Scent, Sound and Light

Scent: separate the mechanism from the marketing

Two things get constantly conflated. A standardised oral lavender preparation has randomised placebo-controlled evidence for anxiety and the sleeplessness that comes with it. Inhaled lavender is a different intervention, with a plausible mechanism, weak studies, and a problem nobody has solved: you cannot blind a smell. If a bedtime scent helps you, it is most likely conditioning — still useful, as long as you know which lever you are pulling.

Sound: masking is the honest claim

White noise contains all frequencies equally, pink rolls off toward the low end, brown more steeply still. The defensible benefit of all three is masking — covering the intermittent sounds that would otherwise surface you. Pink noise has genuinely interesting evidence, but under lab conditions a bedroom speaker does not reproduce. Brown noise is the most viral and has no published randomised sleep trials at all.

Light: real, modest, and not the main problem

Evening light does suppress melatonin and shift the clock. The effect is real and honestly modest — a phone at arm’s length delivers far less light than a laboratory rig. The larger problem is what the screen is for. Something engineered to hold attention has an interest structurally opposed to your falling asleep, and no amount of dark mode changes that.