings managed by the US General Services Administration for seven consecutive days in winter and summer. People who received high circadian-effective light in the morning fell asleep faster at night, especially in winter, and reported better sleep quality; high light exposure across the whole day was associated with lower depression scores. The spectrum of the light matters as well as the amount. Antoine Viola and colleagues gave 94 white-collar workers (104 were enrolled) two four-week periods under blue-enriched white light at 17,000 kelvin and under ordinary white light at 4,000 kelvin. Under the blue-enriched light, workers reported higher alertness, better mood, better concentration and performance, less evening fatigue and better sleep quality. Every outcome in that study was self-reported, and some of the effect could be expectation, which the authors tested and partly ruled out.

For the amount of light, the reference is the European workplace lighting standard EN 12464-1, whose table for offices sets a maintained illuminance of 500 lux for writing, typing, reading and data processing. The standard itself is paywalled; the figure here is taken from a lighting manufacturer’s reproduction of its table. A home desk lit by a single ceiling fixture may fall short of it, and a phone’s light meter app is enough to find out.

Noise: why speech is the problem and 55 decibels is the line

Noise research distinguishes between level and content, and content wins. Simon Banbury and Dianne Berry showed in three experiments in 1998 that office noise containing speech disrupted both memory for prose and mental arithmetic, that office noise without speech disrupted only arithmetic, and that the disruption did not depend on the meaning of the speech. A meaningless conversation in a language you do not know still costs you.

The mechanism was quantified by Annu Haapakangas and colleagues in 2020. Pooling 14 laboratory studies with 34 tests, they related performance to the speech transmission index (STI), a 0 to 1 measure of how intelligible background speech is. Performance begins to decline once STI exceeds 0.21 and reaches its maximum decrement at 0.44; verbal short-term memory deteriorates between 0.12 and 0.51. In plain terms, you do not need to be able to follow the conversation for it to hurt; you only need to half-hear it.

Helena Jahncke and colleagues tested a realistic manipulation in 2011: open-plan office noise at 39 versus 51 decibels, a 12-decibel increase. The louder condition reduced word memory, increased fatigue and reduced motivation. In a 2020 field study, Jahncke and David Hallman measured 64 workers in cell offices and 39 in shared or open-plan offices: noise was 15.3 decibels higher in the open-plan spaces (about 32 decibels in cell rooms, 49 in active zones), and serial recall was about 14 percent lower.

Moderate noise has one documented upside. Ravi Mehta, Rui Zhu and Amar Cheema found in five experiments that ambient noise at 70 decibels improved performance on creative tasks compared with 50 decibels, while 85 decibels hurt, apparently because moderate difficulty pushes people toward more abstract thinking. The finding is about idea generation, not memory or arithmetic, and it is why a cafe can be a good place to brainstorm and a bad place to proofread.

The open-plan office deserves its own line. Ethan Bernstein and Stephen Turban tracked two companies through a move to open floor plans using wearable sociometric badges and server records. In the first company, 52 participants reduced face-to-face interaction by 72 percent, from 5.8 to 1.7 hours per person per day, while emails sent rose 56 percent and instant messages rose 67 percent. In the second, 100 employees cut face-to-face interaction by 67 to 71 percent and email rose 22 to 50 percent. Open plans were sold on collaboration; the measured effect was the opposite.

For a target number, German workplace rules (the ASR A3.7 rule and VDI 2058 Part 3, reproduced in a DGUV report) set a maximum rating level of 55 dB(A) for predominantly mental work such as scientific work, software development and the drafting or translation of demanding texts, and 70 dB(A) for simple or routine office activities. A free sound-meter app will tell you which side of 55 your desk is on.

Temperature: the 22-degree peak and the gender split

Olli Seppanen, William Fisk and Quanhong Lei pooled objective office performance data for Lawrence Berkeley National Laboratory in 2006. Performance rises up to 21 to 22 degrees Celsius, peaks at about 22, falls above 23 to 24, and at 30 degrees is 91.1 percent of its maximum, a reduction of 8.9 percent. Li Lan, Pawel Wargocki and Zhiwei Lian found in 2011 that thermal discomfort from elevated temperature reduced performance on neurobehavioural, addition and typing tasks, with the optimum slightly below the neutral temperature.

The average hides a split. Tom Chang and Agne Kajackaite tested 543 students in Berlin at temperatures from 16 to 33 degrees. For each additional degree, women answered 1.76 percent more maths questions correctly (0.17 questions, p less than 0.001) and 1.03 percent more verbal items (p = 0.036); men showed small, non-significant declines (0.63 percent per degree on maths, p = 0.205, and 0.6 percent on verbal). Neither sex changed on a cognitive reflection task. The practical point is that a shared thermostat has no single right answer, and the 22-degree figure from pooled data is an average of two different curves.

Air: carbon dioxide, ventilation and what 1,000 ppm actually means

Usha Satish and colleagues exposed 22 participants to carbon dioxide at 600, 1,000 and 2,500 parts per million in three 2.5-hour sessions on one day and tested decision-making with a strategic management simulation. At 1,000 ppm, scores fell on six of nine scales; at 2,500 ppm the reductions were large on seven scales, with raw score ratios of 0.06 to 0.56, and five scales fell to marginal or dysfunctional levels. The sample was small, and the task is unusual, but the study changed how CO2 is discussed.

Joseph Allen and colleagues followed with a blinded exposure study of 24 knowledge workers over six full work days in a controlled office, varying ventilation and volatile organic compounds. Cognitive scores were 61 percent higher on the Green day (low VOCs) and 101 percent higher on the Green+ days (low VOCs with doubled ventilation, CO2 around 550 ppm) than on the Conventional day, with significant gains in all nine functional domains under Green+. The study measured a large relative change in a composite score; it does not mean your output doubles when you open a window.

The field evidence is smaller in magnitude and larger in sample. Jose Guillermo Cedeno Laurent and colleagues monitored 302 office workers in six countries (China, India, Mexico, Thailand, the United States and the United Kingdom) for twelve months with desk sensors and phone-based tests. Each interquartile-range rise in CO2 (315 ppm) was associated with 0.85 percent slower Stroop response times and 7.88 percent longer interference times after adjustment; each interquartile-range rise in fine particulates, PM2.5 (8.8 micrograms per cubic metre), with 0.82 percent slower response and 6.18 percent longer interference, and with 1.51 percent lower throughput on an arithmetic test. Small per-unit effects, but they accumulate across a working year, and PM2.5 adds an argument for an air purifier in cities.

One number needs correcting. The widely quoted “1,000 ppm limit” is not a cognitive threshold and is no longer in the US ventilation standard. Andrew Persily of the US National Institute of Standards and Technology documents that ASHRAE Standard 62 set 2,500 ppm in 1981, changed it to 1,000 ppm in 1989, and removed the limit in 1999; the related note that 700 ppm above outdoor air corresponds to about 7.5 litres per second per person was about body-odour acceptability for roughly 80 percent of unadapted visitors, not about thinking, and it was dropped in 2019. Use CO2 as an indicator that ventilation is low, which the studies above say matters, rather than as a line with a cognitive cliff behind it.

Plants and a view of green: 15 percent and 40 seconds

Marlon Nieuwenhuis and colleagues ran three field experiments in the Netherlands and the United Kingdom, two of them longitudinal, comparing “lean” offices stripped of decoration with the same spaces enriched with plants. Enriching a lean office with plants raised productivity by 15 percent, and in the third study tasks were completed faster without a rise in errors. The first study followed 153 employees in total (67 completed every measure) and the second 172 (81 complete).

The view may matter even in tiny doses. Kate Lee and colleagues gave 150 university students a 40-second break in the middle of a sustained attention task, looking either at a flowering green roof or at a bare concrete roof. The concrete group then made more omission errors (p = 0.041, r = 0.14) and showed more variable response times, while the green-roof group’s variability fell; participants also rated the green roof as more restorative (3.46 versus 2.93 on the scale used). The effect sizes are small, and the theory behind them, Stephen Kaplan’s 1995 attention restoration framework, holds that natural scenes let directed attention recover because they engage attention without demanding it. Forty seconds at a window with a tree in it is a cheap experiment.

Order and clutter: the trade-off nobody mentions

The popular claim that a tidy desk makes a tidy mind is half right. Kathleen Vohs, Joseph Redden and Ryan Rahinel ran three experiments in orderly and disorderly rooms. People in the orderly room chose healthier snacks, donated more and preferred the “classic” option; people in the disorderly room produced more creative ideas and preferred the “new” option. Order supports convention and self-control; disorder supports novelty. Which you want depends on the task, which is the same conclusion the noise research reached.

The cost of visual clutter is neural rather than moral. Stephanie McMains and Sabine Kastner showed with functional imaging that multiple objects in view suppress one another’s representation throughout visual cortex, because processing capacity is limited, and that attention has to resolve whatever competition the scene does not resolve on its own. A desk with fifteen objects in the field of view is fifteen small competitions your attention system settles continuously.

The phone on the desk: the best-evidenced fix of all

Adrian Ward, Kristen Duke, Ayelet Gneezy and Maarten Bos asked 520 undergraduates in one experiment and 275 in another to leave their own phone on the desk, in a pocket or bag, or in another room, then measured working memory capacity (operation span) and fluid intelligence (Raven’s matrices). Participants whose phones were in another room outperformed those whose phones were on the desk (p = 0.002), and the effect held in the second experiment on working memory. The phones were switched to silent with ring and vibration off; the cost came from the effort of not attending to them.

A 2023 meta-analysis by Tanja Bottger, Michael Poschik and Klaus Zierer pooled 22 studies with 43 effects. The overall “brain drain” effect was small and significant, Hedges’ g = -0.14 (95 percent confidence interval -0.24 to -0.03), with considerable heterogeneity. The effect was concentrated in memory tasks (g = -0.23, significant) and absent for attention (g = -0.07) and general cognitive tasks (g = 0.10). It was also regional: significant in Asian samples (g = -0.39) and not in European (g = -0.20) or North American samples (g = -0.03). Bill Thornton and colleagues had reported the same pattern in 2014: the mere presence of a cell phone reduced attention and task performance, especially on demanding tasks. The honest summary is that the effect is small, strongest for memory work, and costs nothing to remove. For what the phone’s rewards do and do not do to your brain, see the dopamine myth.

The digital micro-environment: notifications, switches and tabs

The screen is a room too, and its defaults are set by vendors rather than by you.

Notifications disrupt even when ignored. Cary Stothart, Ainsley Mitchum and Courtney Yehnert found that receiving a notification, without touching the phone, disrupted performance on an attention-demanding task by about as much as actively using the phone to call or text. Kostadin Kushlev, Jason Proulx and Elizabeth Dunn had 221 adults spend one week with alerts on and the phone within reach and one week with alerts off and the phone away: with alerts on, people reported more inattention and hyperactivity, and inattention predicted lower productivity and well-being. Nicholas Fitz and colleagues randomised 237 people to notifications as usual, batched three times a day, or switched off entirely. The batched group was more attentive and productive, in a better mood and felt more in control, with fewer interruptions; the no-notification group gained little and reported more anxiety and fear of missing out. Batching, not abstinence, is the evidence-backed setting. Martin Pielot and Luz Rello’s 24-hour no-notification study of 30 volunteers found the same trade: less distraction and more productivity, but some anxiety and a sense of being less connected; about two thirds meant to change their settings afterwards and half were still doing so two years later. The step-by-step method for this is the notification audit.

Switching is the default rhythm. Gloria Mark and colleagues logged 40 information workers at a large technology company for two working weeks. Mean focus on any screen was 47.0 seconds (median 40.2), workers switched between applications 272.7 times a day on average and within applications a further 131.9 times, and the mean focus on email was 61.8 seconds. Microsoft’s Work Trend Index special report of 17 June 2025, based on Microsoft 365 telemetry and a survey of 31,000 knowledge workers in 31 markets, counts 117 emails and 153 Teams messages per day for the average worker and an interruption every two minutes during core hours, 275 a day; the interruption figures describe the top 20 percent of users by ping volume over a 24-hour day, which the report states and most citations omit. What each switch costs, and why the famous “40 percent” figure does not come from switch costs, is covered in single-tasking versus multitasking, and what a switch leaves behind in attention residue.

Tabs are clutter with a search box. Joseph Chang and colleagues studied browser tab use through interviews with ten information workers over two weeks and a survey of 103 people, and found competing pressures: reasons to keep tabs open that range from practical to emotional costs, against the limited attention and resources that push toward closing them. The percentages that circulate from that study could not be verified for this article and are not repeated here. The visual-competition finding from McMains and Kastner applies to a tab bar as much as to a desk. For how screen exposure itself relates to cognition, see screen time and cognitive performance.

The Micro-Environment Evidence Ledger (CEOtudent editorial synthesis)

The table ranks each factor by two things the studies above allow: how strong the evidence is (sample size, design and replication) and how much it costs to change. The priority column is a CEOtudent editorial judgement combining the two; it is not a figure from any study.

Factor	Best-evidenced finding	Sample and design	Evidence strength	Cost to change	Priority
Phone location	Phone in another room beats phone on desk; meta-analytic g = -0.14, memory g = -0.23	22 studies, 43 effects (meta-analysis); 520 and 275 in the original experiments	Moderate to strong; small, heterogeneous effect	None	1
Notification settings	Batching three times a day improves attention, mood and control; off entirely raises anxiety	237 randomised; 221 within-subject	Moderate	None	2
Ventilation and CO2	Scores 61 and 101 percent higher under low pollutants and doubled ventilation; field: +0.85 percent response time per 315 ppm	24 blinded; 22 lab; 302 over twelve months in six countries	Moderate; large lab effects, small field effects	Low (window, CO2 monitor) to medium (purifier, HVAC)	3
Speech noise	Decline begins at STI 0.21; 12-decibel rise cut word memory; open plan 14 percent lower recall	14 studies, 34 tests pooled; 103 workers in the field	Strong for verbal tasks	Low (headphones, door) to high (room change)	4
Temperature	Peak near 22 degrees; 8.9 percent lower at 30; women gain 1.76 percent per degree on maths, men do not	Pooled objective data; 543 in a lab	Moderate; sex-specific	Low to medium	5
Daylight and light level	Windows: 46 more minutes of sleep per night; morning light shortens sleep onset; 500 lux for desk work	49 pilot; 109 in five buildings; 94 under blue-enriched light (self-report)	Moderate; small samples, some self-report	Low (desk position, lamp)	6
Plants and green view	15 percent productivity gain from plants; 40-second green view reduced errors (r = 0.14)	153 and 172 employees; 150 students	Moderate; small effects	Low	7
Order versus clutter	Order supports self-control and convention; disorder supports creativity; clutter competes for visual processing	3 experiments (n not available); imaging study	Moderate for direction, weak for magnitude	None	8

The Micro-Environment Design Checklist (CEOtudent editorial framework)

Twenty items, one point each. Physical and digital halves score separately out of 10. A score of 8 or more on either half means the defaults are working for you; 5 to 7 means a few afternoon fixes; below 5 means your attention is spending its day fighting the room. The checklist is an editorial instrument built from the studies above, not a validated scale, and the targets are the ones the sources give.

Physical (10 points)

No	Item	Target	Source
1	The phone is in another room during focused work, not on the desk or in a pocket	Another room	Ward and colleagues 2017; Bottger and colleagues 2023
2	Desk-level light is measured and adequate	500 lux at the work surface	EN 12464-1, as reproduced
3	The desk receives daylight, ideally in the morning	A window within view; morning light exposure	Boubekri 2014; Figueiro 2017
4	Background noise during focused work is measured	At or below 55 dB(A)	ASR A3.7 / VDI 2058 via DGUV
5	Intelligible speech from others is blocked or masked during verbal tasks	No half-heard conversations	Haapakangas 2020; Banbury and Berry 1998
6	Room temperature is controlled and set for the people in it	About 22 degrees Celsius; adjust for who is working	Seppanen 2006; Chang and Kajackaite 2019
7	CO2 is monitored and the room is aired when it climbs	Treat a rising reading as a ventilation signal, not a cliff	Satish 2012; Allen 2016; Persily 2021
8	Particulate pollution is filtered in urban or roadside rooms	Lower PM2.5	Cedeno Laurent 2021
9	At least one plant and a view of something green are within sight	Present	Nieuwenhuis 2014; Lee 2015
10	The field of view is cleared for focused work and allowed to be messy for idea work	Task-matched	Vohs 2013; McMains and Kastner 2011

Digital (10 points)

No	Item	Target	Source
11	Notifications are batched rather than continuous or fully off	Three deliveries a day	Fitz and colleagues 2019
12	Alerts are silenced during focused blocks	Off for the block	Kushlev 2016; Stothart 2015
13	Email is checked in set windows, not left open	Scheduled windows	Mark and colleagues 2016 (email focus 61.8 seconds)
14	Chat presence is set to do-not-disturb during focused blocks	On	Microsoft 2025 (153 messages a day)
15	Open tabs are limited and saved rather than kept	Closed or saved at the end of each block	Chang and colleagues 2021; McMains and Kastner 2011
16	One application is full-screen during deep work	Single window	Mark 2016 (272.7 switches a day)
17	Meetings are grouped so that focused blocks are not fragmented	Clustered meetings	Microsoft 2025; see the true cost of meetings
18	Screens are matched to task: the work screen carries no feeds	No social or news feed on the work device	Screen time evidence
19	A 40-second micro-break with a green view replaces a phone check	Window, not feed	Lee and colleagues 2015
20	The settings are reviewed on a calendar	Every quarter	Pielot and Rello 2017 (half kept their changes two years later)

Items 17 and 18 rest on the site’s own earlier reviews of the evidence on the true cost of meetings and on screen time rather than on a single study. If you want to measure what changing them does, the deep work deficit index gives a before-and-after number.

Where the evidence is weak

An honest ledger needs this section.

• Samples are small where effects are largest. The 61 and 101 percent cognitive gains come from 24 people; the CO2 decision-making study had 22; the windows study had 49. The large-sample field studies find effects of a few percent.
• Self-report inflates light results. Every outcome in the blue-enriched light trial was a questionnaire.
• Vendor telemetry is not peer reviewed. Microsoft’s 275 interruptions a day describes the top fifth of users over 24 hours.
• The phone effect is heterogeneous. Significant in Asian samples, not in European or North American ones, and absent for attention tasks. It is still the cheapest change on the list.
• Open plans trade noise for something. Bernstein and Turban measured less face-to-face contact, not less collaboration in every sense; the right question for a team is which it needs more.
• Temperature preferences differ by sex. A single thermostat setting optimised for the pooled curve is a compromise, and the study behind the split had one sample in one city.

The CEO move and the student move

A chief executive specifies the plant before hiring people to run it. The CEO move is to write specifications for your own workspace, physical and digital, using the targets in the checklist, and to treat a failed item the way an operations lead treats a machine running out of tolerance: not as a character flaw but as a setting. Most of the items cost nothing. The three that cost money (a lamp, a CO2 monitor, a purifier) cost less than a day of your time, which is what the research says they return over a working year. For a wider view of what drains and restores capacity during the day, see energy auditing.

The student move is to run the experiment on yourself, because the pooled numbers are averages of people who are not you. Score the checklist, change the three cheapest failures, and keep the same output measure for two weeks: words drafted, problems closed, pages read. The studies above suggest the gain will be modest and real. Your own log will tell you whether it is the phone, the noise or the air that matters most at your desk, and that is the number no study can give you.

Frequently asked questions

What single change has the best evidence?
Putting the phone in another room during focused work. A meta-analysis of 22 studies found a small, significant cost to having it nearby (g = -0.14), largest for memory tasks (g = -0.23), and the fix is free.

Is 1,000 ppm of CO2 dangerous for thinking?
It is not a safety threshold, and it was removed from the US ventilation standard in 1999. In one 22-person study, decision-making scores fell at 1,000 ppm on six of nine scales, and a 302-person field study found small slowdowns per 315 ppm rise. Treat a rising CO2 reading as a sign that ventilation is low and air the room.

Should I turn notifications off completely?
The randomised evidence says batch them instead. Delivery three times a day improved attention, mood and sense of control; switching them off entirely gained little and raised anxiety and fear of missing out.

Is a cafe a good place to work?
For idea generation, moderate noise around 70 decibels helped in five experiments; for memory, reading and arithmetic, intelligible speech hurts once it is even half-audible. Brainstorm in the cafe, proofread somewhere quiet.

How warm should the office be?
Pooled data put the peak near 22 degrees Celsius with an 8.9 percent loss at 30. Women in a 543-person study performed better as rooms got warmer and men slightly worse, so a shared room is a negotiation, not a lookup.

Sources

1. Boubekri M, Cheung IN, Reid KJ, Wang CH, Zee PC. Impact of windows and daylight exposure on overall health and sleep quality of office workers: A case-control pilot study. Journal of Clinical Sleep Medicine. 2014;10(6):603-611.
2. Viola AU, James LM, Schlangen LJM, Dijk DJ. Blue-enriched white light in the workplace improves self-reported alertness, performance and sleep quality. Scandinavian Journal of Work, Environment and Health. 2008;34(4):297-306.
3. Figueiro MG, Steverson B, Heerwagen J, Kampschroer K, Barrett C, Rea MS. The impact of daytime light exposures on sleep and mood in office workers. Sleep Health. 2017;3(3):204-215. Abstract.
4. EN 12464-1:2021. Light and lighting. Lighting of work places. Part 1: Indoor work places. Office requirements as reproduced in a lighting manufacturer’s standards reference.
5. Banbury S, Berry DC. Disruption of office-related tasks by speech and office noise. British Journal of Psychology. 1998;89(3):499-517. Abstract.
6. Haapakangas A, Hongisto V, Liebl A. The relation between the intelligibility of irrelevant speech and cognitive performance: A revised model based on laboratory studies. Indoor Air. 2020;30(6):1130-1146. Abstract.
7. Jahncke H, Hygge S, Halin N, Green AM, Dimberg K. Open-plan office noise: Cognitive performance and restoration. Journal of Environmental Psychology. 2011;31(4):373-382. As summarised in the first author’s doctoral thesis.
8. Jahncke H, Hallman DM. Objective measures of cognitive performance in activity based workplaces and traditional office types. Journal of Environmental Psychology. 2020;72:101503.
9. Mehta R, Zhu R, Cheema A. Is noise always bad? Exploring the effects of ambient noise on creative cognition. Journal of Consumer Research. 2012;39(4):784-799. Abstract.
10. Bernstein ES, Turban S. The impact of the “open” workspace on human collaboration. Philosophical Transactions of the Royal Society B. 2018;373(1753):20170239.
11. Hertwig R, Maue J, Neumann HD. Noise. In: Indoor Air, IFA Report. Deutsche Gesetzliche Unfallversicherung (DGUV), 2016. Reproducing VDI 2058 Part 3 and ASR A3.7 rating levels.
12. Seppanen O, Fisk WJ, Lei QH. Effect of temperature on task performance in office environment. Lawrence Berkeley National Laboratory, report LBNL-60946, 2006.
13. Lan L, Wargocki P, Lian Z. Quantitative measurement of productivity loss due to thermal discomfort. Energy and Buildings. 2011;43(5):1057-1062. Abstract.
14. Chang TY, Kajackaite A. Battle for the thermostat: Gender and the effect of temperature on cognitive performance. PLOS ONE. 2019;14(5):e0216362.
15. Satish U, Mendell MJ, Shekhar K, Hotchi T, Sullivan D, Streufert S, Fisk WJ. Is CO2 an indoor pollutant? Direct effects of low-to-moderate CO2 concentrations on human decision-making performance. Environmental Health Perspectives. 2012;120(12):1671-1677.
16. Allen JG, MacNaughton P, Satish U, Santanam S, Vallarino J, Spengler JD. Associations of cognitive function scores with carbon dioxide, ventilation, and volatile organic compound exposures in office workers. Environmental Health Perspectives. 2016;124(6):805-812.
17. Cedeno Laurent JG, MacNaughton P, Jones E, Young AS, Bliss M, Flanigan S, Vallarino J, Chen LJ, Cao X, Allen JG. Associations between acute exposures to PM2.5 and carbon dioxide indoors and cognitive function in office workers: A multicountry longitudinal prospective observational study. Environmental Research Letters. 2021;16(9):094047.
18. Persily AK. Indoor Carbon Dioxide Metric Analysis Tool. NIST Technical Note 2213, 2022; and Persily A. Please don’t blame Standard 62.1 for 1000 ppm CO2. ASHRAE Journal. 2021;63(2).
19. Nieuwenhuis M, Knight C, Postmes T, Haslam SA. The relative benefits of green versus lean office space: Three field experiments. Journal of Experimental Psychology: Applied. 2014;20(3):199-214.
20. Lee KE, Williams KJH, Sargent LD, Williams NSG, Johnson KA. 40-second green roof views sustain attention: The role of micro-breaks in attention restoration. Journal of Environmental Psychology. 2015;42:182-189.
21. Kaplan S. The restorative benefits of nature: Toward an integrative framework. Journal of Environmental Psychology. 1995;15(3):169-182. Abstract.
22. Vohs KD, Redden JP, Rahinel R. Physical order produces healthy choices, generosity, and conventionality, whereas disorder produces creativity. Psychological Science. 2013;24(9):1860-1867. Abstract.
23. McMains S, Kastner S. Interactions of top-down and bottom-up mechanisms in human visual cortex. Journal of Neuroscience. 2011;31(2):587-597.
24. Ward AF, Duke K, Gneezy A, Bos MW. Brain drain: The mere presence of one’s own smartphone reduces available cognitive capacity. Journal of the Association for Consumer Research. 2017;2(2):140-154.
25. Bottger T, Poschik M, Zierer K. Does the brain drain effect really exist? A meta-analysis. Behavioral Sciences. 2023;13(9):751.
26. Thornton B, Faires A, Robbins M, Rollins E. The mere presence of a cell phone may be distracting: Implications for attention and task performance. Social Psychology. 2014;45(6):479-488. Abstract.
27. Stothart C, Mitchum A, Yehnert C. The attentional cost of receiving a cell phone notification. Journal of Experimental Psychology: Human Perception and Performance. 2015;41(4):893-897. Abstract.
28. Kushlev K, Proulx J, Dunn EW. “Silence your phones”: Smartphone notifications increase inattention and hyperactivity symptoms. Proceedings of CHI 2016:1011-1020. Abstract.
29. Fitz N, Kushlev K, Jagannathan R, Lewis T, Paliwal D, Ariely D. Batching smartphone notifications can improve well-being. Computers in Human Behavior. 2019;101:84-94. Abstract.
30. Pielot M, Rello L. Productive, anxious, lonely: 24 hours without push notifications. Proceedings of MobileHCI 2017.
31. Mark G, Iqbal ST, Czerwinski M, Johns P, Sano A. Neurotics can’t focus: An in situ study of online multitasking in the workplace. Proceedings of CHI 2016:1739-1744.
32. Microsoft. Breaking down the infinite workday. Work Trend Index Special Report, 17 June 2025.
33. Chang JC, Hahn N, Kim Y, Coupland J, Breneisen B, Kim HS, Hwong J, Kittur A. When the tab comes due: Challenges in the cost structure of browser tab usage. Proceedings of CHI 2021. Abstract.

The Evidence Ledger, its priority ranking and the Design Checklist are CEOtudent analyses built on the sources above. All other figures are reported as printed in those sources; sources read in abstract form only are marked.

---

This content was compiled with the support of AI following in-depth research, then written and prepared for publication by the CEOtudent editorial team.

This post is also available in: