Recognition

A simulation exploring various gene-based models of recognition of friend or foe between cells. Each model represents a distinct combination of a cell's genes involved in recognizing its neighbours.

The interaction between two cells determines whether one cell can recognize the other as a friend or foe. Recognition is evaluated through gene comparison. If the genes are sufficiently similar, the other cell is recognized as a friend. Conversely, if the genes are sufficiently opposite, the other cell is recognized as a foe.

Gene comparisons use a lenient approach: for similarity-based comparisons, the probability of a match decreases as genetic differences increase. Conversely, for opposite-based comparisons, the probability of a match decreases as genetic differences decrease. Based on the classification of the neighbour, fitness may be gained or lost.

Cell Attributes and Life Cycle

Cell Properties

Four Circular Genes (A, B, C, D, Z)
Fitness (0-255)
Age (+1 with each cycle)

Life Cycle

Fitness +1 for each neighbouring empty cell.
Fitness gain or loss from each neighbour depending on selected model (see below) and corresponding gene comparison matches.
At fitness >= 255 fitness is reduced to 255 minus random value in between 0 and 60.
At fitness >= 128 cell tries to reproduce, dividing fitness equally to offspring and parent.
Reproduction may fail if several cells try to divide into an empty cell. Winner is the offspring with the highest fitness.
During reproduction, there is a 1 in 10,000 chance that a cell mutates a gene, with a mutation range of ±64.
At fitness <= 0 or age > 40, the cell dies.

Models

Each model code represents a gene-matching pattern between cells, where single letters (A, B, C, D) indicate genes compared sequentially, and the "-OR" suffix means any single match suffices instead of requiring all matches. For example, "AA" matches if gene A matches the neighbour's A, while "ABCA-OR" matches if A matches B, B matches C, or C matches A.