How do I protect deep work when I write?

The short answer

Five things, each backed by research:

1. Give your hardest thinking an uninterrupted block. Switching tasks leaves part of your attention behind.
2. Close or park the last task before you start. Unfinished work pulls your mind back to it.
3. End each session with a note on where you stopped and what comes next. A short plan for resuming measurably reduces the pull.
4. Keep the argument your own. The thinking that builds understanding cannot be handed to a tool.
5. Rest on purpose. A walk, especially outdoors, restores the attention deep work spends.

What is deep work?

Cal Newport, who coined the term in his 2016 book Deep Work, defines it as “the act of focusing without distraction on a cognitively demanding task.” For a nonfiction writer, that is the work of building an argument: deciding what you think, which evidence holds, and how the pieces connect. It is a popular book rather than a study, but the research below explains why the idea holds.

Why does switching tasks hurt deep work?

Because attention does not switch cleanly. In two experiments, Sophie Leroy found that “people need to stop thinking about one task in order to fully transition their attention and perform well on another,” and that it is “difficult for people to transition their attention away from an unfinished task” — so performance on the next task suffers (Leroy, 2009). She called what is left behind attention residue. People describe it in the words researchers use to measure it: “My mind kept on drifting back to the first task” (Newton and colleagues, 2020).

How do I stop the last task following me into the next?

Finish it if you can — and if you cannot, write down how you will pick it up. Across four studies, a brief “ready-to-resume” plan, in which you reflect on and plan your return to the interrupted task, reduced attention residue so that performance on the new task did not suffer (Leroy and Glomb, 2018). For a writer, that is a note at the end of the day: where you stopped, and the next sentence you meant to write.

Why does resisting distraction feel so hard?

Because it is more costly than switching. People are “generally biased to avoid” the effort of cognitive control, and a 2025 study found that shielding working memory from distraction feels more costly than simply updating it with whatever comes in (Papadopetraki and colleagues, 2025). The easy path is always the next interruption; protecting a block of time removes the choice.

Why does the hardest thinking suffer most from interruption?

Because a busy working memory has less left to hold distraction off. In a study published in Science, higher working-memory load led to greater interference from distractors (de Fockert and colleagues, 2001). The middle of a complex argument is exactly when a notification costs the most.

How long can I do deep work each day?

Less than you might think. In their classic study of expertise, Anders Ericsson and colleagues found essentially no benefit from deliberate practice beyond about four hours a day, and noted that famous novelists “tend to write only for 4 hr during the morning, leaving the rest of the day for rest and recuperation” (Ericsson and colleagues, 1993) — an observation drawn from author interviews rather than an experiment. Deliberate practice, they wrote, requires keeping “full attention during the entire period.”

Why can’t AI do the hard part for me?

Because the struggle is how the understanding is built. Cognitive load theory separates the load that does not help you learn — formatting, retrieval, mechanics — from the processing that builds understanding (Sweller and colleagues, 1998). A 2026 paper applying it to research writing puts it plainly: when a tool does that work, “the processing does not transfer to the writer; it does not occur” (Chauncey, 2026). The right way to use AI for nonfiction writing →

Does rest help, or is it wasted time?

It helps. Attention Restoration Theory proposes that directed attention is a limited resource that natural settings restore (Kaplan, 1995). In two experiments, a walk in nature — or even looking at pictures of nature — improved directed attention more than a walk in town (Berman and colleagues, 2008). And letting the mind wander between sessions can help the writing itself. Is mind-wandering good for writing? →

What does the research not say?

Some popular claims about deep work did not survive a check of the sources:

• “Switching costs up to 40% of your productivity.” A spoken estimate quoted on a web page, not a finding of the study it is attributed to.
• “Focus builds myelin, so a few focused minutes beat months of practice.” From a popular book. Activity-dependent myelination is real research, but it is shown causally in mice, and human evidence is correlational.
• “The brain works in 90–120-minute focus cycles.” The rhythm is established in sleep; evidence for it while awake is mixed, and one study is titled “no evidence for a 1.5-h rhythm.” A useful session length, not proven biology.
• “Your peak is the first two to four hours after waking.” No source found. Peak time depends on whether you are a morning or an evening person, and the first stretch after waking is dulled by sleep inertia.

How does Wretify protect deep work?

Its design follows the research above:

• No notifications, and one room at a time. Nothing inside Wretify asks for your attention while you write.
• A bookmark with a note in each chapter — your ready-to-resume plan — and the chapter opens scrolled to it.
• At most three active books, so fewer unfinished projects pull at your attention.
• An AI that refuses to write for you, so the thinking stays yours; it finds sources and checks grammar instead.
• Focus Mode, which fades every paragraph except the one you are in.

Why Wretify’s AI refuses to write →