Emergence

[Reivax]

After reading Michael Crichton's novel, Prey, I've become fascinated with the concept of 'emergence'. I’m sure some of you have read it.

Some common examples of emergence include:

• The swarming of bees (or any swarming organism for that matter): which appears to create a somewhat ‘conscious’ entity, but can be explained as an emergent characteristic caused by the domino like effect between individuals in the swarm/school/and perhaps herd?
  
• Neurons in the brain amassing together to create human consciousness.
  
• Atoms joining together to create: molecules > amino acids > proteins > cells etc.

Basically following the idea that "The whole is greater than the sum of its parts."

Is it reasonable to assume that this concept can be applied to things both now and in the future such as the:

• Growing global population; for example, to assume that a global consciousness is to emerge once the population reaches a certain critical point. (Sounds pretty New Age, I know )
  
• Clusters of galaxies: for instance the analogy between atoms/galaxies or atoms/star systems & planetary systems?

Thoughts?

 

[roger]

You'll note that in the working systems you mentioned there are very strong rules and limitations on the interaction of the parts. Bird swarms follow very simple and unvarying rules. Neurons have very specific functionality. Etc. I submit, without proof, that you need such limitations to build a more robust system on top of the underlying parts. Any system without such constraints do not form emergent behavior, that I can think of.

If you look at the more formal definitions of emergence, they require things like coherence - maintainance of the system for a period of time. We can see crowd behavior in people at a concert, but then the concert ends and people disperse. No emergence there. It also requires causal connections - galaxies/stars can move each other a bit through gravitation, but nothing else is going on - there are no inputs except gravity and position.

 

[Reivax]

So essentially the complexity, of say the human consciousness effectively eliminates any sort of consistency and thus the possibility of emergent characteristics?

I'm currently thinking of any examples that go against this idea, but I think you're right in saying that there has to be strict limitations in place, otherwise it won't happen.

Does the same apply for artificial structures such as the internet. Would you consider the internet as being an emergent property which stems from more complex parts, or are the rules and limitations strict enough?

Thanks for your input.

 

[madurobob]

Have you read "Godel, Escher, Bach" by Hofstadter?

Its been many years since I have, but I remember an interesting discussion in it about ant colonies and an analogy to the individual cells of the human brain. I'll have to go find the book and re-read, but a theme of the book is how "thinking" comes about from many seemingly random neurological processes. IIRC, at one point one of the book's characters carries on a conversation with the ant colony while also noting that trying to communicate with any given ant would be impossible.

 

[Reivax]

No, I have not.

It sounds very interesting, I'll be sure to add it to the list and check it out.

Thanks for the recommendation!

 

[Wowbagger]

After reading Michael Crichton's novel, Prey, I've become fascinated with the concept of 'emergence'. I’m sure some of you have read it.

I read it. I thought it was a huge wasted opportunity: You have a very unique and (potentially) frightening concept for a monster, yet it didn't really get to do very many interesting things, at least not in any way that mattered.

Growing global population; for example, to assume that a global consciousness is to emerge once the population reaches a certain critical point. (Sounds pretty New Age, I know )

It would not be the size of the population that matters. But how its members connect to each other.

There is emergent behavior associated with crowds of people. There is no reason why we could not observe some form of emergent behavior in the gobal population. But, in this case, it is the "protocols" that matter more than the number of people.

Clusters of galaxies: for instance the analogy between atoms/galaxies or atoms/star systems & planetary systems?

Have you read about galaxy filaments?
http://en.wikipedia.org/wiki/Galaxy_filament

Not sure if there is much in the realm of "behavior" associated with filaments. But, something seems to be emerging from clusters of galaxies.

Its been many years since I have, but I remember an interesting discussion in it about ant colonies and an analogy to the individual cells of the human brain.

Probably reinforced paths.

When ants travel, they leave behind trails of scent marks. The more popular the trail, the stronger the mark. Over time, the more efficient routes are "carved" in this manner. Though, they could also be re-routed over time, when necessary.

Neurons work in a similar manner. Paths are re-inforced with usage. But, there is still flexability to re-route mental processes over time.

 

[Mark6]

It also requires causal connections - galaxies/stars can move each other a bit through gravitation, but nothing else is going on - there are no inputs except gravity and position.

Not sure what you mean by "causal connections", but all examples in OP other that astronomical bodies involve feedback loops -- when birds in a flock get too far from each other they tend to move closer (swarming behavior), but when they get too close they move apart (avoidance behavior). Most if not all of emergent behavior involves these kind of competing vectors, as the animals (or whatever) follow the vector sum. I don't think anything competes with gravity as far as galaxies/stars are concerned.

 

[Dinwar]

The outward push from nova and supernova shockwaves would counter gravity to some extent, in that they'd provide some push outward away from the center of the galaxy (the shockwave would be roughly spherical, meaning one part is going to face the center of the galaxy and the other away from it). There's also inertia--galaxies tend to spin, which creats angular momentum, which tends to push things away from the center (I know, not EXACTLY away from the center, it's actually tangential to the motion, but it's still a force pushing things away from the galactic center, while gravity pushes them into the galactic center).

 

[Aepervius]

It is relatively to get flock/swarm behavior with very simple rules (there are resource for that one the net).

Build vector for each part of the flock as follow :
* go away from the next neighbor (for each neighbor make a vector going away from it , weighted by the invervse of the distance : the further away the lowest weight)
* add part of the vector representing the average speed of all neighbor
* add part of a vector going toward the barycenter
And poof, you have a flocking swarm behavior.

There are many algorithm on the net, and a few simulation showing such behavior.

So from very few rules and limitation, you can actually have a complex result.