Do you update your model with your actual experiences?


I do a lot to control my internal state

  • 8-hour sleep and daily routine ✔
  • Meditation and various cognitive habits ✔
  • Healthy food ✔
  • Sport ✔
  • No stimulants ✔
  • Rigorous work-life balance ✔
  • Therapy, friends, commitment devices, etc. ✔

Doing all this does indeed provide me with a higher baseline happiness1. Still, I’ve managed to fool myself into treating all living-in-the-moment advise as ‘a list of goals that undoubtedly will bring happiness when completed’. I was too dependent on ideas of what happiness should be and was ignoring what I really felt in the process.

There are times when, even after meticulous completion of todo-list items, I feel down for some reason. Maybe I wanted to accomplish more, maybe someone was a dick to me, maybe I was a dick to someone, maybe it’s just a mysterious hormone fluctuation.
Sometimes, all it takes to bring my mood up are compassion and a permission to feel bad.

There are also times when I did a lot of prep work, read articles by bestselling authors and listened to podcasts, made a model of myself and a plan how to improve my life; discussed and validated this plan with other people. But when I start executing, it feels like a grind, leaves me unmotivated with not much to show for my effort.

At that point, I usually concluded that my plan is flawed and I need to improve it. Or I’m not following the plan well enough because I’m a lazy asshole and I need to figure out how to motivate me first.

Recent therapy helped me to get out of my bubble a little and understand that I didn’t give enough credit to the direct experience and my actual (vs. best guess) feelings. It’s like I was learning to ride a bicycle only by reading some physics textbooks.

If you’re like me, the overanalysing type, it can take a while to understand. This doesn’t mean that planning and analysis are unnecessary, I will still do them. But I will err on the side of (validation through) action.

I can formulate the main thought as update your model with your actual experience; plan for raw experience inputs and corrections early and throughout the plan.



I’m applying this newly learned principle for my hobbies/vocations search: I’m searching for things to do besides my day job, and start with no idea of what I could be doing.

Before: elaborate planning, with weighted criteria list based on my best approximation of what I like; using said list as a scoring method for a vast list of hobbies2.
After: same plan scaffolding but the criteria list and their weights come from a direct experience of doing stuff.

If you’re familiar with software methodologies, I see this idea as applying the lean concept. It’s like iterating on an MVP instead of waiting for the waterfall to end.

This is all still very new, but it feels uncomfortable and right – the best kind of feeling, meaning I’m broadening my comfort zone. Anyway, I’ll see where this particular rabbit hole goes and will change the course if I don’t like it.


  1. I’ve logged my mood and activity during random parts of the day for more than a year now. This allows me to see correlations between my activities and mood. Maybe someday I will make a more thorough proof that there is indeed causation, but right now I’m comfortable being sure that, e.g. dancing and lifting weights reliably make me happy, and playing computer games makes me sad. 

  2. https://en.wikipedia.org/wiki/List_of_hobbies, http://www.notsoboringlife.com/list-of-hobbies/