How Long Does It Actually Take to Get Back Into Focus After an Interruption?
Written By Aftertone Team
Thursday, May 14, 2026
15 min read
How Long Does It Actually Take to Get Back Into Focus After an Interruption?
On average, 23 minutes and 15 seconds. That figure describes a population average, not a fixed biological constant, and understanding what drives the variation is more useful than the number itself. The research behind it, conducted by Gloria Mark and colleagues at UC Irvine, is the most cited finding in the science of workplace interruption. It's also one of the most frequently misunderstood.
Where the 23-minute figure comes from
Gloria Mark, Daniela Gudith, and Ulrich Klocke conducted a field study at UC Irvine in 2008, tracking 36 knowledge workers at second-by-second resolution across their working days. They measured not just interruptions but recovery — specifically, how long it took workers to return to the same task they had been doing before an interruption occurred.
The headline finding: it took an average of 23 minutes and 15 seconds before workers resumed their original task after an interruption. But the fuller picture is more revealing. Workers didn't return directly to their original task. They typically worked through an average of two other tasks before coming back. The 23 minutes is not 23 minutes of sitting unproductively — it's 23 minutes of appearing to work while cognitively still in transit, the original task's context only partially rebuilt.
A second key finding from the same research: interrupted work was resumed on the same day 81.9% of the time. The interruption didn't kill the task. It delayed it, but the delay was costly in ways that don't show up in a simple time accounting.
Why recovery takes so long
The 23-minute figure isn't arbitrary. It reflects what the brain actually has to do to return to depth after a context switch.
Deep, complex work requires a specific cognitive configuration: the task's context loaded into working memory, the relevant background and prior thinking active, the current problem state clear, the next action understood. This configuration doesn't exist pre-formed waiting to be resumed. It's built progressively over the first 15 to 20 minutes of a focus session. By the time you reach genuine depth, you've assembled a fragile cognitive structure through sustained effort.
An interruption doesn't pause this structure. It dismantles it. The social processing required to respond to a colleague, handle a request, or read a message activates an incompatible cognitive state — one that cannot coexist with the deep work state. When the interruption ends, the deep work state has to be rebuilt from scratch. The 23 minutes is primarily the rebuilding time, not recovery from fatigue.
Sophie Leroy's 2009 research on attention residue adds a further mechanism: the interrupted task leaves active cognitive residue — partial processing that continues running in the background, occupying working memory that the recovery effort needs. The residue from the interruption and the effort of rebuilding the prior context compete for the same limited resource. This is why return to depth takes significantly longer than return to the task — you're rebuilding under load.
What actually determines recovery time
The 23-minute average conceals significant variation. Several factors push the recovery time longer or shorter:
How complete the interrupted task was. Leroy's research found that attention residue is strongest when the prior task is incomplete or unresolved. An interruption that arrives when you're in the middle of a complex problem leaves more residue than one that arrives at a natural stopping point. The more unresolved the prior work state, the more the brain continues processing it during recovery, and the longer full engagement with the new task takes.
How cognitively demanding the interruption was. A two-second glance at a phone notification generates less residue than a 10-minute conversation that required active participation. The more the interruption itself demanded cognitive engagement, the more cognitive residue it leaves — adding a second layer of competing processing to the recovery effort.
Whether a ready-to-resume note was written. Leroy and Glomb's 2018 follow-up to the attention residue research found that writing a brief note about where you left off before switching — one sentence about the current state and the next action — significantly reduced residue at the point of switch. The mechanism: the note satisfies the brain's need for completion by providing a concrete plan for the unfinished task, which reduces the automatic background processing that would otherwise continue generating residue throughout the interruption.
How many prior interruptions have accumulated. Recovery from a single interruption from an otherwise clean cognitive state is faster than recovery from the third interruption in an hour, because each incomplete recovery leaves its own residue layer. The compound effect of multiple interruptions means that later interruptions have longer recovery times than earlier ones, even when the interruption itself is identical, because the starting state for recovery is progressively more degraded.
Whether the recovery happens in a distraction-free environment. If the 23-minute recovery period is itself populated with smaller interruptions — the ambient noise of an open office, visible notifications, colleagues passing — the recovery takes longer because the cognitive rebuilding process is itself being interrupted. Recovery requires the same uninterrupted conditions that entry into depth requires. Attempting to recover in a noisy, interruptible environment is attempting to rebuild a structure while it's being dismantled again.
The two interruption types and their different recovery costs
Mark's research and subsequent work by Victor González and Mark (2004) identified an important distinction between self-initiated interruptions (you choose to switch tasks or check something) and externally imposed interruptions (a colleague, notification, or unexpected demand pulls you away).
Self-initiated interruptions — choosing to check email, deciding to take a break — tend to have lower recovery costs because they occur at moments of natural stopping points or reduced momentum, and because the brain has advance notice of the switch. The task context can be partially stored rather than abruptly dismantled.
Externally imposed interruptions have higher recovery costs because they arrive without warning, at arbitrary moments in the task flow, and require immediate social or communicative engagement that is cognitively incompatible with the prior task state. The phone ringing, the colleague appearing, the urgent Slack message, these produce the most severe task-set disruption and the longest recovery times.
This is why the design of the work environment matters as much as the behaviour within it. An environment where externally imposed interruptions are structurally possible at any moment is an environment where 23-minute recovery costs are the recurring norm. An environment where externally imposed interruptions are structurally blocked during focus periods eliminates the highest-cost interruption type at the source.
The practical arithmetic
The 23-minute figure has a practical implication that changes how most people think about their schedule: every externally imposed interruption to complex work effectively costs 25 minutes, not the duration of the interruption itself. A 2-minute conversation costs 25 minutes. A glanced-at notification costs at minimum several minutes and potentially the remainder of the focus session.
For a four-hour morning with three such interruptions: 3 × 23 minutes = 69 minutes of recovery overhead, before accounting for the degraded quality of the recovery periods themselves. The four-hour morning produces something closer to 2.5 hours of actual focused work under even moderate interruption load, and that assumes full recovery between interruptions, which Gloria Mark's 275-interruptions-per-day average makes structurally unlikely.
The planned versus actual comparison over two to three weeks makes this concrete for any individual: comparing intended output to actual output on interrupted versus uninterrupted days reveals the specific cost in your work, rather than the population average. Most people find the difference larger than they expected before tracking it.
How to reduce recovery time specifically
Write a ready-to-resume note before any anticipated interruption. Before leaving a task for a meeting, a break, or any planned switch — one sentence: where you are, and what the first action is on return. Leroy and Glomb's research showed this simple intervention measurably reduces the residue generated at the switch point, which directly reduces the recovery time required afterward.
Create a clean stopping point before unavoidable interruptions. Rather than stopping mid-thought when a meeting starts, spend 60 seconds reaching the end of the current sentence, paragraph, or logical unit before switching. The more complete the stopping point, the less residue it generates. An interruption at a natural stopping point produces lower recovery costs than one mid-flow.
Build transition buffers into the schedule. After a known interruption — a meeting, a phone call, a collaborative session — schedule 10 to 15 minutes of low-stakes admin or email before the next focus block, rather than expecting the focus block to resume immediately from depth. Explicit transition time accepts the recovery cost rather than hiding it and wondering why the focus block didn't produce what was expected.
Eliminate the passive interruption load during recovery. The recovery period is as vulnerable to interruption as the original focus session. Phone away, notifications off, Slack closed — during the recovery, not just during the original focus block. Recovery from an interruption requires the same environmental conditions as entry into depth from scratch.
Aftertone's Focus Screen applies this at the software level: when a focus session resumes, it presents only the current task. Every additional element on screen during re-entry is a potential activation point for the residue from prior tasks or anticipatory attention toward upcoming ones. The simpler the environment at re-entry, the lower the competing cognitive load, and the faster the 23-minute clock runs down.
Frequently asked questions
How long does it take to get back into focus after an interruption?
An average of 23 minutes and 15 seconds, based on Gloria Mark, Daniela Gudith, and Ulrich Klocke's 2008 field study at UC Irvine tracking 36 knowledge workers at second-by-second resolution. The return typically involves two intervening tasks before workers resume their original task. The 23 minutes is not lost time — it's the time required to rebuild the cognitive state the interruption dismantled, during which apparent productivity is actually degraded recovery.
Is the 23-minute focus recovery time really accurate?
It's an average, not a fixed constant. Recovery time varies significantly based on how complete the interrupted task was (incomplete tasks generate more residue), how demanding the interruption was, whether a ready-to-resume note was written before switching, and how many prior interruptions have already accumulated. Single-study findings should always be treated as directional rather than precise, but the direction is well-established: recovery from deep work interruptions takes far longer than the interruption itself.
Does the 23-minute rule apply to all types of interruptions?
The 23-minute figure does not apply equally to all interruption types. It is specific to externally imposed interruptions to complex cognitive work mid-flow. Self-initiated switches, interruptions during routine tasks, and interruptions at natural stopping points all have lower recovery costs. The 23-minute figure applies most accurately to the most damaging case: an unexpected external interruption arriving during a complex focus session in mid-flow.
How do I speed up focus recovery after an interruption?
Three interventions with research backing. First, write a ready-to-resume note before the interruption — one sentence about where you are and what the first action is on return. Leroy and Glomb (2018) found this measurably reduces the residue generated at the switch point. Second, reach a natural stopping point before switching rather than stopping mid-thought. Third, eliminate all secondary interruptions during the recovery period — phone away, notifications off, because recovery requires the same distraction-free conditions as original entry into depth.
Why do interruptions feel like they ruin the whole session rather than just costing 23 minutes?
Interruptions feel like they ruin the whole session because recovery from one interruption is rarely complete before the next arrives. Gloria Mark's research found approximately 275 interruptions per day on average — a rate at which full 23-minute recovery between interruptions is mathematically impossible. Incomplete recovery compounds: each subsequent interruption begins from a progressively more degraded cognitive baseline, so the session doesn't feel ruined by one interruption but by the accumulated incomplete recoveries.
