Why We Left the Screen Out

Part of the field guide · Scent, Sound and Light

The last thing many of us see before we close our eyes is a screen. A phone held a few inches from the face, brightness dimmed but still glowing. One more episode. One more scroll. We built Mythrae to be the opposite of that, and the reasons are worth setting out properly, because they go further than “screens are bad.”

Two separate things make a bedtime screen poor company for sleep: the light it puts out, and the way it is built. They are worth pulling apart.

First, the light

Your body keeps time partly by light. Specialised cells in the retina, distinct from the ones you see with, report the brightness and colour of your surroundings to the brain’s master clock, which in turn governs the evening release of melatonin, the hormone that tells the body night has arrived. Light late in the day, particularly the short-wavelength “blue” light that screens are rich in, can dampen that signal.

The lab evidence here is reasonably consistent. Ordinary room light before bed has been shown to suppress melatonin and shorten the window in which it’s released, compared with dim light.¹ An often-cited study found that reading on a light-emitting e-reader before sleep, versus a printed book, delayed melatonin, pushed back the body clock, and left people less alert the next morning.² In a small laboratory study, an evening in front of an LED-backlit screen suppressed the usual rise in melatonin and left the 13 young men taking part less sleepy, and sharper on tests of attention and memory, at the hour you would rather be winding down.³

It would be easy to overstate all this. A phone held at arm’s length on low brightness delivers far less light than a bright laboratory rig, and sensitivity varies a good deal from person to person. A systematic review of the field concluded that evening light does shift human rhythms, while noting that effect sizes vary considerably with intensity, duration and timing.⁴ So the effect appears real, and at household brightness probably modest. It is also easy enough to sidestep: don’t hold a light source against your face in the last hour of the day.

The design matters more than the light

If light were the whole story, a dark-mode screen or a pair of amber glasses would settle the matter. They don’t, because the other half of the problem is what the screen is for.

The apps most of us reach for at night are not neutral surfaces. They are engineered, with great skill and considerable budget, to hold attention: autoplay that removes the decision to stop, feeds with no bottom, notifications timed to pull you back, the small unpredictable rewards that make a scroll feel like a slot machine. None of that is accidental, and none of it was drawn up with your sleep in mind. Researchers have a name for what follows — bedtime procrastination: going to bed later than you meant to when nothing outside you is stopping you, a lapse of self-regulation rather than of intention.⁵ What people are reluctant to give up, that work suggests, is not sleep so much as whatever they are doing instead; a device engineered to be hard to set down is, we would add, unusually good at supplying reasons not to stop. A systematic review of 67 studies in school-aged children and adolescents found screen time adversely associated with sleep in 90% of them — shorter sleep and later bedtimes, chiefly — with interactive use, computers, games and phones, more consistently implicated than television.⁶ Almost all of that work is observational, so which way the cause runs remains open.

A tool that profits from your attention can never quite be on the side of you letting it go.

A business whose success is counted in time-on-device has an interest that runs against your falling asleep. It can add a “bedtime mode” and a soothing voice, and the machine underneath still does better when you stay. The conflict sits in the business model, which is why no amount of patching quite reaches it.

So we left it out

Mythrae is audio you never have to look at. You press play, or ask your speaker aloud, and the screen can go face-down or stay in another room. There is no Mythrae app to open, no account to tend, no streak, no notification, nothing waiting underneath. Your pieces arrive as a private feed that plays through whatever you already own: a podcast app, a speaker, a pair of headphones. That is deliberate.

The measure we care about is that you stop listening — that if you drift off somewhere in the middle and forget the audio was ever playing, we built the right thing. As engagement metrics go, that is the worst one available, and we chose it knowingly. That is the mark we design toward: a piece you can’t remember ending.

What this asks of us

Building this way costs us something. We give up the nudges, the daily check-ins and the comforting dashboard of “active users.” That was the trade we wanted. The screen is the one thing almost every sleep product keeps, because the screen is how those products hold on. Leaving it out is the clearest way we know to say what we are after: we would rather hand the last hour of your day back to you, and take you somewhere that feels like yours.

← Back to The Lab