Why do swarms build complex brains but keep their movements simple?

Researchers evolved neural controllers for robot swarms under realistic constraints and found that the most coordinated collectives develop two surprising traits simultaneously: complex hidden-layer processing but linear, simple outputs. This split mirrors human problem-solving—we think through messy internal models but act on clear rules. The finding suggests swarms optimize by doing the hard work internally while keeping motor commands direct.