AI as Attention Anchor

You needed to check one thing. You opened a browser tab to look it up. Forty minutes later you are eleven tabs deep, you have read about something unrelated and fascinating, and you have forgotten what you originally needed. The trip to look something up was the trip that ended the work. For an ADHD brain, the gap between leaving a task and returning to it is where the task often dies.

The fourth mechanism is about closing that gap: keeping support in one place so you never have to leave the work to get help.

The real cost of switching

Every context switch carries a cost. You drop what you were holding, reorient to the new thing, then have to reorient back. For most people this is a small tax. For an ADHD brain it is expensive in two ways. The reorientation back is harder, because working memory dropped the thread, and the new context is more likely to capture you completely, because novelty is exactly what an interest-based attention system chases. So the small look-it-up trip becomes a one-way door. You meant to step out and come back. You stepped out and the work was gone.

The 2025 ISCAP analysis identified reducing attention-switching cost as one of its four mechanisms, describing how AI provides real-time contextual support that removes the need to leave for external documentation searches. That phrasing undersells it for an ADHD brain. The point is not only convenience. The point is that the trip away is itself the danger, and a tool that lets you stay put has removed not a minor inconvenience but a frequent cause of the whole session collapsing.

Staying in the chair

In practice this means doing as much as possible without leaving the surface you are working on. When you hit the thing you would normally open a tab to look up, you ask in place instead. The answer comes to you. You never stand up, never open the browser, never pass the row of bookmarks that each whisper a different distraction. You stay in the chair, and staying in the chair is most of the battle.

This is also why the design of focus tools matters. A tool that hands you help without sending you out into the open ocean of the internet is doing something protective. A tool that, to help you, drops you into a feed or a browser is handing you the exact exit you were trying not to take. The medium is not neutral. Where the help lives determines whether getting it costs you the session.

The catch worth naming

There is an obvious tension here, and honesty requires saying it. The same AI that keeps you from wandering to eleven tabs is itself one of the most engaging surfaces ever built. The chat window that answers your question can also become the place you wander, asking it about something unrelated and fascinating until the work is gone after all. The anchor can slip its mooring.

The discipline is to use AI as a place to get answers and return, not a place to roam. Ask the question, take the answer, go back to the work. The moment you notice the conversation drifting away from the task, that is the same failure as the eleven tabs, wearing a smarter costume. The tool only anchors you if you treat it as an anchor rather than another sea to sail off into.

Next in the series: AI as body double, the always-available presence that makes starting less lonely.