After thinking about this over the night, cellular automata modelling can be used to graphically display more complex modelling. The rules that determine which cells turn on, or off, or toggle each turn can be thought of as biological processes, or actions of individual people, etc.
For example, let's say I believe that voting doesn't matter, because my vote counts only if it's the tie-breaker. (This assumes nobody knows the vote counts until after the voting has ended). Normally, that would have no effect on the voting of the rest of the population.
But what if I spread that idea? It may induce others to not vote, but more importantly they have a chance to spread the idea.
And so if I care about the voting process, and I want people to vote, I had better not tell anyone about my belief.
Furthermore, the rules can be more complex if each cell represents more than an on/off state. What if each cell is a hex pair, giving you 256 possible states in each cell? You can encode one piece of information with 256 possible choices, or maybe two pieces of info with 16 choices each. If you're clever or if you track this using binary instead of the hex character themselves, you can encode many on/off states in each cell.
Perhaps you could use this to model the spread of disease in a population. But to make good rules, you'd need good data on how the disease actually spreads. But you could look at the disease cases, and test out a huge range of different rule sets, until you discover a rule set that works to model the real world events.
Perhaps economic modeling, like trying to predict stock market outcomes.
Certainly it would have a use in modeling very low-level physics, but you have to count each cell as an incredibly small space, and have a massive swath of cells, otherwise it's not granular enough.
But maybe I don't know what I'm talking about. If I were you, I'd choose something with more obvious applications to write about.
Author
Topic: Cellular Automata (Read 4773 times)