“Depth is becoming increasingly rare and increasingly valuable in our economy.”
— Cal Newport, Deep Work, 2016
The Open-Plan Hypothesis — What the Design Was Supposed to Do
The open-plan office was designed around a specific theory of organizational value: that spontaneous, informal interaction between employees generates creative cross-pollination and collaborative problem-solving that structured meetings cannot replicate. The theory has roots in mid-century industrial design — Frank Lloyd Wright's Larkin Administration Building (1906) was among the first large-scale open office designs — and became dominant in American workplace design during the 1990s and 2000s as technology companies promoted it as a visible expression of collaborative culture.
The argument for open-plan design was intuitive and compelling. If employees can see and hear each other constantly, they will encounter unexpected ideas, form serendipitous connections, and collaborate in real time rather than waiting for scheduled meetings. The walls between offices were barriers to collaboration. Removing them would release the collaborative potential that hierarchical, siloed structures suppressed. The design change was presented as simultaneously pro-employee — more egalitarian, more social, more dynamic — and pro-organization. By 2017, approximately 70% of American offices had some form of open or semi-open floor plan.
The premise of the design had a testable empirical prediction: that face-to-face interaction between employees would increase after open-plan adoption. This prediction was testable. In 2018, researchers tested it.
The Harvard Study — What Actually Happened
Ethan Bernstein and Stephen Turban at Harvard Business School published a study in 2018 in Philosophical Transactions of the Royal Society B examining actual interaction patterns before and after open-plan office transitions at two large American companies. They used sociometric badges and electronic communication logs to track interaction patterns with precision that earlier studies had lacked. The results inverted the hypothesis.
Face-to-face interaction decreased by approximately 70% after companies transitioned to open-plan layouts. The workers who moved into open-plan environments did not interact more in person — they interacted substantially less. Electronic communication (email, instant messaging) increased to compensate. The researchers' interpretation: when exposed to constant ambient social presence and continuous potential interaction, workers developed coping behaviors to protect their ability to work. They donned headphones. They avoided eye contact. They turned to screens as a signal of unavailability. The open physical environment produced a closed behavioral environment.
The finding was counterintuitive and uncomfortable for organizations that had invested heavily in open-plan redesigns. It suggested that the design theory — that removing physical barriers would increase collaboration — had made a fundamental error about human social behavior in work contexts. The error was not irrational: the hypothesis that proximity produces interaction is reasonable. But it did not account for the distinction between productive collaboration and ambient social exposure, or for the protective behavioral responses that workers develop when their concentration is constantly at risk of interruption.
The Interruption Economy — How Often and At What Cost
Gloria Mark at the University of California, Irvine has studied workplace interruption patterns for over two decades. Her research involves observational methods — watching knowledge workers in their natural work environments and tracking when they switch tasks, when they are interrupted, and how long they take to recover. The core finding has replicated across multiple studies and multiple years: workers in interrupted environments switch tasks on average every three to five minutes. They are rarely in a state of uninterrupted focus for more than a few minutes at a stretch.
The more consequential finding concerns recovery time. After an interruption — whether externally imposed (a colleague stopping by, a notification appearing, an ambient noise in the environment) or self-generated (the worker switching tasks voluntarily, often in response to accumulated interruption stress) — the average time required to return to the pre-interruption level of cognitive engagement with the original task is 23 minutes. Some individuals return faster; many take longer. The 23-minute figure is an average across a substantial observational dataset.
The implication for an eight-hour workday in a high-interruption environment is severe. If workers in open-plan offices experience task-switching events every three to five minutes, and if each interruption incurs a 23-minute recovery cost, then uninterrupted deep work in such an environment is mathematically scarce. The interruptions are not evenly distributed — many are brief and their recovery costs are smaller — but the cumulative cognitive load of continuous task-switching is a fundamental feature of the open-plan work environment, not an exception to its normal operation.
Deep Work and Cognitive States — What Requires Uninterrupted Time
Cal Newport's concept of "deep work" — defined as professional activity performed in a state of distraction-free concentration that pushes cognitive capability to its limit — provides a framework for understanding what open-plan environments structurally prevent. Deep work, by Newport's formulation, is not the only valuable form of work: shallow work (logistical tasks, coordination, email, routine decisions) has genuine organizational value and does not require deep concentration. But the high-value knowledge work that produces the outputs most consequential to organizations — original analysis, creative problem-solving, complex writing, technical development, strategic synthesis — requires sustained cognitive engagement that cannot be achieved in three-to-five-minute windows.
The neurobiological basis for this claim is the concept of cognitive flow, documented by Mihaly Csikszentmihalyi across decades of research. Flow states — characterized by complete absorption in a task, loss of self-consciousness, heightened performance, and intrinsic reward — require a specific relationship between task challenge and operator skill. They also require uninterrupted time to enter: flow states do not occur in the first minutes of engagement with a task; they emerge after a period of sustained concentration that allows working memory to fully load the relevant context and pattern-recognition systems to activate at full capacity.
An environment that interrupts concentration every three to five minutes structurally prevents flow states. Workers in such environments do not spend less time in flow — they spend effectively no time in flow. The cognitive work that requires flow is either deferred (workers find ways to work outside the open-plan environment: before office hours, after hours, at home, in private spaces) or it is never done. The organizational consequence is that the most valuable cognitive outputs are produced in spite of the designed work environment, not because of it.
The Interruption Tax — A Named Condition
The productivity differentials documented between high-privacy and open-plan work environments range from 40-50% across multiple studies, depending on the nature of the work, the degree of interruption, and the quality of the focus environment. Studies of software developers find that programmers in private offices produce substantially more debugged, functional code per unit of time than those in open-plan environments. Studies of researchers find similar output differentials. The differential is not uniform — some roles are not sensitive to interruption in the same way that deep cognitive work is — but for the class of knowledge work that requires sustained concentration, the environment matters substantially.
The cognitive cost of task-switching in interrupted environments: a documented recovery period of 23 minutes to return to deep focus after each interruption, producing a cumulative productivity differential of 40-50% between high-privacy and open-plan office environments. The Interruption Tax is not paid by the organization that designed the environment — it is paid by the cognitive worker whose attention is consumed by the recovery process. The tax is structural, not behavioral: it is not a function of individual self-discipline but of the properties of the environment that shapes what self-discipline must overcome.
The term "tax" is deliberate. A tax is an involuntary transfer of a resource from the person who holds it to the system that extracts it. The interruption tax transfers cognitive capacity from the worker to the environment — not through deliberate extraction, but through structural features of the workspace that impose attentional costs on every person who works within it. The worker who arrives with full cognitive capacity and works in an open-plan environment for eight hours does not return home with the output that eight hours of sustained deep work would have produced. The environment extracted the difference.
Who Benefits — The Real Economics of Open-Plan Design
The stated rationale for open-plan offices is collaboration and culture. The revealed rationale, visible in the economics of commercial real estate, is density. Open-plan offices accommodate more employees per square foot than private office configurations. The cost per employee per square foot is lower. For organizations managing significant real estate costs, open-plan design reduces facilities expense substantially. A private office might require 150-250 square feet per employee; an open-plan workstation can be configured at 50-75 square feet.
The critique of open-plan offices can be overstated. Many knowledge work roles genuinely benefit from ambient collaboration: sales teams, customer service functions, operational coordination roles, and creative teams working on joint projects all have legitimate reasons to work in close proximity where communication is fluid. The research documenting productivity losses in open-plan environments focuses primarily on solo deep work — the kind of work that software developers, researchers, and writers do. It does not characterize all knowledge work equally.
The response is not that all offices should be private. It is that workspace design should match the cognitive requirements of the work being done, and that the current dominant model — one open-plan configuration for all work types — optimizes for real estate cost rather than cognitive output. The Harvard study's finding is not that all interaction is bad; it is that forced ambient presence does not produce collaboration and does prevent deep work. Organizations that are serious about cognitive output would design environments that enable switching between high-focus and high-collaboration modes, rather than locking all workers into the mode that is cheapest to build.
Protective Environments — What Cognitive Infrastructure Requires
The research base on cognitive performance and workspace design is sufficient to describe what a cognitively protective work environment would require. It is not a universal prescription — different roles have different requirements — but the core elements are identifiable. Acoustic privacy sufficient to prevent ambient conversation from entering working memory. Visual privacy sufficient to prevent social monitoring anxiety. Control over interruption — the ability to signal unavailability and have that signal respected. Natural light, which is associated with better circadian regulation and self-reported wellbeing. Temperature control in the individual work zone. Access to nature or natural views, which is associated with attention restoration.
None of these requirements are technically difficult to meet. Private offices meet most of them. Acoustic pods, phone booths, and focus rooms within otherwise open-plan environments partially address them. The barrier to implementation is not technical — it is that providing genuinely cognitively protective environments increases the cost per square foot of the workspace. The organization that adopts open-plan design to reduce real estate expense cannot simultaneously provide the acoustic and visual privacy that would restore the cognitive productivity the open-plan environment removes. The trade-off is not hidden. It is the point.
The Infrastructure of Thought series begins with the workspace because it is the most proximate environment to the cognitive work that organizations claim to value. The Interruption Tax documented in this paper is paid every day, in every open-plan office, by every knowledge worker attempting to do work that requires sustained concentration. The size of the tax — 40-50% of productivity for deep cognitive work — is not a marginal inefficiency. It is a structural feature of the dominant form of knowledge work environment in the American economy, built and maintained by organizations whose stated goal is to maximize the productivity of their employees.
Selected References
- Bernstein, E. S., & Turban, S. (2018). The impact of the 'open' workspace on human collaboration. Philosophical Transactions of the Royal Society B: Biological Sciences, 373(1753).
- Mark, G., Gudith, D., & Klocke, U. (2008). The cost of interrupted work: More speed and stress. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. ACM Press.
- Mark, G., Iqbal, S. T., Czerwinski, M., & Johns, P. (2014). Bored Mondays and focused afternoons: The rhythm of attention and online activity in the workplace. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems.
- Newport, C. (2016). Deep Work: Rules for Focused Success in a Distracted World. Grand Central Publishing.
- Csikszentmihalyi, M. (1990). Flow: The Psychology of Optimal Experience. Harper & Row.
- DeMarco, T., & Lister, T. (1999). Peopleware: Productive Projects and Teams (2nd ed.). Dorset House. [Original research on programmer productivity and privacy]
- Stokols, D., et al. (2002). The ecology of team innovation: Linking the human environment and group creativity. In Group Creativity: Innovation through Collaboration. Oxford University Press.
- Hedge, A. (1982). The open-plan office: A systematic investigation of employee reactions to their work environment. Environment and Behavior, 14(5), 519–542.
- Kim, J., & de Dear, R. (2013). Workspace satisfaction: The privacy-communication trade-off in open-plan offices. Journal of Environmental Psychology, 36, 18–26.