Voting Patterns

We like to think that our votes are the product of careful reasoning about the issues of the day. It's reassuring to learn that most of our neighbors vote the way we do. The few who don't are obviously idiots and shouldn't be invited to parties. Those vast blocks of idiots in other states who vote for the other party are worrisome. What could they be thinking? The question is easy to answer if we are willing to admit that our political opinions are influenced by the opinions of our neighbors. After all, we all like to get invited to parties.

We can demonstrate this using the Game of Life model developed earlier. We now interpret cells to be households. The state of the household indicates the political party affiliation of the owner of the house:

1 = Democrat
0 = Republican

The ambience of a household is the number of Democratic neighbors.

The set T01 characterizes strong Democratic neighborhoods. A Republican living in such a neighborhood would be inclined to switch parties.

The set T10 characterizes weak Democratic neighborhoods. A Democrat living in such a neighborhood would be inclined to switch parties.

For example, the following assignments say that a strong democratic neighborhood consists of four or more democrats, while a weak democratic neighborhood consists of four or fewer Democrats:

set T10 [0 1 2 3 4]
set T01 [4 5 6 7 8]

Note that a neighborhood consisting of four democrats (and therefore four Republicans) is both weak and strong.

Initially, the distribution of democrats (white) and Republicans (black) is random:

After twenty or so rounds of the observe-update cycle Republican and Democratic regions have formed:

Interestingly, the boundaries between Republican and democratic regions oscillate between the two parties on almost every cycle.

Experiment with different definitions of T01 and T10. Also try to make the party switch probabilistic. In other words, interpret the ambience as the probability that the occupant will switch party affiliations.