Learning and growth
Herbert Simon
Bounded rationality
Herbert Simon (United States · 1916-2001) — Bounded rationality.
What actually happened?
In 1935, still an undergraduate at Chicago, he went home to Milwaukee to study how the city's public recreation programme divided its budget. The money had to be split between keeping up the playgrounds and paying the leaders who ran activities for children. The textbook said to split it until the last dollar did equal good on either side. Nobody he met thought that way. The people who looked after the grounds wanted more for the grounds; the people who ran activities wanted more leaders. Neither side was confused. Each saw the problem from inside its own patch. He later traced his whole subject back to that study.
He replaced the rational man
Economics assumed a chooser who knows every option, can work out every consequence, and picks the best. Simon's people have incomplete information, limited computing power and limited time, so their reasoning works inside those bounds. He called it bounded rationality: reasoning has a cost.
A plan that three hours of meetings cannot settle is rarely stuck because somebody is being unreasonable.
Good enough, then stop
Since the best cannot be computed, people do something else. They set an acceptable line, look at options one at a time, take the first that clears it, and raise or lower the line as the search proves easy or hard. He called it satisficing.
Choosing a builder, the first one whose price and references both clear the line spares you meeting a sixth.
Where you stand is what you see
Milwaukee left him a second idea. A person deciding inside an organisation takes the angle of the unit he belongs to. Administrative Behavior gives this identification its own treatment: it spares people from weighing everything afresh, and hides whatever lies outside their patch.
When sales and engineering fight over the budget, neither side is usually playing games.
A wealth of information, a poverty of attention
In 1971 he wrote the lines quoted ever since: information consumes the attention of its recipients, so a wealth of information creates a poverty of attention. The question for any new system is not how much it tells you, but how much it keeps away.
Someone getting hundreds of group messages a day gains more by muting them than by learning to skim.
He tested thinking on a computer
In 1955 and 1956 he, Newell and Cliff Shaw wrote a program called the Logic Theorist and set it to prove theorems from Principia Mathematica. It did not try every path, only the few that looked promising. People think the same way, he held: by selective search, not by exhausting the options.
An experienced engineer chasing a fault checks the two likeliest places first, not line one onwards.
How do I use it today?
If you are stuck between two or three decent options, don't go looking for a fourth today. Write down the three things you cannot do without, check each option against them, and take the first that clears all three. If none does, the line is too high: move it and check again. Keep a second job for your reading: go back over what you read this week, count the items that actually changed a decision, and drop half of the sources that supplied the rest. He used one test for both: your computing power and your attention are finite, so don't spend them on the chance that something better is still coming.
Deep read
Read alongside
Further
Lines to keep
You are not looking for the best. You want the first good enough.
Reasoning costs computing power, and nobody has much.
More information makes attention the thing that runs short.