Definition: Attention residue is the persistence of thinking about a previous task after you have moved on to a new one. Sophie Leroy, a professor at the University of Washington Bothell, introduced the idea in a 2009 paper, "Why Is It So Hard to Do My Work? The Challenge of Attention Residue When Switching Between Work Tasks", in Organizational Behavior and Human Decision Processes. Part of your attention stays on Task A while you try to do Task B, so Task B gets less of it.
TL;DR: Attention residue explains why switching tasks costs more than the seconds it takes to open another window. Leroy's two experiments found that people struggle to disengage from unfinished work, and their performance on the next task suffers. Finishing the first task, or writing down where you stopped, helps you let go. As AI agents multiply the threads you supervise, the switching cost is still yours. Build one free →
Picture a Tuesday. You are halfway through a budget review when a message asks for a quick decision on a hire. You answer in two minutes, return to the budget, and find you must reread the same page three times. The hire question is closed, but part of you is still turning it over. That leftover is the residue.
Why Attention Residue Matters in 2026
Modern work is fragmented, and the fragments carry a cost that calendars do not show. UC Irvine researcher Gloria Mark reports in her 2023 book Attention Span that time on a single screen before switching fell from roughly two and a half minutes in 2004 to about 47 seconds in her recent measurements. That figure describes digital switching behavior, not a biological limit on thought.
You will often see "it takes 23 minutes to refocus after an interruption." Handle it with care. The number comes from Mark's earlier interruption research and interviews, and it measures how long people took to return to the interrupted work, which includes the other tasks they did first. It is not a clean measure of recovery time, so do not quote it as one. Attention residue is the better-supported mechanism behind the feeling: Leroy tested it directly.
Leroy's work suggests residue is not just a distraction problem. In her studies, the harder it was to disengage from an unfinished task, the worse people did on the next one. Her practical advice, reported by the University of Washington, includes working offline to avoid interruptions, taking breaks between tasks, doing complicated work early in the day, and writing things down so the brain can let go and return more easily.
How Attention Residue Works
The mechanism is a mismatch between the task you are doing and the task your mind is still holding.
- You leave a task open. It is unfinished, urgent, or emotionally loaded.
- You switch. You open something else, often after a ping.
- The old task keeps running in the background. Your mind rehearses it while you read the new one.
- The new task gets a reduced share. You reread, miss details, and take longer.
- Closing the loop helps. Finishing the first task, or capturing the next step in writing, lets your attention release.
Attention Residue vs Related Ideas
| Idea | What it describes | Difference |
|---|---|---|
| Attention residue | Lingering thought about the previous task | The hidden cost of any switch |
| Context switching | Moving between tasks or tools | The action that creates residue |
| Deep work | Long, distraction-free focus | What residue prevents |
| Flow state | Full absorption in a task | Residue makes it harder to enter |
| Mental load | The burden of tracking open items | Open items feed residue |
Worked Example: A Day With Six Hops
A product manager plans a 90-minute block to write a spec. The day looks like this.
| Time | Event | Residue risk |
|---|---|---|
| 9:00 | Starts the spec | Low |
| 9:25 | Slack question about a bug, answers in place | High, spec is mid-paragraph |
| 9:40 | Returns to spec | Rereads the last page |
| 10:05 | Calendar alert for a standup | High, task unfinished |
| 10:30 | Back from standup | Spent 10 minutes recovering context |
The fixes are small. Before leaving the spec, write two lines: what is done, and the very next sentence to write. Batch the Slack answers into two set windows. Hand the bug triage to a system that gathers the facts, so the manager decides in one read instead of one conversation.
Common Mistakes
- Leaving tasks open on purpose to "stay in the zone." Unfinished work is what creates residue.
- Checking messages between every step. Each check is a switch.
- Trusting the 23-minute number. It is a rough return interval, not a law.
- Assuming AI removes the cost. Supervising several AI threads is still switching. No study cited here tests this, so it is reasoning from Leroy's mechanism, not a measured result.
Connection to Taskade
Two Taskade features address the same problem from different sides. The first is a place to park the state of a task. Your notes live in projects, so "where I stopped and what comes next" sits next to the work instead of in your head. The second is delegation: a Taskade AI Agent can gather updates, draft a summary, or watch a source so you are not the one who checks it. An automation can move information between tools you already use, such as Slack, Gmail, and Google Drive, and write the result back into a project.
Taskade does not reduce the cost of switching itself. It helps you make fewer switches by letting agents and automations carry the routine steps, and by keeping the resume note in the same workspace.
What You Would Build in Taskade
You would describe a daily focus brief. Each morning an agent reads your task list and the overnight updates, groups them into "decide now," "handle in a batch," and "ignore," and writes a two-line resume note for each open task. You start your first focus block with the state of your work in front of you.
Describe yours and build it free →
Related Concepts
- Deep Work: the focus that residue erodes
- Flow State: what fragmented attention blocks
- Batching: grouping small tasks to cut switches
- Mental Load: the open items that feed residue
- Time Management: protecting focus blocks
- Burnout: the long-run cost of constant switching
- Reduce Context Switching: a practical guide
- Cognitive Offloading: what happens when AI holds the load
Frequently Asked Questions About Attention Residue
What is attention residue?
Attention residue is the leftover attention that stays on a previous task after you switch to a new one. Sophie Leroy named it in a 2009 paper. It reduces the attention available for the new task.
Who discovered attention residue?
Sophie Leroy, of the University of Washington Bothell, introduced it in 2009 in Organizational Behavior and Human Decision Processes (volume 109, issue 2, pages 168 to 181).
How long does attention residue last?
There is no single figure. Leroy's experiments showed that unfinished tasks leave more residue and hurt later performance. The popular "23 minutes" number is a return interval from Gloria Mark's interruption research, not a measured residue duration.
Is attention residue the same as context switching?
No. Context switching is the act of moving between tasks. Attention residue is one of its costs: the part of your mind that stays behind.
How do you reduce attention residue?
Finish tasks before switching when you can. When you cannot, write down where you stopped so your brain can let go. Leroy also recommends offline time, breaks between tasks, and doing demanding work early in the day.
Does multitasking cause attention residue?
Switching between tasks does. Leroy's argument is that people are not truly doing two demanding things at once. They alternate, and each switch leaves residue behind.
Do AI agents reduce attention residue?
They can reduce the number of tasks you switch between by handling routine gathering and drafting. Supervising many agents is still switching, so batch your reviews.
How can Taskade help with attention residue?
Keep the resume note in the project, let a Taskade AI Agent gather updates, and use automations to move routine information so fewer things interrupt you.