Pursue

Research Notes

What published research says about attention, interruptions, timing and trusting an AI with your day. Plain answers, every claim tied to a source you can check.

published studies and classic papers
40
universities, hospitals and research labs
50
countries
10
years of research, 1966–2026
60
citations checked against DOI, arXiv, publisher or library records
40/40

Institutions where the studies behind these notes were written. They are not affiliated with Tre-Up and do not endorse it.

Updated 10 October 2026

Start with a question

Questions people ask about focus, interruptions and AI. Answered with sources.

  1. 01Do phone notifications hurt focus, even if I don't check them?Yes. In a 2015 Florida State University experiment, simply receiving a notification hurt performance on an attention-demanding task about as much as calling or texting, even when people never touched the phone. In a field experiment with 221 adults (Kushlev et al., 2016), people reported more inattention in the week their alerts were on.Interruptions and attention
  2. 02Should I turn off notifications completely, or batch them?The best evidence favours batching. In a randomized field experiment with 237 people (Duke and Georgetown, 2019), getting notifications in three batches a day made people feel more attentive, productive and in control than the usual stream, while getting none brought few of those benefits and more anxiety and fear of missing out.Interruptions and attention
  3. 03Why is it so hard to get back to work after an interruption?Because part of your attention stays with the task you left. Sophie Leroy's 2009 experiments at the University of Minnesota call this attention residue: people who switched before finishing did worse on the next task, and a 2008 study by UC Irvine and Humboldt University researchers found interrupted people sped up to compensate, at the cost of more stress and frustration.Interruptions and attention
  4. 04When is the right moment for a reminder or a nudge?When the person is able to take it in, which changes from moment to moment. In a trial with 1,255 users of a workplace well-being app (Bidargaddi et al., 2018), the same message made people about 4% more likely to use the app in the next day, and about 12% more likely at 12:30 on weekends.Timing and receptivity
  5. 05Do reminder apps change what people do, or just how often they open the app?Often only the second. In a six-month University of Michigan trial with 287 adults with high blood pressure (Golbus et al., 2026), notifications made people use the app more but did not raise steps in the next hour or improve food choices the next day. Opening an app is engagement, not benefit.Engagement is not benefit
  6. 06Why do I push back when an app tells me what to do?Because being pushed feels like losing a freedom, and people act to win it back; psychologists call this reactance. In an experiment with 394 people (Richards et al., 2022), simply reminding them the choice was theirs reduced the pushback against health messages, and the effect was still there a week later.Autonomy and wording
  7. 07Should I trust an AI's advice about how to spend my day?Trust it as far as it has earned, and no further. Across six experiments, people followed advice more readily when they believed it came from an algorithm rather than a person (Logg et al., 2019), yet watching an automated aid make mistakes made them distrust even a reliable one (Dzindolet et al., 2003).Trust and uncertainty
  8. 08How should an AI say it isn't sure?Plainly, and without inventing a confidence number. Modern neural networks are often over-confident (Guo et al., 2017), and a system allowed to decline uncertain cases makes far fewer errors on the ones it answers (Geifman and El-Yaniv, 2017); a 2019 review led from the University of Cambridge found that admitting uncertainty does not necessarily cost trust.Trust and uncertainty
  9. 09How can an app help me if it doesn't know me yet?By asking, briefly, instead of guessing. A 2023 systematic review (Yuan and Hernandez) treats the 'cold start', a new user with no history, as its own problem, and asking the person directly is one approach that works from the first day. Later, fresh answers should outweigh old patterns, because preferences drift (Koren, 2009).Learning a person
  10. 10When is it better for an AI assistant to say nothing?Whenever speaking would cost more attention than it gives back: when the person can't take it in, the evidence is thin, or nothing needs to change. The 2018 framework for just-in-time support (Nahum-Shani et al.) counts giving nothing as a designed option, and declining uncertain cases makes a system's other answers more reliable (Geifman and El-Yaniv, 2017).Timing and receptivity
  11. 11If I delete my data from an app, is it deleted from backups too?Eventually, but usually not at the same moment. EU and UK law require erasure without undue delay, and the UK regulator, the ICO, says backups count but may be put 'beyond use' until they are overwritten; Google Cloud commits to deleting customer data within about six months, and some backups are locked so that no one can delete them early.Privacy and deletion
  12. 12Is there an app that notices when my day is slipping and suggests one next step?Tre-Up, an iOS app in development, is being built for exactly that, and it is not released yet: its AI, Pursue, is designed to notice when your focus or energy shifts and to answer with one small next step, one question, or nothing. Until it ships, your phone's focus modes and scheduled notification summaries cover part of the job.Timing and receptivity

Topics

Seven groups, one subject: the moment a day slips.

Where the research comes from

50 institutions in 10 countries.

The studies behind these notes were written at universities, hospitals and industry research labs. Institutions are the authors' affiliations as printed in each publication. Naming them says where the research was done. They are not affiliated with Tre-Up and do not endorse Tre-Up or Pursue.

United States28 institutions

United Kingdom5 institutions

Canada3 institutions

China3 institutions

Israel3 institutions

Japan3 institutions

Germany2 institutions

Australia1 institution

Slovenia1 institution

Spain1 institution

All 54 sources, grouped by topic

How these notes are made

Read, checked, written in our own words.

Three steps, shown on one real source: Mark et al., 2008, behind the note on getting back to work after an interruption.

  1. 01

    Find

    Pursue Lab, Tre-Up's research process, looks for published studies that bear on one question at a time.

    The question

    Why is it so hard to get back to work after an interruption?Mark G, Gudith D, Klocke U (2008). The cost of interrupted work: more speed and stress. Proceedings of CHI 2008, 107–110.
  2. 02

    Check

    Every citation is matched to its DOI, arXiv or publisher record. A source we cannot confirm is marked as unconfirmed, never quietly kept.

    The record

    Confirmed against the DOI record: doi.org/10.1145/1357054.1357072Written at University of California, Irvine and Humboldt-Universität zu Berlin.
  3. 03

    Summarise

    We write what a study found in our own words, say what it did not show, and link to the original. We don't republish papers, figures or paywalled text.

    In our words

    Interrupted people finished their tasks faster and just as well, but paid for it: more stress, frustration, time pressure and effort. They seem to make up for interruptions by speeding up.Where it stops: A short lab task. It does not measure how long it takes to get back on track, and gives no minutes figure.The full source page

Research Notes are general information, not medical advice.

Our full method

About Tre-Up

Tre-Up is an iOS app in development, built by its founder in Istanbul, Türkiye, and not yet incorporated. It is a behavioral decision system: it is designed to help people notice the moment their day starts to slip and decide what to do next.

Pursue is the AI inside Tre-Up. When your focus, energy or rhythm shifts, Pursue is designed to answer with one small next step, a single question, or nothing at all when nothing needs to change. The research on this page is why it is designed to stay quiet more often than it speaks.

Tre-Up does not sell your data. Not personal, not anonymous, not pooled.

Pursue

The AI inside Tre-Up

Tre-Up is built on this research.

Pursue, the AI inside Tre-Up, is designed to notice when your day starts to slip and offer one small next step. The iOS app is in development. Invitations go out in small groups; leave your email and we will contact you when yours is ready.

Your email is used only for your invitation and important product updates. It is never sold. Unsubscribe at any time. Privacy policy