Do you have any thoughts about emergence as it relates to consciousness? Or, maybe better, would emergence as it relates to Conway's Game of Life give any insight to how consciousness might emerge? I am not familiar, and will look at the link you provided, so I defer to you.
I think GoL can potentially give insight into the difficulty of accounting for consciousness as brain activity.
GoL involves a grid of binary cells which can be either 'on' or 'off' or ('alive' or 'dead'). This is usually indicated visually by a cell being black when 'on' and blank/white when 'off'.
The on or off state of each cell depends on the state of the 8 adjacent 'neighbour' cells according to a few simple rules:
- Any live cell with two or three live neighbors survives.
- Any dead cell with three live neighbours becomes a live cell.
- All other live cells die in the next generation. Similarly, all other dead cells stay dead.
The game proceeds by iterating over the grid cell by cell and setting its state according to the above rules. A generation is an iteration of the whole grid. For obvious reasons, a computer is necessary to perform this process for more than trivially simple arrangements and short runs.
The interest in GoL depends on the initial state of the grid cells. When a decent sized grid (say, 100 x 100 or more), with cells set on or off at random, is iterated repeatedly, characteristic patterns of activity appear where live cells are clustered together. These patterns have been given names and can be generated at will by setting an initial cluster of live cells in a given configuration. What's more, although the cells are static, fixed in place on the grid, some of these patterns move across the grid in characteristic ways, and when they meet other cell patterns, they interact with them in characteristic ways. See
GoL Demo.
The relevance to emergence is that the patterns and their activity are not predictable from the initial configuration of the cells. You can understand the rules and the binary nature of the individual cells, but unless you apply the rules over several iterations, you don't know what patterns will result from a given starting configuration. IOW, the patterns are emergent.
A whole new language and set of concepts can be used to describe these patterns, their behaviours, and their interactions, completely independent of the language and concepts used to describe the cells and their behaviour in the grid. The cells in the grid have become simply the substrate on which these interacting patterns of activity appear. A reductionist approach won't be informative where those interacting patterns are concerned - not least because what they do doesn't depend on the properties of the grid cells themselves but on their initial state configuration.
If you know the language and how to use it, you can make
self-replicating pattern structures, or
emulate the GoL itself:
So far so cool and geeky...
The discovery that I thought was particularly relevant to the question of consciousness and the brain, was that the interactions of these patterns can be made to process information - if you set the initial state of the grid appropriately, and decide on your representations of inputs and outputs, you can make the interacting patterns in GoL do anything a computer can do - as in
a programmable computer:
in fact, you can emulate a
Universal Turing Machine, the fundamental programmable computer:
So the patterns themselves can be arranged to act like a computer processor and can process any inputs you provide according to a set of instructions you provide (as patterns of cells).
Now one can make an (admittedly very crude) analogy between GoL and the brain, between the cells of the automaton and the neurons in the brain. The neurons are complex, of many different types, with many different functional rules, with complex, dynamic connectivity, and organised into many specialised structures, but it's not hard to imagine that interacting patterns of activity between neurons and networks of neurons can represent information processing orders of magnitude more complex and sophisticated than those in GoL - and correspondingly more difficult to elucidate beyond a crude functional analysis of the specialised structures.