Clerestory

Dry November: Day 6

November 06, 2019

Today I’m thinking about a friend’s article on resourcefulness, which is great and I recommend reading it. In it, he argues that there are various ways to train one’s own decision-making and thinking.

I’m especially interested in the claim that roguelike computer games might be good for you. I agree with his sense that most games are something like trigger/response, and hook you via dopamine, without engaging much creativity.

A corollary to this could be that challenging games like roguelikes provide their pleasure primarily via serotonin, like other effortful activities (exercise, meditation). If that’s true, they might improve learning. I had forgotten (or perhaps never known) that Sebastian Marshall wrote a book on this topic, which I’m resisting the temptation to read today due to an already encumbered and delinquent list.

At first I was resistant to the idea that these games might be good for you, for two reasons:

  1. I’ve spent plenty of time playing roguelikes, and don’t need excuses to sink any more hours in them.1
  2. It seemed like the choices in roguelikes are artificial and constrained, which doesn’t seem to map well onto decisions in real life.

Then again, they do emphasise lateral strategies, probabilistic thinking, and careful planning. They also readily produce flow states, and my friend believes that familiarity with flow states is itself helpful in producing this kind of concentration in the pursuit of other goals.

I also started thinking about the fact that choices in roguelikes are far less constrained than they are in that more familiar turn-based game, i.e., chess. Especially as the game progresses, you might have hundreds of options open during a given turn. You also take a lot more turns. Chess lasts roughly 40-80 turns, with a long game lasting maybe 300, whereas the average NetHack ascension takes 50,000 turns, with a record-settingly short game lasting 2,130.

This question has also made me wonder whether chess improves decision making.

Certainly I have the impression that chess is good for you. On the other hand, I’m not aware of chess pros being great at life in other regards (they may well be excellent, but I’m just not aware of it). Conversely I’ve never heard non-chess-high achievers, who routinely recommend things like meditation and exercise, recommending long hours of chess. I can think of a few who enjoy it, but I can’t recall any who are particularly evangelical about it.

It got me thinking about constraints in games in general. My intuition was that less constrained games must be better. Though I’ve never played either, I would just conceptually guess that Minecraft is better for you than Candy Crush. But is this true?


  1. I must have started playing NetHack around 2004, when I first installed Linux. I was attracted to it because of its reputed difficulty. I ascended eventually in AceHack, a minor interface tweak to NetHack. Then I committed to beating the vanilla game. My only NetHack ascension was in 2012, meaning that it took me roughly eight years of (casual off and on) playing. I’m half-ashamed and half-proud to say that I did it in the office, and then had to spend a considerable amount of time explaining why I was so excited to colleagues, who could not have cared less.


I'm Bryan Kam. I'm thinking about complexity and selfhood. Please sign up to my newsletter or see more here.