My Chameleon Switches Colour in Seconds. I Spend 23 Minutes Recovering From a Single Tab Switch.

My veiled chameleon, Ptolemy, can shift from a calm muted green to a vivid yellow-and-teal display in under three seconds. He does it on demand. He does it without any visible effort. He does not appear to spend the next half hour half-paying attention to whatever he was doing before, still mentally lodged in the previous conversation. He transitions completely and immediately, and when he shifts back, that is also complete and immediate.

I have been thinking about Ptolemy a lot since I read the research on attention residue.

What attention residue actually is

In 2011, organisational psychologist Sophie Leroy published research examining what happens cognitively when people switch tasks. The finding was specific and uncomfortable: when you move from Task A to Task B, part of your attention does not come with you. It remains with Task A — processing it, chewing on unfinished aspects of it, generating small, intrusive thoughts about it — while you are nominally doing Task B. Leroy called this attention residue, and its size is proportional to how unfinished or unresolved Task A felt at the moment of switching.

The 23-minute figure is from related work by Gloria Mark at UC Irvine: that is the average time workers needed to fully return to a primary task after an interruption — not just to navigate back to the window, but to reach equivalent depth of engagement. Interruptions do not cost 30 seconds. They cost 23 minutes.

The difference between Ptolemy and me

Ptolemy’s colour changes are complete, involuntary, and neurologically effortless. A chameleon does not carry an “attention residue” from its previous colouration. The mechanism is structural — pigment cells called chromatophores physically change shape in response to signals from the nervous system. There is no cognitive layer holding onto the old colour while displaying the new one.

Human attention does not work this way. We do not have a switch that moves us cleanly from one context to another. What we have instead is a process — and that process takes time, and costs working memory, and generates mental noise proportional to how unfinished the previous thing felt.

This means that every tab switch, every app hop, every “I’ll just quickly check” carries a debt. The debt is not the 10 seconds you spent looking at the other thing. The debt is the attention residue that trails you back into whatever you were doing before, and the time it takes to either clear it or overwrite it with enough depth in the new task.

The cost that does not show up in a time audit

If you tried to audit where your time goes — and many productivity frameworks encourage exactly this — you would probably not find the 23-minute attention-recovery cost in the log. You would record 10 minutes of writing, 4 minutes of email, 12 minutes of writing. The audit shows 26 minutes of productive work. What it does not show is the quality of the writing on either side of the email break, or the fact that it took 8 of those 12 “writing” minutes to get back to where you were before.

This is why the number of task switches matters more than the number of tasks. Two tasks with no switching between them cost less than one task with three interruptions.

What actually reduces the residue

Leroy’s research also identified something useful: planned stopping points reduce attention residue significantly. When people stopped Task A at a point they had intentionally pre-selected — a natural break, a section end, a specific moment they had decided in advance would be their pause — they showed less residue on Task B than people who were interrupted mid-flow.

This suggests a more practical question than “can I switch less?” — which is often genuinely not possible — and instead asks: when I switch, am I switching at a planned point or at a random one?

Ptolemy does not get to choose when he changes colour. That decision is made for him by his nervous system, responding to his environment. The colour change is always complete and immediate because he has no planning layer to contribute. Humans have that planning layer, which is both the problem and the potential solution: unlike Ptolemy, we can decide when the switch happens, even if we cannot control how long recovery takes.

A few things that help:

• Finishing a natural unit before switching — a paragraph, a completed thought, a clearly defined step — rather than breaking mid-sentence or mid-problem.
• Writing a two-sentence note about where you were before you switch. It externalises the context so your attention can release it rather than holding onto it.
• Setting explicit windows for different tasks, even loosely — “email before 9am and after 4pm” — so the switch itself is planned rather than reactive.
• Closing tabs you are not currently using rather than keeping them open in your peripheral vision. An open tab is not a neutral object; it is a small, persistent claim on the attention that should be on the current task.

Ptolemy has it easy. But he also cannot help it.