The night before last, I couldn’t stop reading The Giver. I knew I had to take my kids to school first thing the next morning, but I kept reading until almost 4 a.m. and forced myself to finish the whole thing. I got maybe two or three hours of sleep.
Yesterday, the miserable day after, I was completely exhausted. So I started thinking: why does the same thing keep happening over and over? What do I actually need to do to fix my refusal to sleep?
Why the immediate reward always wins
First, we all know about opportunity cost. If you do one thing, you can’t do another. Money is the most obvious example. If you only have $1,000 and spend it on an iPhone, you can’t invest that same money.
But staying up late has an opportunity cost too. If I stay up all night, my schedule and energy suffer for the entire week afterward. The strange part is that, to me, this enormous cost somehow matters less than “finding out how this book ends right now.”
At the root of it, I think there may be a disconnect between my present self and my future self.
What happens to future me seems to have nothing to do with present me. Today’s me is selfish. He wants to maximize his own benefit before he disappears. And he firmly believes that tomorrow’s me is very capable. Whatever problem I dump on him, he’ll figure it out himself.
The underestimated cost of staying up late
As far as today’s version of me is concerned, my brain is probably doing this calculation:
Whenever “finish now” is worth more than “finish tomorrow,” my brain chooses to stay up.
The biggest problem is that today’s me gives far too much weight to “finishing tonight,” while giving far too little weight to the cost of staying up and everything that comes after tomorrow.
In other words, a reward I can only get in the future is worth almost zero, and so is any suffering I’ll face in the future. From the present moment’s perspective, the best choice is obviously not to sleep and to keep doing whatever I want to do right now.
So what can I do to make my brain calculate the cost of staying up and the value of future rewards correctly? Maybe I can learn something from reinforcement learning.
Reinforcement learning adjusts its next choice based on the reward or penalty that follows an action. If my brain keeps getting the calculation wrong, I need to put the feedback it has been leaving out back into the equation.
Remember the cost of staying up
When I wake up exhausted, I need to spend some time really feeling that exhaustion and really feeling its consequences.
If there was something I could have done but was too tired to do, I should write that down too. I can’t only count the cost of the next day. I have to add up the costs that keep falling, like dominoes, all the way through day N.
Then I can tell yesterday’s me: you were heartless, childish, and reckless, and now I’m miserable because of you. But I forgive you. I’m going to make a different choice.
Empathize with my future selves
Humans are social animals. Today’s me is selfish because he can’t quite grasp that there are still so many future versions of me.
But if I imagine myself as an organization, where what I do today affects thousands and thousands of future selves, then it starts to feel at least a little more important.
In the best case, today’s me and all my future selves forgive that shortsighted version of me from the past. Instead of taking revenge, we make the best choice we can for the thousands and thousands of selves still to come.
Set limits for myself
If I still can’t control myself, then before I start doing something highly addictive, I have to set limits in advance.
For example, I can schedule my computer to shut down, or use my favorite sunrise alarm clock, which turns the light off on a timer.
Before I start reading a novel, I can set the light to dim in 30 minutes. If present me wants to turn it back on, he has to spend a little extra effort and openly defy the decision he made 30 minutes ago.
That little bit of extra friction won’t work every time. But at least it stops “keep reading” from being the unconscious default.
The past has already happened. I can’t change it. My future self, meanwhile, is like Schrödinger’s cat. Until the box of time opens, I don’t know whether he’s doing well or suffering.
Whether he gets to be happy is up to the person I am right now.
