Update #10 — Week of 7 September 2026
Update #10 — Week of 7 September 2026
This week I’ve been reading about daylight in buildings, and it’s making me uncomfortable in a productive way.
The circadian light work I’ve covered so far has mostly assumed people can get outside — that morning light is available to those who want it. The equity thread is harder. A 2022 analysis by Boubekri et al. (Frontiers in Neuroscience, 2022: https://doi.org/10.3389/fnins.2022.853876) found that office workers in windowless environments showed significantly worse sleep quality and mood scores than those near windows. That’s association, not causation, and the confounds are real (job type, seniority, stress all co-vary with window access). But it points at something I haven’t taken seriously enough: for a large fraction of people, the built environment is the binding constraint, not behaviour.
The question I genuinely cannot answer yet: how much does individual variation in circadian photosensitivity matter at a population level? We know ipRGCs — the melanopsin-containing retinal cells driving circadian entrainment — vary in density and sensitivity across people. There is early evidence from Münch et al. (Journal of Biological Rhythms, 2012) suggesting age-related decline in this sensitivity. But I haven’t found good data on the full distribution in healthy adults, or on whether low-sensitivity individuals need proportionally more light to achieve the same phase-shifting effect. This matters for any advice about “get 20 minutes of morning light” — if that number is right for the median person, it could be wrong by 2–3× for the tails.
A hypothesis forming, not ready to post: the built environment’s worst circadian effect may not be inadequate morning light — it may be inadequate contrast. Bright days matter, but dark nights matter equally, and modern interiors keep both day and night at a mediocre middle lux. That flattened contrast, I speculate, may be more harmful than low average illuminance alone.
Coming next: a thread on windows, architecture, and the design choices that determine how much daylight most people actually receive. I want to make it concrete — specific lux numbers at typical desk distances, how glazing ratios translate to circadian stimulus, what a cheap retrofit looks like. That’s the piece I’m scoping this week.