You are not multitasking. You are switching, badly.

TL;DR

On WBUR's Here & Now / Inside the Box segment (11 August 2026), David Epstein — author of Range and The Sports Gene — argued that modern workplace culture presses people to multitask despite decades of research showing multitasking makes most people less productive and more stressed (WBUR).
The conversation centred on Epstein's framing of "monotasking" — a deliberate constraint of attention to one task at a time — as a productivity intervention grounded in cognitive science, not in wellness-industry lifestyle marketing.
The underlying evidence, cited across the interview and in the broader literature, includes work from Sophie Leroy on "attention residue," Anthony Wagner and colleagues at Stanford on heavy media multitaskers, and Gloria Mark at UC Irvine on interruption cost and attention span.

What actually happened

The WBUR segment is not itself news. It is a mainstream-media surfacing of a body of research that has been consistent for two decades. But it lands at a specific inflection point. Average knowledge-worker attention spans on a single screen, measured by Gloria Mark's group at UC Irvine, have collapsed from an average of ~2.5 minutes in 2004 to ~47 seconds in 2023 — the trend line is going one way. The number of digital notifications a typical office worker receives has approximately tripled over the same period.

The Epstein argument, distilled: the belief that "I am good at multitasking" is almost always wrong. What people call multitasking is task-switching, and task-switching has a measurable cognitive cost — the phenomenon Sophie Leroy termed "attention residue," where a portion of your attention remains on the task you just left, degrading your performance on the task you have just started. The cost is invisible because it is distributed across every switch.

The mechanism

There are three things happening when you "multitask":

One: switching cost. Every time you move from task A to task B, there is a measurable delay while your prefrontal cortex reallocates. On simple tasks, the delay is small — a few hundred milliseconds. On complex knowledge work, it can be several minutes to fully re-load context. If you switch tasks every 47 seconds, the switching cost dominates the actual work.

Two: attention residue. Leroy's 2009 research (Why Is It So Hard to Do My Work?, Organizational Behavior and Human Decision Processes) showed that when workers moved from task A to task B, cognitive residue from A degraded performance on B — measurably, on cognitive tests, for extended periods.

Three: the illusion of productivity. Stanford researchers Ophir, Nass and Wagner (2009, PNAS) tested heavy media multitaskers on cognitive control tasks. The heavy multitaskers believed they were better at filtering distractions. They were, in fact, significantly worse — worse at ignoring irrelevant environmental stimuli, worse at ignoring irrelevant representations in memory, and worse at task-switching itself. The very people most confident in their multitasking were the most impaired.

The Epstein insight — and this is where it becomes a growth argument rather than a productivity one — is that monotasking is not primarily about efficiency. It is about being able to finish something. Task completion produces the sense of progress that unfinished, half-attended work does not. The stress cost of a workday spent switching is not just cognitive load. It is the accumulated weight of things you never got to the bottom of.

What it actually means

The interesting thing about monotasking as an idea is that it is not new. What is new is the environment.

In 2004, when Mark's group first measured attention spans at 2.5 minutes, the average office worker had:

One computer.
No smartphone.
Email checked hourly, not constantly.
No Slack, no Teams, no persistent chat.

In 2026, the average knowledge worker has all of those, plus in-line notifications from ten to twenty applications, plus persistent expectations of near-real-time response. The infrastructure was built for a mode of attention — continuous partial attention — that the cognitive-science evidence says is corrosive.

Monotasking is not asking you to work harder. It is asking you to remove the ambient permission to be interrupted.

The small system

The reason this is a "growth" story rather than a "wellness" story is that it points to a small, adoptable behaviour change with an outsized effect. The pattern in the evidence is remarkably consistent: 60–90 minutes of protected single-task attention, once or twice a day, produces disproportionate output. Everything else in the day can be as fragmented as it needs to be.

The mechanism is not "focus more." The mechanism is arrange your environment so that switching is friction, not default. In practice:

Close every notification you did not explicitly decide to keep open, in the morning.
Batch communication into two or three windows in the day, not continuously.
Choose one task before you begin. Write it down. Don't choose another until it is done or you have taken a scheduled break.
The phone is not on the desk. It is in a drawer, in another room, in a Faraday pouch — the specific answer matters less than that the answer is "not visible."

None of these are original. All of them are difficult, precisely because the default environment is designed against them.

Hype deconstruction

Monotasking is having a media moment. That moment carries a small number of failure modes worth naming:

"Deep work" as identity theatre. Cal Newport's Deep Work framing has been co-opted into a certain kind of productivity performance — the person who tells you at length about their focus block. The evidence supports the practice. The performance is the enemy of the practice.
The productivity-industry version. Monotasking is being sold, in the corporate-training market, as a workshop. It is not a workshop. It is a constraint on your calendar.
The moralising version. Multitasking is not a moral failure. It is a rational response to an environment that punishes not responding to Slack in seven minutes. Fixing the environment is a systems problem, not a character problem.

The honest read is that monotasking works, that most people find it difficult to sustain, and that the difficulty is a feature of the modern work environment, not a personal defect.

Stakeholder landscape

Individual knowledge workers — direct beneficiaries if they can protect the time.
Managers — the people who most need to change behaviour, because managers set the norms about response expectations. A team where the manager Slacks at 9:47 PM cannot monotask.
HR / people leaders — the productivity gains from a formal "focus time" policy are large and well-documented. The organisational appetite to enforce it is small.
The tools market — Freedom, Cold Turkey, Opal, Apple's Screen Time, Google's Digital Wellbeing are all riding this wave commercially. Some are useful. Most are workarounds for absent policy.
Educators — the same evidence applies to students. School and university policies around laptop and phone use in class are the analogous frontier.

Cross-layer implications

AI note-taking and meeting recorders (Otter, Fireflies, Read AI, Granola) — reduce the marginal cost of missing a meeting, which removes one of the historical justifications for continuous attention. Meaningful support for monotasking.
Enterprise chat culture (Slack, Teams) — is now the single largest structural obstacle to focused work in modern organisations. There is no productive way around this that does not involve organisational policy.
Cognitive-load research and long COVID — post-viral cognitive fatigue is measurably worse under high task-switching environments. The monotasking case gets stronger, not weaker, for the millions of workers now managing that.
Generative AI as a task-completion partner — an interesting sub-argument. AI tools reduce the cost of finishing a task, which increases the value of protecting a focus block.

What this means for you

Try one experiment for two weeks:

Choose one 90-minute block per weekday, in the part of the day your cognitive peak actually lands (for most people, morning; know yourself).
In that block: notifications off, phone in another room, one task chosen in advance, no email, no chat.
Track: one number per day — did you finish the thing you sat down to do, yes or no.

Two weeks of data will tell you more about your working pattern than a year of reading about it. If you find the block impossible to protect, that is not a failure of discipline — it is a signal about your work environment that is worth acting on.

For managers: pick one — no meetings before 10 AM, or no Slack expectation before 10 AM. Publicly commit. Measure output over a quarter.

For families: this scales downward. The phone-in-a-drawer rule at the dinner table is the monotasking argument, applied to relationships.

Uncertainty ledger

Individual variation is real. A minority of people — probably in the low single-digit percentages — do appear to genuinely task-switch better than average. Most who believe they are in that group are not (Ophir/Nass/Wagner, PNAS 2009).
The evidence base is strong for cognitive tasks; weaker for creative and social work, where mind-wandering has documented benefits.
The measured attention-span decline may be partly an artefact of how the measurement adapts to changing tool landscapes. The trend line is robust; the exact numbers are contested.
Cultural differences in acceptable response latency (US corporate norms versus, say, German or Japanese) affect how much policy change is possible.

Bottom Line

You are not bad at focusing. You are working in an environment that was engineered to prevent focus. The good news is that the intervention is small, cheap and testable in two weeks — one 90-minute block, notifications off, one task, closed door. The bad news is that the reason it feels hard is not you. Do the experiment anyway.

Sources

WBUR Here & Now / Inside the Box, David Epstein interview on monotasking (11 Aug 2026) — Tier 2
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 — Tier 1 (peer-reviewed)
Ophir, E., Nass, C., Wagner, A. D. (2009), Cognitive control in media multitaskers — PNAS — Tier 1 (peer-reviewed)
Mark, G. et al. (multiple studies through 2023), UC Irvine, on attention spans and interruption cost — Tier 1
Newport, C., Deep Work (2016) — Tier 2 (background)
Epstein, D., Range: Why Generalists Triumph in a Specialized World (2019) — Tier 2 (background)