Week 3 Lab: Two Instruments, One Night
This week
Your phone, a paper diary and your own memory will disagree about last night, and none of them is the referee.
Before you run this
- When to stop
- Stop recording and tell us if you find yourself checking your sleep data before you have got out of bed, or lying awake wondering what the device will say.
- What this is not
- Nothing in this week asks you to sleep less, sleep more, or go to bed at a different time. You are measuring the sleep you were already going to have.
- If this one isn't for you
- Compare a paper diary against your own morning recall, with no device at all. Write the two times into the diary as they happen, at lights out and at waking, then write the same two times again from memory that evening and report the gap. Two instruments, one night, and still no referee.
What you bring to the lab
The lesson below is how you get there. The lab is where you show it.
- you record the same night with two different instruments
- you can state which one you trusted before you looked, and why
- you report the disagreement rather than resolving it
Tonight you measure one night twice.
A phone or watch on one side. A paper diary filled in at the moment you turn the light off and the moment you get up, on the other. Same night, same person, two numbers.
They will not match. That is not a malfunction, and the week is not about deciding which one won.
There is no referee in this room#
In a sleep laboratory there is a third measure (electrodes, a technician, a recording of your actual brain) and it settles the argument. You do not have one. Neither do the millions of people making decisions about their lives based on a number a wristband produced by guessing from wrist movement.
So you have two instruments and no arbiter. The honest report is not “I slept 6 hours 12 minutes”. It is “one instrument said 6:12, the other said 7:05, and here is which one I believed before I looked.”
That last clause is the assessable part. Write it down first. It is remarkably easy to discover, after seeing the numbers, that you always trusted the one that agreed with you.
The risk in this particular week#
This is the first week with a real way to hurt yourself, and it is not sleep deprivation. It is the measuring.
In 2017 a group of sleep clinicians gave a name to something they kept seeing: patients arriving at clinic distressed about their sleep, whose sleep was fine, and whose actual complaint was their tracker. They called it orthosomnia: anxiety about sleeping correctly, produced by the instrument meant to help. Some of those patients were sleeping worse because of the data.
So: the stopping rule at the top of this page is not decoration. If you catch yourself reaching for the number before you have got out of bed, stop. You have found the effect this week is about, and you do not need to keep going to prove it.
In the lab#
We put every pair of numbers on the board and look at the spread. Then we ask a question that will run for the rest of the semester: given two instruments that disagree by fifty minutes, what size of effect could you possibly detect next week?
The answer is usually “a much bigger one than I was hoping to find”.
Evidence on file
- 669 people blinding not possible
Wrist actigraphy and self-report disagreed about the same nights. Measured sleep averaged 6.0 hours against 6.8 hours reported, the two correlated at only 0.47, and people sleeping five hours over-reported by about 1.2 hours.
Lauderdale DS, Knutson KL, Yan LL, Liu K, Rathouz PJ, Self-reported and measured sleep duration, Epidemiology 2008;19(6):838-845- History, not measured
Sleep clinicians proposed the term "orthosomnia" in 2017 for patients arriving at clinic whose presenting complaint was their own sleep-tracker data rather than their sleep.
Baron KG, Abbott S, Jao N, Manalo N, Mullen R, Orthosomnia: Are Some Patients Taking the Quantified Self Too Far?, Journal of Clinical Sleep Medicine 2017
Teaching team
- Dr Wren Halliday, convenor
- Priya Ramanathan, tutor