Being able to continuously adjust a decision given new information provides a thing with a measure of free will. We normally call this the ability to “change one’s mind”. In this actual final post on free will, I will argue that our inability to make informed decisions in the time allotted for making them–but having to act anyway–is what gives us the experience of free will.
The easiest way for me to demonstrate this is with an example, and the example I’ve chosen is the Theseus and the Minotaur maze by Robert Abbott, over at logicmazes.com. Coincidentally, the first incarnation of the online version of this maze depicts both Theseus and the Minotaur as spheres. (Not intentional, I swear!) Anyway, in this maze, the Minotaur moves twice every time Theseus moves once, but will also move horizontally before moving vertically.
Before I get further into the maze game–and before I lose you, reader, to the game–I should explain the analogy. If our sphere creature from the previous free will posts was fleeing from a predator, you’d have a situation similar to this game. After each move, the environment of the sphere changes, providing new information used to adjust its escape strategy. However, simply adjusting isn’t going to be enough; it has to be able to look several moves ahead to avoid getting caught.
In other words, the creature needs to simulate possible outcomes of its actions. Simulations require memory, and more (working) memory means more processing power. That is, as this creature grows, it can make the same decisions more quickly. The same holds for you, as you play the Theseus maze.
In the logic maze, there’s no rush to act. You can leisurely sit at your desk and work out all possibilities in your head before pressing your cursor key. In doing this, you could potentially learn, computationally, everything there is to learn about a particular maze. You can then select the winning simulation and act it out.
Now imagine that you only had five seconds to act each turn, starting from the moment the screen loads, and if you didn’t act, the Minotaur would get to move again. And just to make it more fun, when the Minotaur moves under these conditions, it will move around obstacles to reach you (in other words, cheat). Thus, you can no longer carry out every possible simulation–or more realistically, implement your entire decision-making strategy–before making a decision. You are forced to act before you’re ready. Would you agree that your inability to evaluate all the relevant information–but choosing to act anyway–makes you feel more free-willed? (Well, the time limit itself is relevant information, but whatever.)
Most every moment of our day is like this. We take in a lot of information every moment, and even after all our sensory and cognitive filters do their work, we still don’t have enough time to be certain–really certain–that our choices are the “right” choices. There are too many variables. Often we oversimplify our decision-making algorithms to fool ourselves into certainty (like finding the general notions that account for human experience, ahem), or give special attention to certain criteria, but that is merely a necessity to getting on with life.
Anyway, do I think we have free will? I think it’s clear by now that I think it’s kind of free. Could it be more free? Yes, I think it could. If we had better memories, faster neural networks, and access to more information, we’d have freer will. It’s not a question of whether free will exists, but how much we have.
Naturally, the next question is: what about absolutely free will? Well, think about that and I’ll get back to that question someday.
Tags: action, choice, free will, indecision, memory
http://home.earthlink.net/~btodd2/freewillnote.html
Was there something specific you wanted to add, Bob, or were you just doing some drive-by proselytizing?
This was a very interesting article. A lot of the ideas you expressed are ideas used in the field of artificial intelligence. The minimax algorithm is used in games to find perfect play. The most simple way of increasing the search depth is to use pruning. Pruning can massively increase the number of nodes the minimax algorithm can visit each turn. So increasing the memory and speed of the processor (neural nets in humans) isn’t the only (or the most efficient) way of finding the best possible answer.
Now for my question. We have designed algorithms and computers that can out think most humans. Do you believe that these computers have more free will than humans?
Thanks for the comments, Patrick. I don’t even know who reads this, so feedback is always a treat. And funnily enough, just last weekend I was just brainstorming some pathing algorithms for a little game my friend is working on (www.tarchongame.com) and I thought of this post at the time. I’m sure your suggestion on pruning will help.
I admit I did not mention anything like pruning in my post above, but given the generalities I’m working with, I don’t see anything that precludes pruning, either. I guess I took it for granted that evolving organisms would develop different strategies, some of which akin to pruning. Ultimately, however, with a given quantity of information, memory, and processing power, there is a limit, yes? Nonetheless, efficiency within those bounds is not a topic that should be ignored, so thanks for pointing it out.
As for your question, my answer is yes and no. ;) No, because I disagree that we have computers that can out-think or out-perform most humans. Not beyond a specific task that they were programmed for, anyway. But whether we agree on that is moot, because I think that yes, if someday there are machines that are convincingly intelligent in a way similar to ourselves, then they would have the potential to have more free will.
That said, there shouldn’t be anything stopping us from ultimately granting ourselves more free will, either. Unfortunately machines will have an advantage in that it is easier for a machine to reprogram itself than for a human to reprogram itself. Substrates being what they are.
Hi Ergo,
I see that this is an old topic here, but something I don’t believe I’ve seen in your posts on free will is an examination of the factor of what is underlying everything. I mean, does a greater ability to navigate and evaluate options really equate to greater free will if the *reason* for doing so is clearly based in something deterministic, like a need for food? If I appear to have all of these great, apparently free-will abilities, but am only making use of them because I am compelled by biological imperatives, is that really free will? I think not.
Does a human ever really do much that has nothing to do with some state of being that has been thrust upon him/her? We only do anything because we’re here, and we’re here not of any free will of ours. The closest support of a notion of free will there might possibly be what is really more of an unthinking urge within the functional biology of the gametes that surge forth to find that egg.
- Doug
Hi Doug, thanks for reading. I’ve been so busy this past year that I’ve totally forgotten this blog. I coincidentally received ten email notifications for comments on here this morning, the first in a long time. Nine were spam, and the last was yours. I didn’t expect a real comment in there!
Anyway, you raise a good point, one I’ve fortunately had time to think about. Someone similarly put it to me last year over lunch, asking “Can we choose not to eat?” You’re both right, of course, for if we do, we ultimately die and our wills die with us. We can still hypothetically not eat, but I’ll concede that the probability of that is negligible.
However, unlike simpler organisms, we have on the whole more choice in what/how, when, and where we eat, thanks to our relative growth and freedom. Those varieties are where we exercise our “freer” will.
Or as Douglas Adams said, “…the first phase is characterized by the question ‘How can we eat?’ the second by the question’ Why do we eat?’ and the third by the question ‘Where shall we have lunch?’”