Trevor Bedford attended the University of Chicago, where he received a B.A. in the Biological Sciences in June of 2002. Since then, he has become very interested in studying the origin of biological complexity. He believes that NKS methodology and intuition provide powerful tools to examine this question. He hopes to integrate such NKS approaches into the Ph.D. research that he’s currently conducting at Harvard University.
Project: Multi-Agent Interaction Modeling
The final goal of this project is to create an abstract system that is subject to the effects of natural selection. I’m attempting to implement this system in a universal automaton capable of supporting persistent localized structures (such as elementary rule 110). This automaton and the rule behind it represent the physics behind the system. I hope to find a discrete group of cells (which we’ll call an organism for lack of a better word) that is capable of robustly computing its own replication. In this way, the organism’s replication is not arbitrary; it lies entirely within the bounds of the cellular automaton’s physics. This concept relies heavily on the Principal of Computational Equivalence, in that the physics of my abstract system should be capable of producing just as much complexity in its behavior as we observe in the real world. If replication is not perfect and these replicators do not exist on an infinite background, then the system will have everything it needs to undergo natural selection: heredity, variation and competition.
Favorite Two-Color, Radius-2 Rule
Rule chosen: 3327554226
I liked this rule because, starting from a single black cell, it is able to emulate a slightly skewed version of elementary rule 110. However, with other initial conditions, two separate patterns emerge. The original 110 emulation is still present, along with dark branching structures that move to the right. And thus the standard localized structures in rule 110 are able to interact with these new structures in interesting ways.