the speed of execution, for example, to leave more time to observe details as
necessary. The model world is shown on the right-hand sight. In this particu-
lar case, the patches of the world correspond to individual households in
Mediterranean antiquity and can have two states. Initially they are white
indicating that the corresponding households are“pagan,”that is, they have
not yet become followers of the Christ movement. During the execution of the
model, they can turn green, indicating that the households have been con-
verted. After pushing the“setup”button, the patch at the center of the world
turns green (at least at the initial settings of the model). If we now push the
“go”button, the movement starts to spread until all patches turn green.
A counter at the top left corner of the model world shows the“ticks”that
have passed, that is, the number of times the go procedure has been carried
out. We are interpreting a“tick”as a day, where the correspondence to days in
the real world is only symbolic and does not establish any prediction by itself.
Another counter on the left shows the number of the followers (to be more
precise, the number of households converted, where we can heuristically think
of each patch represented by the head of the household). The plot diagram
at the left bottom part of the Interface shows the number of days elapsed
on the x (horizontal) axis and the number of followers on any given day on the
y (vertical) axis.
In the upper region of the left-hand side of the Interface,five parameters can
be set. The“mission goal”parameter tells the model what percentage of all
households have to be converted before the execution of the go procedure
stops. The“conversion rate”parameter sets the probability of conversion
given that a household communicates with another household. In real life,
this probability is relatively low. We can assume that people had to be exposed
to some religious innovation repeatedly before deciding to adopt it. (Again, we
assume that households became followers in an all-or-none fashion, that is, all
members of the household were converted simultaneously.) An important
parameter is the learning strategy used when deciding whether to adopt the
new religion. There are three possible learning strategies. (1) The setting“any”
makes households convert (with the probability dictated by the“conversion-
rate”) any time one of their eight neighbors (to the north, south, west, east,
north-west, north-east, south-west, or south-east) becomes a follower. Thus if
the conversion-rate is 0.5, as in the initial settings, all eight neighbors of the
initially green patch will have a 50 percent probability to convert on thefirst
day. (2) The learning strategy called“probabilistic”means that a household
checks the status of all of its neighbors and the chance of converting increases
with the ratio of the neighbors that are already converted. For example, if two
out of eight neighbors are already converted, the chance of the original
household to convert is 25 percent; combined with the conversion-rate of
50 percent, this gives an overall chance of conversion of 12.5 percent at each
tick in this example.
194 Cognitive Science and the New Testament