Selfey Science

The science behind taking care of yourself.

Explore the research behind rest, connection, movement, mental load, joy and the everyday things that help us feel more like ourselves.

36 studies, read properly — what each one found, what it can't tell us, and what it might mean on an ordinary Tuesday.

a woman working on a laptop in a busy shared space

Worth knowing

an unmade bed in a dim room with balcony doors beyond
RestStrong confidence

Seven hours isn't a wellness trend. It's a floor.

The short version

A panel convened by the American Academy of Sleep Medicine and the Sleep Research Society reviewed the published evidence and agreed that adults should regularly sleep seven hours or more. Below that, the reviewed research links short sleep to higher risk across weight, heart health, mood, immune function and mortality.

Why it matters

It reframes a habitual six-hour night as a health decision rather than a personality trait — closer in weight to how we already think about diet or movement.

Watson, N. F., Badr, M. S., Belenky, G., et al. (American Academy of Sleep Medicine and Sleep Research Society) · Sleep, 2015

Read the research →

Evidence snapshots

Four findings, drawn out.

Every number here comes from the study it cites. Where a study describes an idea rather than a measurement, we draw the idea instead.

  • Connection

    Being close to people shows up in the health data.

    Likelihood of survival over the study period
    • Weaker social relationshipsreference
    • Stronger social relationships+50% likelihood of survival

    The 50% figure is the pooled average reported by Holt-Lunstad, Smith & Layton (PLoS Medicine, 2010). It is an association across 148 studies, not an individual prediction.

    A meta-analysis pooling 148 studies found that people with stronger social relationships had, on average, a 50% higher likelihood of survival across the study periods than people with weaker ones. The association held across age, sex, starting health and cause of death.

    Read the research →
  • Life

    Why mental load can be exhausting when you haven't moved all day.

    What mental load is actually made of
    1. Anticipatingnoticing what will be needed
    2. Rememberingholding it until it's relevant
    3. Planningworking out how and when
    4. Decidingchoosing, often for other people
    5. Following upchecking it actually happened

    A Selfey diagram illustrating the concept described in this review. Not a measurement, and not a claim about sequence in the brain.

    A review of executive-function research concludes that planning, deciding, remembering and initiating draw on a shared, limited pool of resources. When total demand on that pool is high, performance on tasks that would normally feel easy measurably declines.

    Read the research →
  • Joy

    Two hours outside a week, however you get them.

    Recreational time in nature, per week
    120
    0360 minutes

    120 minutes — where the association appears

    200300 minutes · benefits levelled off around here

    Both figures are reported in White et al. (Scientific Reports, 2019). Association only — the study cannot show the nature itself is doing the work.

    In a large England-wide survey, people who reported at least 120 minutes of recreational time in nature in the previous week were significantly more likely to report good health and high wellbeing than people who reported none. It made no difference whether those minutes came in one long visit or several short ones.

    Read the research →
  • Connection

    Contact and connection are not the same thing.

    Contact ≠ connection

    Contact

    • Being around people
    • Group chats
    • A full week
    • Interaction counted

    Connection

    • Feeling people are available
    • Being known by a few
    • Somewhere to bring it
    • Support perceived

    A Selfey diagram illustrating the distinction this research draws between received contact and perceived support.

    Across studies of adolescents, higher perceived social support — the sense that people are available and care — was consistently linked to better psychological adjustment under stress. Support from parents and close friends showed the strongest associations; sheer amount of contact did not stand in for it.

    Read the research →

Keep in mindOne study gives us a piece of the picture. Selfey looks for patterns across the broader evidence rather than treating a single finding as a rule.

Explore the science of

From the evidence desk

Everything we've read, in plain language.

36 of 36 studies

Science in real life

What does the research look like on an ordinary Tuesday?

  • a row of shopping trolleys outside a store at night

    The research

    Psychological detachment during off-hours was independently associated with lower exhaustion.

    Real life

    Leaving the phone in another room during dinner is not a productivity technique. In this literature it is closer to the mechanism itself.

    The study behind this →
  • two women talking on swings in a park

    The research

    At least 120 minutes a week in nature was associated with better self-reported health and wellbeing.

    Real life

    Twenty minutes of the school run done the long way, on foot, five days a week. Nobody would call it being in nature. It is the same number of minutes.

    The study behind this →
  • a child pausing on a bike at the edge of a woodland path

    The research

    Executive functions share a limited pool of resources; when total demand is high, easy tasks get harder.

    Real life

    Sitting in the car for ten quiet minutes before walking into the house may not look productive — and that may be exactly the point.

    The study behind this →

How we read research

Evidence isn't the same as certainty.

Every study here carries a confidence label — an editorial judgement about the whole body of evidence, not a score for a single paper. Where a finding is commonly overread, the page says what it does not mean.

  • Strong confidence

    A consistent body of evidence — reviews, meta-analyses or established guidelines pointing the same way across populations.

  • Moderate confidence

    Carefully designed research that still needs context or replication: trials, longitudinal work, well-conducted single studies.

  • Emerging evidence

    An early signal worth knowing about: preliminary studies, smaller samples, designs that can't yet settle the question.

  • Mixed findings

    Genuinely conflicting findings. The literature doesn't agree, and we say so rather than pick a side.

Evidence is context. A plan is what you do with it.