owleye

DRFseven...

"The difference is only that we have access to more alternatives than the kitten does."

Of course, but what's your point?


"It's still a trip swich that activates our response, even if the response route is much shorter in the kitten."

I assume you mean here the response time between a stimulus and a response. In first place, I would disagree with this characterization. Stimuli, by definition, of course, are such as to cause something to happen, but I would contend that we process information from the world largely on the basis of our own conceptual scheme, acting on it, not, for the most part, because of what bombards us on a daily basis. We are not passive and impressionable creatures so far as we have learned enough to make our own way in the world. It is we who are largely in charge of the environment, not the other way around. Alternatively, you could say that fewer things in the world stimulate us in way it does a cat, though, of course, many of us are indeed so captured, which in my mind, does not say much for us.

In the second place, there is probably no direct connection between this "trip switch" (sometimes called a readiness potential) and stimuli. What makes you think there is?

But, more significantly, the line of argument you seem to be insisting on reflects a source of explanation that can only be physical and causal, somewhat like you might explain the behavior of computers. Computers, in this way of thinking of them, are solely physical, what some refer to as 'hardware,' abiding by the causal principles of electronics. Software has no bearing on their performance since software depends on the logic of computers, and computers have no logic. Indeed, there is no such thing as software and no such thing as logic. It would be a wonder that you would think that we can make decisions at all. Indeed, I wonder that you even say that humans have more alternatives than a cat. In this causal way of explaining human behavior, there can be no alternatives. It is fully determined.

Have I been misled by the line you are taking or is this the jist of it?

owleye

John Page

Owleye:

A fully deterministic system can display choice/preference. By storing prior outcomes we learn to "predict the future". Our minds can evaluate a current situation and try and steer reality toward a desired outcome. This happens both consciously and subconsciously.

I contend a Turing Machine with the right sensors could achieve this. If you suck in all the data you can about how the world works and analyze it internally IMO you will be able overcome the commonly supposed limitation on machine consciousness (Penrose et al).

So, our apparent free will arises from the fact that we cannot predict our individual actions very well. When dealing with a specific advertising experiment to a mass audience and a limited number of variables we can achieve a % of the desired outcome.

The complexity of our minds, based on our deterministic brains etc. allows us to make good predictions about inanimate matter, less succesfully with worms and so on. Thus, free will is relative to complexity and not "absolute".

How are the desired outcomes generated in the first place? Don't know - maybe Freud was right!

If you see any holes in this please fire away.

Cheers.

DRFseven

You said we are not captured by stimuli as a kitten is. My point is that we are still reacting to stimuli, no matter how vast our reportoire in comparison.

No, by shorter route, I was referring to actual length of neural connections involved in a behavioral response. In other words, in most cases, we must think (we have no choice) before we act, which necessitates lots of connections. The kitten is probably unencumbered with what we would call thinking, but it is still responding to a stimulus. Instead of shorter, I should have said "more direct."

And what makes you think these mental schema are not also stimuli? In fact, almost all human responses involve intermediate mental stimuli; we don't respond to external stimuli so much as we respond to what we think about the external stimuli. In many cases, the external stimulus triggers the cascade of internal stimuli.

If you are simply saying we respond differently from cats, of course, I agree, but we are not simply talking about quality of responses, we are talking about control. We do manipulate our environment in ways that other animals cannot, yet we do it according to the extent we have been enabled by happenstance biology and experience. Just because someone decides to do something, such as to become educated, doesn't mean that person controlled the forces that resulted in such a decision. On the contrary, we come into the world with a biological inheritance and begin accumulating experience that is converted into mental schema without our permission.

Well, is the electrochemical action the thought, or does it cause the thought? I admit I don't know the answer to that, but I'm in pretty good company. I do know, however, that we have not controlled the formation of that thought; its basis was established by happenstance prior to any triggering stimulus.

LinuxPup

This is an interesting topic, and I was thinking:

If "I" am merely a conglomeration of particles (matter and energy), and if I am completely determined to do what I do, whether I like it or not. That is, I think I have free will, but I really don't. Is there really an "I" to begin with? Do "I" really exist?

DRFseven

Yes, it seems almost every discussion of how the brain works involves a discussion of consciousness, doesn't it?

owleye

John Page...

"A fully deterministic system can display choice/preference. By storing prior outcomes we learn to "predict the future". Our minds can evaluate a current situation and try and steer reality toward a desired outcome. This happens both consciously and subconsciously.

How do our minds "evaluate" and "try" and "steer" a fully determinate system? One possible answer is that our mind is a separate entity from that which it evaluates, tries, and steers, something like an automobile being driven by a driver. Surely you cannot mean this.

"I contend a Turing Machine with the right sensors could achieve this."

How is a Turing Machine a fully determined system? I suppose what you have in mind is not that it is causally determined, but rather that it is an ideal state machine that processes data in accordance with instructions that necessarily place it in the sequence of states. Formally, it is no more than a sequence of states. Is it because it is logically determined that makes possible the choices? I hardly think so.

In any case, as you indicate, humans are able to control their environment, just as robots can. As a control/information processing system, a robot or a human mind can exercise a certain amount of control over the variables in their environment. That our mind is a control system, is undoubtedly a better way of understanding what's going on that thinking it is a fully determined system. Of course, even with this concept, there remains the problem of consciousness, and how it is able to control anything.


"If you suck in all the data you can about how the world works and analyze it internally IMO you will be able overcome the commonly supposed limitation on machine consciousness (Penrose et al)."

The problem term here is 'consciousness.' As Chalmers indicates, why is consciousness necessary in the first place? Couldn't we create a machine that could do what Penrose requires without the need for consciousness?

"So, our apparent free will arises from the fact that we cannot predict our individual actions very well."

What role does "apparent" play in this conclusion? In any case, I don't really follow how you arrived at this conclusion. Perhaps you can fill in a few more details here. (I gather that you do some of this in the sequel.)

"When dealing with a specific advertising experiment to a mass audience and a limited number of variables we can achieve a % of the desired outcome."

So?

"The complexity of our minds, based on our deterministic brains etc. allows us to make good predictions about inanimate matter, less succesfully with worms and so on. Thus, free will is relative to complexity and not "absolute".

Well, I think the relationship you are making, with increased knowledge being related to decreased free will, is precisely backwards. The more I know, the greater my choices, the more likely I am to be characterized as having free will. Indeed, it is precisely the choices available to us that accounts for our understanding of free will. We excuse those who have no choice, but blame or credit those who do. Your way of thinking of this needs much more work for it to make sense.

owleye

owleye

DRFseven

"You said we are not captured by stimuli as a kitten is. My point is that we are still reacting to stimuli, no matter how vast our reportoire in comparison."

It is true that between two different things there are always similarities. However, that there are similarities does not refute the differences. It was for this reason I found the statement rather odd.

"No, by shorter route, I was referring to actual length of neural connections involved in a behavioral response. In other words, in most cases, we must think (we have no choice) before we act, which necessitates lots of connections."

What is the relationship between the length of neural connections and behavioral response? I don't get it? I had mentioned time and you dismissed this. What is it about the length of a neural connection (which I gather means the distance between neurons) have to do with behavior?

More importantly what does all this have to do with thinking we have no choice before we act? I don't see the relevance of one to the other. Secondly, I think you are confusing what makes a will free. It is not because decisions are in fact made, it is rather that the decisions were deliberated on and a decision was made. If there was no deliberation, and we merely reacted to stimuli, the way you seem to think we do, we would never think we had choices at all.

"The kitten is probably unencumbered with what we would call thinking, but it is still responding to a stimulus. Instead of shorter, I should have said "more direct."

What do you mean by "thinking" and why is it important to your point? Most folks believe that thinking is what makes us free. You seem to believe otherwise.

"And what makes you think these mental schema are not also stimuli?"

If you fail to distinguish stimuli from the environment from stimuli from our conceptual scheme, then our conceptual scheme does not belong to us. We can blame our conceptual scheme for causing us to act in a certain way. Let me acknowledge that many of us are trapped within our conceptual scheme and when we realize this we often remark that we made a mistake. Well, the same thing holds for normative ideas. We can know when we were wrong. We do this because of the way our conceptual scheme works. By virtue of its being our representation (framework) of the actual and ideal world, we certainly act and judge in accordance with it. But this is what makes it possible for us to make decisions, either in conformity with the scheme or by denying the applicability of the scheme in a given instance. It is a capability that is, for the most part, exclusively human.

"If you are simply saying we respond differently from cats, of course, I agree, but we are not simply talking about quality of responses, we are talking about control. We do manipulate our environment in ways that other animals cannot,"

What makes you think this? How many creatures are you aware of that can clone genes? Indeed, the entire array of cultural behavior is quite unlike that of almost all the animal kingdom. What is it exactly you are saying here?

"Just because someone decides to do something, such as to become educated, doesn't mean that person controlled the forces that resulted in such a decision."

So you say. But if you must rely for this conclusion that the forces that are not subject to control were the firing of your synapses, then most folks would think this is irrelevant to it being a decision to become educated. Who cares what the sequence of firings were that led to this decision? If it made a real difference, I suspect we would try to do something about it. Indeed, many of us try to do just that sort of thing by using drugs. It is quite apparent to me that you think that because we are causally determined that we lack free will. Your mere statement of your belief is insufficient to convince me. I am quite satisfied that we can be free despite that we might be fully determined.

"On the contrary, we come into the world with a biological inheritance and begin accumulating experience that is converted into mental schema without our permission."

This is ridiculous. It may be, as Sartre says, that the "situation" we are thrown into is so overwhelming that we can hardly do anything else but what we do, but there is just no way to escape the burden of our being free to face the situation in one way or another. Your world is sterile, meaningless, and, in my mind, of little consequence, since it completely fails to take into consideration that we are responsible for our actions, just because they are actions, and not just behavior. In your world, a causally determined one, there really are no choices. There is only one future. Indeed, whatever attitude we might have about our future it cannot be mistaken, since it is part of that one future. We can never be right or wrong. Indeed, we cannot have knowledge of things. Knowledge, if it exists, does not belong to us. It exists merely as an influence on us. This world that you are making for yourself is, I confess, something quite radical, but also, I suspect, not particularly helpful to anyone's life.


"Well, is the electrochemical action the thought, or does it cause the thought? I admit I don't know the answer to that, but I'm in pretty good company."

First of all, I don't think you are in good company. But what company you keep is not relevant to the arguments you present.

"I do know, however, that we have not controlled the formation of that thought; its basis was established by happenstance prior to any triggering stimulus. "

As far as i can see there is no thought in your way of thinking of things. Thoughts, if they existed, don't do anything -- neurons do, and they can't be controlled, or so you say.

owleye

John Page

You're right, I don't mean that. How exactly, obviously I don't know but my "model" of the mind is an information entity that contains working hypotheses about its environment. By internally monitoring sense data and hypothetical outcomes action can steer reality toward a desired outcome.

I only consider mind separate in the sense it is abstract and the body physical. I believe it resides on the physical substrate of the body so I'm not a dualist in the traditional sense of the word, but neither am I a strict determinist.

I also suggest that the mind can be considered as separate parts with several layers of abstraction, in this sense parts of the mind evaluate/control/monitor control other parts of the mind.

I don't know how a Turing Machine is deemed a fully determined system. However, if the states it contains bear relation to states in the outside world and are stored in a manner that is contextual with other states (e.g. x happened at the same time as y but produced a different color) this data can be used by a simple process to handle a wide array of situations. By experimentation with the state data, learning of optimal actions can ensue. I know this because I have created a multi-user multi-tasking non-preemptive scheduler for MS-DOS (which is non re-entrant) where all of the necessary functions were contained within a finite state automata (written in assembler, triggered on machine interrupts).

I have to admit that the "learning" to optimize system behavior was done manually but one of my colleagues was playing with Occam and transputers and we spent some time figuring out how a self-optimizing system might be implemented.

So how do choices get determined? (Note that I shy away from logically determined). In humans I suggest this is likely due to reward/punishment systems that have developed stepwise to optimize behavior. Ultimately the goal may be simply be survival - if you don't do that then your behavior traits don't get passed on. In this sense we could be considered programmed by our environment. In a Turing Machine, hmmm, if you could provide it the means to stop you unplugging it and an understanding that it would cease to be when you unplugged it....

I'm not sure I have the answer to Chalmers, but look at the advantages consciousness might confer. We have a real-time convergence of data from reality merged with experience of past situations codified to provide us an understanding of cause and effect. A creature with this capability is going to out-predict a simpler creature that only has unconscious or reflexive actions. This brings us back to the advantages of increased levels of free will (I'm not saying that consciousness is a pre-requisite for free will though).

I'm not really into what Penrose requires, real world requirements are the drivers.

I desired to indicate that free will is not a binary, on/off, property. Perhaps "apparent posession of absolute free will..." would have been clearer.

My wording indicates a direct relationship, not an inverse function, so I think we're in agreement. (Increased complexity == Increased free will, providing of course that the complexity is appropriately organized!)

Good questions! I hope the above makes clearer the thinking behind my words.

Cheers!

[ April 03, 2002: Message edited by: John Page ]</p>

DRFseven

You find it odd because, to you, the differences account for the free will you think we possess. I find that, as great as the differences are, there is still no freedom from prior causation in our own behavioral mechanisms. The thoughts that govern our behavior all come with strings attached; just because many of the strings are conceptual does not make them nonexistent.

Time, as far as this discussion goes, is not important in determining capacity of behavioral response; it is the number of connections. Abstract thought involves activation of constellations of neural connections all over the brain because thought comprises memory associations. Most human behaviors are predicated by this type of neural activity, while the cat's behavior involves comparatively simple sensorimotor function without, we presume, abstract thinking and the concomitant invocation of accrued hierarchical schema.

It's not strictly true that I think we don't have choice. It's that I see choices as an evolved mechanism whereby the individual's responses are determined according to previous experiential factors instead of being determined by hard-wired biological factors. To many people, an individual's choice is not a choice, an individual's freedom is not free, if it is determined by antecedent causes, so what I call a choice may not mean choice to you. I think a response is a choice if it reflects an evaluation based on the individual's conclusions. Is this what you think?

I think the previous paragraph addresses this.

Surely you see, now, that the whole point of my talking about the complexity of abstract mental thought with its dependency on hierarchical conceptions as compared to the cat's direct sensorimotor responses is to note the distinction and to point out, at the same time, that neither type is free.

Certainly our mental schema belong to us; certainly we think our thoughts and make our decisions; indeed, we have no say in the matter, we are bound by our physiology to do this.

I don't deny a bit of this. You are describing how we are bound to "come to" conclusions based on experience; not to freely design conclusions unfettered by bases.

I don't know why you ask this. My statement, again, was "We do manipulate our environment in ways that other animals cannot,"....

Owleye, the firing of the synapses either are or are caused by (depending upon which view of the neurophysiology of cognition is assumed) the various conceptual schema held by an individual at any point in time. Either way, once a thought-provoking stimulus is received by the senses, the whole rigamarole that is sensitized to that specific associative coding is engaged. We have groupings of firings going off all over the place ("knowledge"; nested beliefs, conceptualizations) over which we have not the slightest control and many of which we are not even aware. We might end up saying something like "I retied my shoe because I just felt it needed retying.", without being aware of all those thousands of connections that started accumulating back when other people tied our shoes for us and what aspects of shoe-tying and peripheral issues seemed rewarding to us throughout our lives and what attitudes and motivations subsequently accrued and were activated by the act of feeling a certain looseness and looking down at a shoelace. Even in much more complex issues, we are aware of only the "top layers" of the attitudes that form our opinon on the issue; the deep-down layers have long-since become shortcuts straight to feelings and are unavailable to us in words. You ask "Who cares?" about the sequence of firing. I can't identify who those are who care (though there are quite a few!), but I can tell you that the importance of the sequence of firing is that the sequence and the specific thought are two sides to a coin. Without that specific sequence, the specific thought would not exist to the thinker.

What is ridiculous is the idea that we might somehow design a set of beliefs and thoughts before we believe and think them! "Now I shall choose to think education is good before I have any idea what it is, and I will also decide without evidence that impulsive behavior should be avoided., I think I shall decide to think denistry is a more appropriate field for me than engineering and I shall come to that decision at age 23. Without meeting any of either species, I shall decide to prefer cats over dogs and before tasting it that lasagna is my favorite food. At the end of my life I will decide I have been correct. "

Obviously, we would have no way of coming to these conclusions if experience had not provided the means to acquire them. Sure, we can manipulate our environment, provided OTHER appropriate experience has taken place.

Ludicrous; thought determines almost all our behavior.

I think neural activity IS thought; what do you think thought is? And how do you think neural activity or thought is controlled? When, for instance, you perform some task and you don't achieve the results you wanted, can you alter that perception? Can you decide to think that the task was done appropriately, or are you bound to your conceptions of what constitutes a correctly performed task? Could you decide to think you should perform the task again in exactly the same way?

owleye

John...

"You're right, I don't mean that. How exactly, obviously I don't know but my "model" of the mind is an information entity that contains working hypotheses about its environment. By internally monitoring sense data and hypothetical outcomes action can steer reality toward a desired outcome."

I think it very reasonable to conclude that our brain is an information processor. I also think the kind of processor it is is a control system. Indeed, all of our subsystems are control systems, and the brain is no different in this regard. Though there is a large analog basis to most of our control systems, I think the brain does much of its work because information is decoded. What are highly varying data are classified in accordance with a conceptual scheme without which our brain could hardly function. However, though the human brain has this feature, making it subject to rules, rules themselves can be lifted out of their context and treated independently, making it possible to know that data doesn't fit its category and can learn, perhaps from a third party, that the category is wrong. This is probably what you are thinking about by the term 'working hypothesis'. Our brain does more than just perform a discrimination function, a feature that we hold in common with all or most animate life.

"I only consider mind separate in the sense it is abstract and the body physical."

In what way is the mind 'abstract'? The usual meaning of abstraction is a feature of our mind that discards any concrete influences. For example, when considering the concept of cat, we are in the abstract realm of concepts, they being general, arrived at by a process of abstraction, paring away all the individual characteristics that are not held in common among those individuals that count for us as cats. (Note that an abstraction is not some generalized "image" we hold when we imagine a general cat. It is the set of common conditions we have learned which belong to all cats. They amount to a set of rules or principles that can be applied to sense experience or to behavior or to discursive formulations, or to judgements, generally, or other.)

Let me acknowledge, though, that our own ability to process abstract ideas is something that, if I'm not mistaken, is not yet possible by software. Perhaps, with your knowledge of software, you could convince me otherwise.

"I believe it resides on the physical substrate of the body so I'm not a dualist in the traditional sense of the word, but neither am I a strict determinist."

I think what you may be groping for is something similar to the software/hardware facet of computation. In philosophical circles, this is known as the type/token distinction. If this is not what you have in mind, let me know.

"I also suggest that the mind can be considered as separate parts with several layers of abstraction, in this sense parts of the mind evaluate/control/monitor control other parts of the mind."

Again, I have difficulty comprehending this, but if what you mean is that there is an element of layering (much as we might understand the OSI layers in network protocols), I think you may be onto something. I suspect not, though. I do understand, from the experiments of Libet, that consciousness in some sense "rides" on top of a sea of subconscious activity, extracting data from the world, mostly tossing it, and letting the most relevant data percolate up to consciousness. This fascinating story is interesting, especially the part about the time leads and lags that consciousness itself is involved in. At NASA, in the research conducted in flight simulation, we had also identified these lags and leads.

However, as fascinating as this is, there remain signficant problems with it if we wish, as Norretranders does, to regard it as a "User Illusion." His interpretation is that consciousness is situated in a such a way that makes free will possible. That is, he thinks consciousness has a function -- namely to veto our actions (presumably he would say the same thing for our judgements). However, if it is an illusion, how can it even do this?

"I don't know how a Turing Machine is deemed a fully determined system."

My question was based on your apparent need to tell us that a "fully determined system" could exhibit choice. To date, I have yet to understand what compelled you to suggest this.

"However, if the states it contains bear relation to states in the outside world and are stored in a manner that is contextual with other states (e.g. x happened at the same time as y but produced a different color) this data can be used by a simple process to handle a wide array of situations."

The use of 'handle' is a term from control theory. The wide variety represents the free variables that are processed by this system, notwithstanding that there is presumably a determinate process that is handled by each class. One feature of a control system is that it is never fool proof (thus, not fully determinate). Like a river bed that has too much river flowing down its banks, the bed can only hold so much and will either give way or flooding will occur. In many well-engineered software systems, and some "hardware" systems, there is a hierarchy of handlers for each such contingency, and one would hope that most problems could be gracefully handled. Because computation is relies on a physical process, we can readily imagine that it can be damaged in myriad ways, many of which probably haven't been considered, at least for most information systems.

"I have to admit that the "learning" to optimize system behavior was done manually but one of my colleagues was playing with Occam and transputers and we spent some time figuring out how a self-optimizing system might be implemented."

The problem child here is what does it mean to be a "self." I have seen database systems (or so-called knowledge-base systems) attached to a natural language interface, in which the language used to communicate to human users made it seem like its database was part of its self. Is this all it takes?

"So how do choices get determined? (Note that I shy away from logically determined). In humans I suggest this is likely due to reward/punishment systems that have developed stepwise to optimize behavior."

Well, yes, I suppose this is one way of looking at it. But this leaves completely unexplained the role of mental activity, including pains and pleasures. We are reduced to the status of any other creature. We would assume a dog who has maimed a child that got too close did so on the basis of their having learned (or been trained) in a behavioral way (or that it was instinctual and hadn't been trained otherwise, or whatever). In any case, we would not assume the dog had a choice and could have done otherwise. Indeed, we would typically find some person to blame for this incident -- someone who would be held responsible for it. Moreover, the same is generally true for children. They are susceptible to such conditioning by stimuli, which is why we usually blame the parents or the school, for the misdeeds of children. Once we are adults, however, we are expected to have moved on beyond such conditioning, or at least to be able to control ourselves better than when we were children.

"I'm not sure I have the answer to Chalmers, but look at the advantages consciousness might confer. We have a real-time convergence of data from reality merged with experience of past situations codified to provide us an understanding of cause and effect."

I don't see consciousness as necessary for this. What part of the above requires consciousness? (Note, I think that it would be more accurate to say "near real-time", since it has been determined that there is a signficant lag (about 1/2 second, according to Libet -- about .4 second in NASA studies) in what we are conscious of compared to what a real-time system says it is. This might impact your theory -- I can't say.

"I desired to indicate that free will is not a binary, on/off, property. Perhaps "apparent posession of absolute free will..." would have been clearer."

This muddies the water, I'm afraid. But let me ask you why you need to indicate 'apparent' in the first place. Second, what is the need to specify 'absolute' here? Is this the right word? Absolute is usually meant to be distinguished from 'relative'. It could very well be the case that each of us have different degrees of free-will or that we exhibit free-will differently at different times, or whatever. Is this something that came up with another interlocutor?

owleye