Automatic checking and Lost Intention · practical guide / pillar

How to stop automatic phone checking without deleting everything

If your hand reaches for the phone before you have decided what you want from it, do not start by deleting half your life.

Mike Salari · Updated 22 September 2026
A closed loop with a gold wedge driven through the top, breaking the circuit
Automatic, until something is put in the way.

If your hand reaches for the phone before you have decided what you want from it, do not start by deleting half your life.

Start by finding where the automaticity enters the sequence.

The Check Loop uses four stages:

Cue -> Reach -> Drift -> Repeat

A good intervention changes the stage that is actually failing.

That sounds simple, but it prevents a common mistake: applying the same solution to every kind of phone problem.

A notification problem is not the same as a boredom problem. A reflexive unlock is not the same as a useful message reply that becomes twenty minutes of scrolling. A phone that interrupts coding is not the same as a phone that takes over bedtime.

The aim is not "less phone" as a moral achievement. The aim is more deliberate use.

Step 1: separate automatic checking from useful phone use

Before changing settings, list what your phone genuinely needs to do for you.

For example:

This matters because an intervention you constantly fight is unlikely to survive.

Research on digital self-control tools repeatedly finds a tension between effectiveness and autonomy. Users want enough support to change behavior but not so much coercion that legitimate use becomes irritating.

So do not begin with "How do I make my phone unusable?"

Begin with:

Which phone actions do I want to stay easy, and which actions do I want to stop happening automatically?

Step 2: identify the cue

A cue can be external:

Or internal:

Research on spontaneous smartphone checking is useful because it explicitly recognizes phone-directed attention that occurs without an external prompt and without a specific conscious goal.

For two days, log only the moment before the reach.

Do not count total minutes yet.

Write:

context -> cue -> app opened

Examples:

build running -> waiting -> Reddit

partner pauses -> silence -> Instagram

email difficult -> discomfort -> news

phone vibrates -> alert -> WhatsApp

You are looking for repeated conditions.

Step 3: make the reach slightly less automatic

Once you know the cue, add the smallest useful amount of friction.

Physical friction is often underrated:

Software friction can help too:

The word slightly matters.

If the friction is too weak, automatic behavior passes through it.

If the friction is too strong, you may remove the system because it also blocks useful behavior.

Step 4: create an intention before unlocking

Use a short question:

What am I opening the phone to do?

Do not turn this into a mindfulness ceremony.

One sentence is enough:

The last example is important. Intentional use can include entertainment.

The Check Loop is not a purity system.

The problem is not that you enjoyed an app. The problem is when the app selected itself.

Step 5: stop the drift after legitimate use

Many phone sessions begin correctly.

You open the phone to answer a message. That task finishes. Then you stay.

I call this Lost Intention.

If this is your main pattern, access friction alone can be badly targeted. You needed access.

Instead, add an exit:

If the answer is no, the original purpose probably ended.

Read: Lost Intention.

Step 6: change the repeat condition

The final stage is Repeat.

Even if one check is short, the environment can immediately prepare the next one.

Examples:

After a check, reset the environment.

Put the phone back.

Clear the physical cue.

Return to a neutral screen if that helps.

Silence the source that does not need to call you back in five minutes.

The repeat stage is where a single check becomes a day-long pattern.

Why willpower is a weak default strategy

Willpower can help in a moment. The problem is asking it to solve the same predictable problem hundreds of times.

If a cue repeatedly appears in the same environment, redesigning that environment is usually more practical than staging the same internal argument again.

That does not mean people have no agency. It means agency can be used earlier, when the environment is being configured.

You decide once where the phone lives during coding.

You decide once which notifications deserve interruption rights.

You decide once whether the most reflexive app belongs on the first screen.

Then the environment carries part of the load.

Do app blockers work?

Sometimes.

It would be inaccurate to say digital self-control tools never work. Systematic reviews and meta-analyses have found evidence that some interventions can reduce unwanted technology use, while also noting substantial variation in design, evidence quality, long-term evaluation, and user fit.

The practical question is not "Are blockers good or bad?"

It is:

Does this intervention act on the stage of my loop that needs help?

A blocker can be excellent if access itself is the problem.

It is less useful if the problem is what happens after a legitimate task, or if you override it every time because it blocks work you actually need to do.

A seven-day Check Loop experiment

Pick one context, not your whole phone.

Day 1-2: observe

Record Cue -> Reach -> Drift -> Repeat without changing anything.

Day 3: choose one friction point

Change one thing before the reach or immediately after the intended task.

Day 4-6: keep the same experiment

Do not add five more rules because you had one bad afternoon.

Day 7: review

Ask:

If not, redesign it.

A sustainable system is better than a severe one you disable on Tuesday.

What success looks like

Success is not never checking your phone.

Success can look like:

You are not trying to win against your phone.

You are trying to make your environment more likely to produce the behavior you already prefer.

That is how automatic checking becomes deliberate use.

Continue

For the full method and the seven-day field test, see The Check Loop book.

Sources

  1. Berntsen, D., Hoyle, R. H., Møller, D. M., & Rubin, D. C. (2023). Digital daydreaming: Introducing the spontaneous smartphone checking scale. https://doi.org/10.1002/acp.4034
  2. Monge Roffarello, A., & De Russis, L. (2023). Achieving Digital Wellbeing Through Digital Self-Control Tools: A Systematic Review and Meta-Analysis. https://doi.org/10.1145/3571810
  3. Lyngs, U., et al. (2022). The Goldilocks level of support. https://doi.org/10.1016/j.ijhcs.2022.102869
  4. Biedermann, D., et al. (2021). Digital self-control interventions for distracting media multitasking - A systematic review. https://doi.org/10.1111/jcal.12581
The Check Loop book cover by Mike Salari
A practical next step

The Check Loop

The Check Loop is a field manual for noticing automatic phone use, adding useful friction, and keeping the smartphone you actually need.

Get The Check Loop →

← Back to the journal · The Check Loop home