Working Memory Limit

Explicit thought needs a temporary workspace in which facts, constraints, and possible consequences remain available for comparison. That workspace is small. When a problem demands more live pieces than it can hold, some pieces fade while others enter, so the reasoning may feel continuous even though its inputs are no longer all present. This is one root of Bounded Rationality: the constraint is not merely missing information, but limited capacity to manipulate what you already know.

The useful response is compression. Chunking turns several details into one meaningful unit; reduction removes variables that do not affect the decision; externalization moves the remaining load onto paper or a screen. Each method preserves scarce working-memory slots for comparison and inference.

The supplied material gives an approximate capacity, not a precise universal cutoff for every person or task. It also does not show that every difficult decision is a memory problem: uncertainty, bad premises, or an unfamiliar domain can defeat reasoning even when only a few variables are involved.