Figures 8.1 and 8.2 One-dimensional CA

Again, the examples in the book were made using a NetLogo model based on this one in the NetLogo model library. This page provides R code instead.

Model code

The 1D elementary CA model code is in the block below. It returns a matrix of cell states, with each system state a row in the matrix.

The key function is one_d_ca(rule, n, density, t), which returns a t \(\times\) n matrix, with initial density of 1’s given by the density parameter, and the update rule being the Wolfram code specified by rule.

library(dplyr)

# initialise random state, specified size and density
init_state <- function(n = 50, density = 0.5) {
  as.numeric(runif(n) < density)
}

# pre- and post-fix the last and first elements of the state
wrap_state <- function(x) {
  c(tail(x, 1), x, x[1])
}

# binary vector to decimal integer (most significant bits first)
as_decimal <- function(b) {
  sum((2 ^ ((length(b) - 1):0)) * b)
}

# convert decimal integer to vector of bits (most significant first)
as_binary <- function(n, n_bits = 8) {
  result <- c()
  for (i in 1:n_bits) {
    result <- c(n %% 2, result)
    n <- n %/% 2
  }
  result
}

rule_description <- function(rule_num) {
  paste(rule_num, 
        paste(as_binary(rule_num), collapse = ""), 
        sep = ": ")
}

# takes a matrix of CA states (each row a timestep) and appends
# a new row for the next system state
generation <- function(states, rule) {
  state <- tail(states, 1) |> c()
  w_state <- wrap_state(state)
  result <- c()
  for (i in 1:length(state)) {
    nbhd <- w_state[seq(i, i + 2)]
    result <- c(result, rule[as_decimal(nbhd) + 1])
  }
  matrix(c(states |> t() |> c(), result), nc = ncol(states), byrow = TRUE)
}

# generate a matrix of cell states where each row is the system state
one_d_ca <- function(rule = 30, n = 64, density = 0.5, t = 192) {
  s <- init_state(n, density)
  ss <- matrix(s, nr = 1)
  for (i in 2:t) {
    ss <- generation(ss, rev(as_binary(rule)))
  }
  ss
}

Making plots

Figure 8.2

The second figure is a little easier since there is only one of them…

The matrix format is convenient for computation, and fine for plotting a single model result but that has to be converted to a dataframe to be plotted using ggplot2. So here is the pure R version of Figure 8.2, but with time running from top to bottom, more cells, and more time.

library(ggplot2)
library(reshape2)

one_d_ca(rule = 110, n = 192, density = 0.5, t = 384) |>
  reshape2::melt(c("t", "x"), value.name = "state") |>
  mutate(state = as.logical(state)) |>
  ggplot() +
    geom_raster(aes(x = x, y = -t, fill = state)) + 
    scale_fill_brewer(palette = "Paired") +
    coord_equal() +
    theme_void()

Figure 8.1

And here are the results for all Wolfram’s elementary 1D CA.

# convenience function to return a dataframe instead of a matrix
get_df <- function(rule, n = 64, density = 0.5, t = 192) {
  one_d_ca(rule = rule, n = n, density = density, t = t) |>
  reshape2::melt(c("t", "x"), value.name = "state") |>
  mutate(state = as.logical(state),
         rule = rule)
}

# The 32 rules that match Wolfram's selection criteria
rules <- c(  0,   4,  18,  22,  32,  36,  50,  54,
            72,  76,  90,  94, 104, 108, 122, 126,
           128, 132, 146, 150, 160, 164, 178, 182,
           200, 204, 218, 222, 232, 236, 250, 254)

df <- get_df(rules[1], t = 128)

for (rule in rules[2:32]) {
  df <- bind_rows(df, get_df(rule, t = 128))
}

# give the rules a better label
df$rule <- factor(df$rule, levels = rules,
                  labels = lapply(rules, rule_description))

ggplot(df) +
  geom_raster(aes(x = t, y = x, fill = state)) + 
  scale_fill_brewer(palette = "Paired") +
  coord_equal() +
  guides(fill = "none") +
  facet_wrap( ~ rule, ncol = 4) +
  theme_void()

# License (MIT)
#
# Copyright (c) 2023 David O'Sullivan
#
# Permission is hereby granted, free of charge, to any person
# obtaining a copy of this software and associated documentation
# files (the "Software"), to deal in the Software without restriction,
# including without limitation the rights to use, copy, modify, merge,
# publish, distribute, sublicense, and/or sell copies of the Software,
# and to  permit persons to whom the Software is furnished to do so,
# subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included
# in all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS
# OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
# THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
# FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
# DEALINGS IN THE SOFTWARE.

© 2023-24 David O’Sullivan