Your Body Keeps the Average
One heroic early night changes almost nothing, and neither does one wrecked one. A new look at 105 people living ordinary lives suggests the body's clock is listening to the pattern and barely hears the day.
You went to bed at half past nine last night. You did it on purpose, with the phone in another room and a book, because the week had been ragged and you had decided to put it right. And this morning you felt — fine. Ordinary. Much as you felt on Tuesday, after the night that ended at one.
It is a small and familiar disappointment, and it usually gets filed under that sort of thing doesn't work for me. A preprint posted in June suggests a different filing.
Billy Smith and thirteen colleagues, recruiting in Newcastle upon Tyne, gave 105 healthy adults a smartwatch and a continuous glucose monitor and let them live as usual for up to four weeks — just under two thousand days of ordinary life between them. From the heart-rate data they drew out each person's daily rhythm and the hour at which it peaked, a marker of where the body's clock is sitting. Then they split everything a person did into two parts. The habit: their usual wake time, usual first meal, usual exercise. And the deviation: how far any particular day strayed from it.
The habits explained 42.3 per cent of the variation in when the rhythm peaked. The daily deviations explained 0.9 per cent.
The body is not marking each day. It is keeping a running average, and a single entry — in either direction — barely shows.
Why the good night did not count
Most of us run a routine on the logic of a ledger. Each day is an entry, a good one is a credit and a bad one a debit, and the account can be squared by Friday if we are strict enough. Under that logic the early night ought to have paid out by morning.
The rhythm in this study did not behave like that. Where a person's clock sat was explained well by what they habitually did and hardly at all by what they did yesterday. Habitual wake time mattered most; the usual timing of the day's hardest exercise and of the first meal came next. A day that departed from the pattern did very little, because the following day there was the pattern again.
Which cuts both ways, and the second way is the kinder one. The late night did not undo much either. The single wrecked evening that people atone for — with a lie-in, then a nap, then a heroic early night, which is three further departures from the pattern — was, on this evidence, mostly not worth atoning for. A good deal of the damage is done by the repair.
The limits are real and the authors set them out. This is a preprint, not yet peer reviewed. It is observational, so it cannot show that the habits set the clock rather than the clock setting the habits — a person built to peak late will wake late for exactly that reason. A hundred or so people from one English city wore the devices for about nineteen days each. The heart-rate rhythm is one of the body's clocks and not the central one, and light was not measured at all. The deviations, too, were the ordinary kind: an hour here or there in a normal life. Nothing in it says a week of night shifts passes unnoticed. And it measured where the rhythm sat, not how anyone felt at their desk the next afternoon, which is a separate matter and one a short night can certainly settle.
Moving an average
An average moves slowly and only by repetition. That is tedious news and also clarifying, because it says where the effort goes.
Pick the anchors rather than the days. On this evidence the candidates are when you get up and when you first eat: two times, held roughly, most days. Not perfectly. The arithmetic of an average is forgiving in precisely the way a streak is not — one loose morning in a fortnight hardly registers, and there is nothing to break.
Change them by small amounts and then wait longer than feels reasonable. A quarter of an hour earlier, kept for a fortnight, before asking whether anything has happened. The mistake is the one made last night: a large correction, applied once, audited the following morning.
And when a night goes wrong, do nothing clever about it. Get up at the usual time, eat at the usual time, and let it be one entry among thirty.
There is an odd comfort in the body not paying close attention. It did not notice your best night. On this evidence it is not holding much against you for the worst one either.