Driving Cellular Automata

A research introduction on using a genetic programming language to build a cellular automaton model of multicellular development.

Project overview and theme

Framing DNA-like policies as evolvable control programs.

• This is another research introduction, dictated for later rewriting and polishing.
• The subject is using my genetic programming language to build and then write up a cellular automata model.
• Working title: Life, but not as we know it.
• Prior idea: thinking of DNA as encoding a policy; using a “hand language” can provide an evolvable system for guiding the behavior of a simple agent.
• Here, the analogy is with a cell.

Context: multicellular development (evogenesis)

Explaining the control system in a multicellular setting.

• The target of explanation on this website is multicellular development (or evogenesis).
• Key question: how should we understand the role that this control system plays in that context?

Modeling approach: embedding a cell among identical cells

Cells interact locally; neighbors become the environment.

• Proceed by embedding one cell amongst copies of itself.
• A cell acts directly with its environment, and its actions change the environment around it.
• In this case, the environment is other cells running the same control program.
• Outputs from surrounding cells provide environmental cues that shape each cell’s response.

Cellular automata and the Game of Life

Using a simple automaton framework to model local rules.

• The simplest modeling format for this is a cellular automaton.
• The most well-known cellular automaton is Conway’s Game of Life.
• Insert a short explanation of the Game of Life here.
• Next paragraph: emphasize that the rules guiding each cell in the Game of Life are rules that respond to neighboring states.