Free vs. collective will

[aggle-rithm]

This topic probably belongs in Science, but since free will is such a hot issue...


I saw a documentary recently about army ants. These ants are nomadic, crowding together into a dense ball (with the queen at the center) rather than digging a nest. The queen, eggs, larvae, pupae, etc. stay in the ball while the others scout for food in the manner that makes army ants so darned endearing.

What caught my attention was the description of the colony's movement from place to place. At some point, the colony "decides" to pull up camp and go somewhere else. There is no discussion, no vote, no explaining the plan to other ants. They simply all begin, at the same time, the process of moving the colony. The mechanism for how this decision is made is unknown, but it seems likely that if enough ants sense that food in the area is running low, a sort of critical mass is reached, and the colony exercises a collective will to move on.

Insect colonies can be seen as a crude analogy of the human mind. The colony is an emergent property of a large group of individual ants, just as the mind is an emergent property of the huge tangle of neurons in the functioning brain.

I realize there are differing viewpoints on this, but this analogy has always made sense to me. If it is valid, then the collective will demonstrated by an ant colony can be seen as analogous to a human being's free will. Assuming the colony's decision to move isn't simply made by the queen (which seems unlikely, considering her small brain), then this exercise of collective will is de-centralized; there is no focal point of authority (such as a "self") that drives the decision. Could free will, exercised by individual humans, be the same? Could it simply be the resolution of some potentiallity produced by many parallel processes in the brain, rather than a single, unified "command" coming from the ego? Could it be that the ego is the last member of the "mind colony" to find out about the decision?

And is the rhetorical question an overused verbal device?

[aggle-rithm]

Bump, because I'd still like to hear people's thoughts on this really, really ancient thread.

[blobru]

OP could almost be a snippet from Marvin Minsky's "Society of Mind" (though did he use that many rhetorical questions?) ;) I think the ant colony analogy is a great way to model mind and free will as emergence.

[Dilb]

Insect colonies can be seen as a crude analogy of the human mind. The colony is an emergent property of a large group of individual ants, just as the mind is an emergent property of the huge tangle of neurons in the functioning brain.

I always look at ants in the opposite way: an ant colony is a single organism the same way that a 'colony' of human cells is a single organism. Ants just aren't stuck together.

In any case, this really doesn't do anything for free will. Determinists (as far as I can find, anyway) admit we have the illusion of free will, which is why we have the concept of free will at all. Either the parts work together in the same way every time (allowing for enviromental factors), in which case free will does not really exist; or they sometimes work together differently, in which case free will is whatever causes that difference.

[Squeegee Beckenheim]

Either the parts work together in the same way every time (allowing for enviromental factors), in which case free will does not really exist; or they sometimes work together differently, in which case free will is whatever causes that difference.

Not necessarily. If all environmental factors are the same then there could still be a different process in the brain which could be described as other than free will. Let's say that the ant colony view of the brain is right. Let's also say that the idea of "critical mass" is right, in that when x number of ants think it's time to leave (and, say, indicate that by excreting a certain chemical - let's call it chemical L) then the colony will leave.

However, instead of each individual ant, say, deciding it's time to leave if they find fewer than 10 pieces of food on a foraging trip, they have a probability associated with it. So if they find 10 pieces of food there will be a 0% probability of them excreting chemical L. If they find 0 pieces of food there will be a 100% probability of them excreting chemical L. If they find 9 pieces, it'll be 10%, 50% for 5, and so on. Now you can have exactly the same environmental factors - each individual ant finds exactly the same amount of food - and yet it still potentially leads to two completely different decisions.

So, if you have a view of the brain where free will is an illusion with which we tell ourselves we're making decisions, when they're actually being made by unconscious automatic processes within the brain, then you can have a scenario with identical environmental factors which can still have 2 (or more) different outcomes - so long as you treat the firing of a neuron as an event with a probability attached, rather than as an either/or.

I've no idea how true any of this is, as I've not read enough research on the subject, but I do believe that brain processes being influenced by quantum events is something that has, at least, been the subject of some scientific debate.

[Dilb]

However, instead of each individual ant, say, deciding it's time to leave if they find fewer than 10 pieces of food on a foraging trip, they have a probability associated with it. So if they find 10 pieces of food there will be a 0% probability of them excreting chemical L. If they find 0 pieces of food there will be a 100% probability of them excreting chemical L. If they find 9 pieces, it'll be 10%, 50% for 5, and so on. Now you can have exactly the same environmental factors - each individual ant finds exactly the same amount of food - and yet it still potentially leads to two completely different decisions.

That's irrelevant: the ants are still behaiving in a consistent way each time. Either it's truely random, in which case free will still doesn't exist, or free will affects those probabilities (and free will is a very odd and limited thing, in that case).

Admittedly it's not determinism, but as far as I know there isn't a good term for a system that's deterministic except for some entirely random processes. This gets brought up every time someone suggests that quantum mechanics allows free will to exist.

[Lowpro]

Well a fair question. The decision itself to move the colony; does it get "decided" on through a process of positive feedback?

Basically one ant starves, the instinct to move kicks in, however it's not enough to initiate movement.

5 ants get the instinct, they move, colony does not move, HOWEVER their instinct to move could cause a positive feedback which kicks the instinct into the other ants, initiating the migration.

I'm not sure if this is possible, but doesn't this throw a wrench fly in the face of the conventional idea that it's the small parts of the whole brain which consciousness (my definition of will, free or otherwise) derives from? To be conscious is nothing more than the summation of a positive feedback system?

It's a question, not any real evidence-based idea.

[Squeegee Beckenheim]

That's irrelevant: the ants are still behaiving in a consistent way each time.

They're not, actually. If they find 5 pieces of food, then they're behaving in a consistent way 50% of the time.

[nzric]

This theory is described in much more detail in Gödel, Escher, Bach: An Eternal Golden Braid, by Douglas Hofstadter. He talks a lot about the ant metaphor and the mind.

Highly recommended

[aggle-rithm]

OP could almost be a snippet from Marvin Minsky's "Society of Mind" (though did he use that many rhetorical questions?) ;)

I'll have to check it out, thanks.

[aggle-rithm]

This theory is described in much more detail in Gödel, Escher, Bach: An Eternal Golden Braid, by Douglas Hofstadter. He talks a lot about the ant metaphor and the mind.

Highly recommended

I may actually have that book. Someone gave it to me as a gift and I never read it.

[aggle-rithm]

Not necessarily. If all environmental factors are the same then there could still be a different process in the brain which could be described as other than free will. Let's say that the ant colony view of the brain is right. Let's also say that the idea of "critical mass" is right, in that when x number of ants think it's time to leave (and, say, indicate that by excreting a certain chemical - let's call it chemical L) then the colony will leave.

However, instead of each individual ant, say, deciding it's time to leave if they find fewer than 10 pieces of food on a foraging trip, they have a probability associated with it. So if they find 10 pieces of food there will be a 0% probability of them excreting chemical L. If they find 0 pieces of food there will be a 100% probability of them excreting chemical L. If they find 9 pieces, it'll be 10%, 50% for 5, and so on. Now you can have exactly the same environmental factors - each individual ant finds exactly the same amount of food - and yet it still potentially leads to two completely different decisions.


Maybe it's somewhat like the synchronized flight of birds or movement of schools of fish. There are just a few "rules" in the brain of each animal, and if they all follow the rules then there is a large-scale effect.

Suppose each ant has two rules:

1. I will become restless when I grow hungry
2. If (1) is in effect, and other ants around me become restless, then I pick up and move.

Of course, the behavior is a little more complex than that, so there are probably other things going on.

ETA: Ah! Here's the third rule.

3. When the colony moves, I will by default do the same thing I did the last time we moved.

So, an ant who has never moved before will feel the urge to pick up some helpless member of the colony (an egg or pupa) and march along with everyone else. Or, if there is nothing left to pick up, he will simply search for food, moving along with the other ants. Which brings up a fourth rule:

4. When moving, stay with the other ants.

This should really be the first rule.

Once an ant has performed this task once, it becomes his "job", and he will automatically do the same thing next time, which will increase reaction time of the colony.

[CurtC]

Could free will, exercised by individual humans, be the same? Could it simply be the resolution of some potentiallity produced by many parallel processes in the brain, rather than a single, unified "command" coming from the ego? Could it be that the ego is the last member of the "mind colony" to find out about the decision?

My impression of brain research is that this is exactly the predominant view.

[dlorde]

Could free will, exercised by individual humans, be the same? Could it simply be the resolution of some potentiallity produced by many parallel processes in the brain, rather than a single, unified "command" coming from the ego? Could it be that the ego is the last member of the "mind colony" to find out about the decision?

Yes, I believe it could.

[Beth]

OP could almost be a snippet from Marvin Minsky's "Society of Mind" (though did he use that many rhetorical questions?) ;)

I'll have to check it out, thanks.

Let me second this recommendation. Hofstadter's work is also quite thought provoking as well as entertaining.

[Dilb]

They're not, actually. If they find 5 pieces of food, then they're behaving in a consistent way 50% of the time.

If they make a 50/50 decision every time, that is being consistent (in this "monty carlo determinism", if we need a term). This isn't a new idea in the free will/determinism debate. For free will to be a worthwhile thing, it needs to be something much more impressive than pure randomness, so these appeals to random processes don't help. This is just part of the standard argument against free will.

I think we agree on this point, the only question being whether we agree on the terminology we're using. That doesn't mean a collective behaivior model is useless, just that it doesn't allow for free will in any special way.

[Squeegee Beckenheim]

I think we agree on this point, the only question being whether we agree on the terminology we're using. That doesn't mean a collective behaivior model is useless, just that it doesn't allow for free will in any special way.

I think that's true. I wasn't trying to pass the argument off as original with me, and I was trying to make the case that it's possible to have a model without free will that isn't deterministic or dependent on neurons/ants behaving in the same way in response to the same stimuli.

[dlorde]

...I was trying to make the case that it's possible to have a model without free will that isn't deterministic or dependent on neurons/ants behaving in the same way in response to the same stimuli.
Yes; particularly where the brain is concerned, learning from experience (storing new patterns) means that even if it were possible to provide exactly the same stimuli and external context more than once, the response would not necessarily be identical each time. In fact, at a much lower level, one of the most fundamental characteristics of neuronal function, habituation, tells us we should not expect repeated stimuli to elicit the same response or the same level of response.

[JJM 777]

Assuming the colony's decision to move isn't simply made by the queen (which seems unlikely, considering her small brain), then this exercise of collective will is de-centralized; there is no focal point of authority
I am unconvinced that the size of an ant´s brain prevents it from deciding to move on when food runs out. Plants can grow towards light and do many other survival moves _without any brain at all_.

[aggle-rithm]

I am unconvinced that the size of an ant´s brain prevents it from deciding to move on when food runs out. Plants can grow towards light and do many other survival moves _without any brain at all_.

I think they probably have certain hard-wired behaviors that are kicked off in response to certain stimuli. That's what allows this to happen:

http://theoatmeal.com/comics/captain_higgins

[mike3]

If they make a 50/50 decision every time, that is being consistent (in this "monty carlo determinism", if we need a term). This isn't a new idea in the free will/determinism debate. For free will to be a worthwhile thing, it needs to be something much more impressive than pure randomness, so these appeals to random processes don't help. This is just part of the standard argument against free will.


I wonder about this: How would one empirically distinguish "real" free will from "randomness", such that an empirical test can then confirm or refute the idea?

[Squeegee Beckenheim]

I wonder about this: How would one empirically distinguish "real" free will from "randomness", such that an empirical test can then confirm or refute the idea?

I'm pretty sure you couldn't. But the fact that our conscious minds lag behind our senses would seem to be a pointer, I'd have thought (or not have thought, as the case may be).

[blutoski]

I think insect 'colonies' are different than 'swarms' or schools of fish or flocks of birds or herds of animals in terms of the process for significant changes like migration.

Insect colonies have a queen that provides some high-level control. Scientists have removed a queen from her nest to get information about her role many times, and it does seem to massively impact the vitality of the colony. Some colonies will fail without a queen, most merely become a temporary basket case while undergoing a succesion process.