self-control: philosophical questions
Aug. 1st, 2006 01:44 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
guslI've been thinking a lot about self-control. What is the self? Where is consciousness?
I'm asking these questions partly because I've been planning to read an ACT-R paper reading titled "The Minimal Control Principle". But really, it's just one of those nagging questions that never go away.
People obviously have a folk theory of what they do and don't have control over, and also of what *others* do and don't have control over. In fact, *every* moral judgement we make of our own or others' actions is justified by this basic assumption that the person is an agent who is able to make choices, and to control him/herself at that level.
What is a little bit puzzling about this is that this ability to control oneself is a modal notion. How can one prove that one could have acted otherwise? We may be able to beat some predictability tests, but how many levels of determinism can you refute before the laws of physics (or neuroscience!) make you predictable?
This interesting paper (via MR) about willpower sugests that, like patience, willpower is a depletable resource. Does this mean that, after enough temptation, people cannot be expected to control themselves?
This ties back to video game design: what level of control should you give to the player? If the goal is to make the played feel immersed as seamlessly as possible, we should try to minimize the size of the (inevitable) cut between real and virtual world (min-cut <=> max-flow!). A game like Spore is interesting because you different levels of description, but it would be more interesting still (but maybe very annoying) if the game automated away parts of gameplay *just* as the user demonstrated that they reached expertise: that way, you'd always be growing to higher and higher levels (being an expert doesn't make an activity completely effortless).
Much of what we do in life and work is automatic. This means that our control is restricted to the higher-level. We can only be accountable for our automatic actions insofar as managers are accountable for the behavior of their employees.
---
ACT-R folks, when does conscious control correspond to the contents of the buffers? Should we define consciousness as being a meta-level of attention, i.e. attention to one's own cognition?
---
Interesting talk on determinism:
David Steinsaltz - Essay on chance and determinism
I'm asking these questions partly because I've been planning to read an ACT-R paper reading titled "The Minimal Control Principle". But really, it's just one of those nagging questions that never go away.
People obviously have a folk theory of what they do and don't have control over, and also of what *others* do and don't have control over. In fact, *every* moral judgement we make of our own or others' actions is justified by this basic assumption that the person is an agent who is able to make choices, and to control him/herself at that level.
What is a little bit puzzling about this is that this ability to control oneself is a modal notion. How can one prove that one could have acted otherwise? We may be able to beat some predictability tests, but how many levels of determinism can you refute before the laws of physics (or neuroscience!) make you predictable?
This interesting paper (via MR) about willpower sugests that, like patience, willpower is a depletable resource. Does this mean that, after enough temptation, people cannot be expected to control themselves?
This ties back to video game design: what level of control should you give to the player? If the goal is to make the played feel immersed as seamlessly as possible, we should try to minimize the size of the (inevitable) cut between real and virtual world (min-cut <=> max-flow!). A game like Spore is interesting because you different levels of description, but it would be more interesting still (but maybe very annoying) if the game automated away parts of gameplay *just* as the user demonstrated that they reached expertise: that way, you'd always be growing to higher and higher levels (being an expert doesn't make an activity completely effortless).
Much of what we do in life and work is automatic. This means that our control is restricted to the higher-level. We can only be accountable for our automatic actions insofar as managers are accountable for the behavior of their employees.
---
ACT-R folks, when does conscious control correspond to the contents of the buffers? Should we define consciousness as being a meta-level of attention, i.e. attention to one's own cognition?
---
Interesting talk on determinism:
David Steinsaltz - Essay on chance and determinism
The philosopher Daniel Dennett has
compared the free decision-making self in psychology to the
center-of-gravity in physics. Both are fictions, but they are useful
fictions, because they tell you where to apply force to achieve desired
effects. Recognizing that they are fictions does not force us to abandon
them, but it does suggest to us that we may need to put them aside when
the problem at hand depends on the fine structure.
(no subject)
Date: 2006-08-01 04:34 pm (UTC)If you're a compatibilist, predictability is actually essential to making sense of the notion of free choice. So, you exercise your free choice when you act according to your desires, experiences and character. Or alternately, it's equivalent to saying that your actions are fully conditioned upon the decision-making machinery in your brain.
If people were totally unpredictable, then their actions wouldn't depend upon what makes them unique individuals, and it's hard to say they were responsible for their actions. And indeed, people who can't properly process their experiences and make plans (like small children, the insane, and the senile) don't bear full reponsibility for their actions. Likewise, we don't ascribe responsibility to adults for circumstances beyond their control. If you promised to meet me for lunch, but fell violently ill, it wouldn't be your fault that you didn't keep your promise, because the illness means that your action (skipping lunch) was now independent of the your decision-making processes.
Saying that we can't exercise free choice because neuroscience shows our brain function is predictable, is like saying that computers don't run programs because they are really transistors shuffling electrons around.
(no subject)
Date: 2006-08-01 07:05 pm (UTC)Your analysis seems like a refinement of the folk ideas, which is of course, the right way to deal with paradoxes. The paradoxes arise from a rough formulation of the situation.
You seem to be saying that when we blame someone, we should be blaming their decision-making processes, even though they may be determined by physics, their prior experiences, etc. It's interesting, mechanically speaking, that we come up with this notion of "free will" (although it makes perfect sense in terms of evolution), that this notion arises from a deterministic system (not that there is such a thing as a non-deterministic system: it's just an interesting perspective to have).
It's like there's a homunculus inside of me that is my consciousness. If I am sane, then this homunculus functions well and is well-integrated with the body. You wouldn't blame me if my homunculus somehow lost control of the body (e.g. through illness or disability). Likewise, you wouldn't blame me if I were deterred by outside circumstances. But in a sense, illness and insanity are also "outside circumstances": they are outside of my homunculus's immediate control (i.e. what I can control).
A good definition of the homunculus that I am talking about here is: the part of me that is conscious and in control. It seems like a useful analogy: intention is what the little man (i.e. a "mind") does, when it's healthy. Do we use this analogy unconsciously? I certainly think that humonculi are a universally recurring idea (unlike Jaynes, I suppose).
Am I making any sense?
---
It's also interesting that people are bad a random number generation. Good cognitive architectures should reflect this.
(no subject)
Date: 2006-08-02 04:06 pm (UTC)