Varying the initial spatial distribution of agents changes the global form that emerges from identical local rules

When agent-based generative systems follow local rules, the global pattern that emerges depends strongly on the initial positions of the agents — even when every rule stays identical. Seeding agents in a uniform distribution produces a different overall form than seeding them on a circle’s circumference or along a center line. This gives the artist a compositional handle that operates above the rule level: you don’t change what agents do, you change where they start, and the macro shape changes. It is a form of top-down guidance inside a bottom-up system, letting the artist coax an emergent process toward a desired global structure without breaking emergence.

Examples

Reas’s Articulate-based system: ‘Here they start at the circumference of a circle. Here from a line in the center of the screen. Here from the top and bottom’ — same behavioral rules, three visually distinct macro-forms.

Assessment

Run an agent simulation with three different initial distributions (uniform, ring, line). Sketch or describe the global form that emerges from each and explain why the distribution matters.