← /notes

The Economy of Attention

Created Jun 6, 2026 game-designattentionstrategygames

In 1971, long before the feed, the economist and polymath Herbert Simon wrote down the problem the rest of us would spend the next half-century living. “A wealth of information,” he observed, “creates a poverty of attention.” Information consumes the attention of whoever receives it, so an abundance of information produces a scarcity of attention and a need to allocate it carefully among the too-many things competing to consume it. He won a Nobel for thinking like this about how people actually decide — the same mind that gave us satisficing, the idea that bounded creatures search for good-enough rather than best.

Watch a top StarCraft player and you are watching Simon’s sentence acted out at speed. At any instant a dozen things want attention — an army to control, workers to manage, an expansion to defend, a build order to keep. The player can only look at one. Time spent landing perfect commands in a battle is time the economy ran untended.

The game even measures it: APM, actions per minute. Elite players, especially the Korean professionals who turned StarCraft into a national sport, sustain hundreds of actions a minute, commonly in the 250–350 range and spiking past 400 in a fight. The numbers are real, but they mislead, because a lot of high APM is spam — repeated, redundant clicks. The deep skill was never raw speed.

It’s allocation. A pro doesn’t attend to everything; he attends to whatever currently has the highest payoff and lets the rest slide. He’ll let a few workers idle to land a critical attack, leave a fight on autopilot to set up the next expansion. Every decision is really the same question, asked many times a second: where is my attention worth the most right now, and what am I willing to let slip? Every glance has an opportunity cost — to look here is to not look there. Productivity, off the screen, is largely this same discipline of deliberate neglect, a subtractive skill of choosing on purpose what not to attend to.

There’s a structural reason it bites so hard: attention doesn’t parallelize. You can add workers to gather more minerals, but you cannot add a second self to watch a second battle. It’s a hard serial bottleneck. Which is why, in these games, the interface changes that reduce attention cost — control groups, rally points, automation — matter as much as any new unit. This is the logic of the jig: build the fixture once so every later operation costs nothing to supervise. Anything that lets you reach a goal with less attention frees the one beam you have for something else.

That generalizes to any system bottlenecked on a person. The highest-leverage move is rarely “try harder.” It’s usually “spend less attention on this.” Automate the routine, batch the interruptions, set the defaults so the standing work runs without supervision — not to do more in total, but to free the irreplaceable serial resource for the decisions only it can make.

So the master move in an attention economy, on the screen or off it, has two halves: cut the attention cost of everything routine, and then spend the freed attention only where its marginal value is highest, letting the rest deliberately slide. Simon saw the whole shape of it from 1971, before there was a feed engineered to fragment exactly the resource he named.

You will never attend to everything. Choosing what to ignore is the game.

Go Deeper

Books

  • The Attention Merchants by Tim Wu — A history of the industries built to harvest and resell human attention, from the penny press to the feed.
  • The Sciences of the Artificial by Herbert Simon — A book-length way into the mind that first framed attention as the scarce resource.