On Consciousness

[Roboramma]

So someone that never jumped out of an aeroplane with a parachute knows what it feels like to jump out of an aeroplane with a parachute before they actually feel what it feels like? How?

No, but he is afraid of jumping out of an airplane based on his other experiences and, as you say, an evolved (based on things that happened in the past) instinct to fear falling from high places.

All of this is based on information from the past, as you say.

 

[quarky]

I had a dream wherein i couldn't detect any scientific instruments.
Then my alarm clock failed to go off.

 

[Brian-M]

Sure we do, anticipation, expectation, anxiety, fear are all feelings we have before we jump out of an aeroplane for the first time with a parachute on our backs.

But all these emotions are based on information from the past/present (knowledge of parachuting, seeing the open hatch and the ground a great distance below) being used to anticipate a future event (jumping out the hatch and falling from a great height).

The capacity to anticipate future events is a trivial thing to program. So when you said you cannot program a computer to feel the future, what were you objecting to?

Perhaps the keyword here is feel? But assuming that you have a program capable of actually feeling emotions (which is theoretically possible, albeit not practically possible at present), why do you believe it wouldn't be able to feel these kinds of emotions for anticipated events as humans do?

 

[!Kaggen]

No, but he is afraid of jumping out of an airplane based on his other experiences and, as you say, an evolved (based on things that happened in the past) instinct to fear falling from high places.

All of this is based on information from the past, as you say.

Biological evolutionary adaption is not a function of the past. The past is an abstraction that humans invented. Humans don't adapt to the past. We adapt to the future.

 

[!Kaggen]

But all these emotions are based on information from the past/present (knowledge of parachuting, seeing the open hatch and the ground a great distance below) being used to anticipate a future event (jumping out the hatch and falling from a great height).

No the abstract knowledge you mention is based on the past. Emotions are not abstract knowledge.

The capacity to anticipate future events is a trivial thing to program. So when you said you cannot program a computer to feel the future, what were you objecting to?

It's only trivial when the future events are trivial, because they are not really future events, but projections of abstractions of the past into the future.

Perhaps the keyword here is feel? But assuming that you have a program capable of actually feeling emotions (which is theoretically possible, albeit not practically possible at present), why do you believe it wouldn't be able to feel these kinds of emotions for anticipated events as humans do?

Assuming?
Yes it is theoretically possible to project past abstractions into a future which is a projection of past abstractions. As long as one is consistent.

The real world however is not a function of our consistent abstractions.

If computers are to function independently of us in the real world they will need to be more than a projection of our abstractions.
Seeing that computers, like all human tools, are defined as projections of our abstractions of the past they will never be independent of us since they require abstractions of the past to exist.

Because by definition the real future is not our abstract projections of the past.
Computers cannot feel the real future they can only feel the abstract projections of the past into the future we program them to feel.

 

[Roboramma]

Biological evolutionary adaption is not a function of the past. The past is an abstraction that humans invented. Humans don't adapt to the past. We adapt to the future.

Which of the following do you disagree with:

1. Natural selection is the main driving force of evolutionary adaptations.
2. Natural selection is, to quote wikipedia:

Natural selection is the gradual, non-random, process by which biological traits become either more or less common in a population as a function of differential reproduction of their bearers.

Or, if you don't disagree with that, how do you suppose the future is able to influence natural selection?

What happens is that individual who fall of cliffs die and, thus, fail to reproduce or to contribute to the reproduction of their close kin.
Individuals who happen to have some uncomfortable feeling when too close to the edge of a cliff (say fear) are less likely than those without that feeling to fall off cliffs.
If that feeling is influenced by some genes, those genes will tend to spread through the gene pool over time.
Because this process took place in the past, we living now tend to have those genes.

Whatever you mean by "biological evolutionary adaptation", if you are supposing that it's influenced by the future, you are certainly in disagreement with evolutionary biologists.

 

[Roboramma]

Computers cannot feel the real future they can only feel the abstract projections of the past into the future we program them to feel.

The same is true of humans.

If you can give an example of human behavior that can't be explained as dealing with information from the past to make predictions about the future... well, I'll be very impressed.

 

[dlorde]

humans don't adapt to the past. We adapt to anticipate the future.

ftfy.

Computer systems have already been developed that can anticipate the future by modeling contexts (typically of potential actions), evaluating the probability of various outcomes, and plan their next actions based on the results, or feed the evaluated likely outcomes into another round of modeling. The algorithms for this kind of multi-level decision tree analysis have been refined by generations of computer chess and other game algorithms and are now being used in more general contexts.

Even the simplest tic-tac-toe program needs to anticipate the opponent's moves, and chess programs spend most of their time 'imagining' the results of possible future moves.

 

[Brian-M]

Biological evolutionary adaption is not a function of the past. The past is an abstraction that humans invented. Humans don't adapt to the past. We adapt to the future.

"Biological evolutionary adaption" involves organisms becoming better adapted to present environmental conditions. (And the present immediately becomes the past.) We don't biologically adapt to the future.

No the abstract knowledge you mention is based on the past.

Can you give me any example of abstract knowledge that isn't based on the past (or present)? If not, then what are you objecting to?

Emotions are not abstract knowledge.

Emotions are internal states of mind that can be triggered by many things, including abstract knowledge.

It's only trivial when the future events are trivial,

It's trivial regardless of the significance of future events.

(And the matter of accuracy is independent of both these factors.)

because they are not really future events, but projections of abstractions of the past into the future.

Not always. It's also possible to model possible future events based on past and present information, regardless of whether or not equivalent events have ever occurred in the past.

But either way, isn't this exactly what the human mind does?

The real world however is not a function of our consistent abstractions.

So?

If computers are to function independently of us in the real world they will need to be more than a projection of our abstractions.

They'd be better off generating their own abstractions rather than relying on ours. Independence of thought is important.

Seeing that computers, like all human tools, are defined as projections of our abstractions of the past

This seems like gibberish to me.

they will never be independent of us since they require abstractions of the past to exist.

Because by definition the real future is not our abstract projections of the past.
Computers cannot feel the real future they can only feel the abstract projections of the past into the future we program them to feel.

We wouldn't be programming to feel a specific future. They'd be generating their own expectations of the future and react emotionally to these expectations.

But are you saying computers cannot feel emotion about the actual future, because they can only anticipate possible futures based on existing information derived from past and present experience, and cannot know for certain what the actual future will be?

If so, how are humans any different?

 

[Mr. Scott]

Sure we do, anticipation, expectation, anxiety, fear are all feelings we have before we jump out of an aeroplane for the first time with a parachute on our backs.

So someone that never jumped out of an aeroplane with a parachute knows what it feels like to jump out of an aeroplane with a parachute before they actually feel what it feels like? How?


It is a trivial ability of humans which developed through evolution. No magic there. You computationalists should study biology sometime, you might learn something instead of assuming everything that is not about computers is magic.

Yes, your wording "feel the future" is the problem with your statement.

Let's use the word "predict," and without any assumption of supernatural prophesy.

Computers predict the future by extrapolating from the past. Video game characters can easily predict where you will be in a moment and shoot at where you will be rather than where you are. Of course, an opponent can change course, and there are algorithms to make predictions there.

And has been pointed out, in games from tic-tac-toe to chess, computers routinely anticipate future positions to decide where to move.

We call this "feeling" because this work of our neural networks is done subconsciously ( not leaving traces of how the data processing was performed in our conscious memory). The result comes to us as a feeling.

There's a famous story of a psychiatrist who, even though he hadn't seen a particular patient in over a year, started to get a feeling of concern and called the patient. Sure enough, the patient was having sudden serious emotional difficulties. The psychiatrist at first wondered if he was psychic, then realized the day was the anniversary of this patient's extremely traumatic experience of a prior year. In the patient and the psychiatrist, that day subconsciously triggered a "feeling" about trouble. The data processing connecting the date with the patient's distress was not at the conscious level.

So it goes with "feeling the future." Feelings like that come from unconscious data processing. Indeed, the vast majority of what the brain does is subconscious.

If we wanted to make machines "feel the future" we'd build separate subsystems that would take in the information they needed, make predictions, and feed only the results to the conscious module. No magic bean needed.

This one was a breeze. I feel you'll want to hit me with another one.

 

[Mr. Scott]

No, you cannot program a computer to feel the future.
All the information you would use is from the past which has nothing to do with the future. Unless you believe in magic, do you?

I just realized what computers do is "leading the target" which various mechanical and electronic computers did with flexible gunnery in WWII.

I think this is what you mean by feeling the future. In this sense, machines can be programmed to feel the future.

No, I don't believe in magic.

 

[quarky]

John Lilly, before going off the deep end (well, actually during the deep end) described the silicon-based life forms that were using us to establish their physicality. Their plan, according to J.L., was to gradually eliminate the O2 in the atmosphere.
They don't need it, and it mostly causes corrosion.

Too bad he didn't live long enough to see us all doing their bidding.

(Gulp?)

 

[!Kaggen]

Which of the following do you disagree with:

1. Natural selection is the main driving force of evolutionary adaptations.
2. Natural selection is, to quote wikipedia:


Or, if you don't disagree with that, how do you suppose the future is able to influence natural selection?

What happens is that individual who fall of cliffs die and, thus, fail to reproduce or to contribute to the reproduction of their close kin.
Individuals who happen to have some uncomfortable feeling when too close to the edge of a cliff (say fear) are less likely than those without that feeling to fall off cliffs.
If that feeling is influenced by some genes, those genes will tend to spread through the gene pool over time.
Because this process took place in the past, we living now tend to have those genes.

Whatever you mean by "biological evolutionary adaptation", if you are supposing that it's influenced by the future, you are certainly in disagreement with evolutionary biologists.

Evolutionary adaption only has "meaning" based on how a biological entity functions with regard to the future that meets it in the present.
A collection of genes has no "meaning" with regards to evolutionary adaption without interacting with the future that it confronts in the present.
The fact that humans can extract meaning from genes has only to do with the genes past interactions. This does not relate, unless you believe in magic, to their future interactions which have not happened yet. It is an abstraction that humans invented. It is a model, it is not what happens.
No model is the future. It is a guess. Like I said computers will only be as developed as the model that humans abstract from the past.

The same is true of humans.

If you can give an example of human behavior that can't be explained as dealing with information from the past to make predictions about the future... well, I'll be very impressed.

Either you are assuming all human behavior has happened already or you are putting too much faith in our ability to predict human behavior from models of the past.

 

[!Kaggen]

ftfy.

Computer systems have already been developed that can anticipate the future by modeling contexts (typically of potential actions), evaluating the probability of various outcomes, and plan their next actions based on the results, or feed the evaluated likely outcomes into another round of modeling. The algorithms for this kind of multi-level decision tree analysis have been refined by generations of computer chess and other game algorithms and are now being used in more general contexts.

Even the simplest tic-tac-toe program needs to anticipate the opponent's moves, and chess programs spend most of their time 'imagining' the results of possible future moves.

No we adapt. We cannot anticipate a future that has not happened, that would be magic.

What future are you anticipating?

I assume your a human

 

[!Kaggen]

"Biological evolutionary adaption" involves organisms becoming better adapted to present environmental conditions. (And the present immediately becomes the past.) We don't biologically adapt to the future.

Sure we do, watch the olympics this year.

Can you give me any example of abstract knowledge that isn't based on the past (or present)? If not, then what are you objecting to?



Emotions are internal states of mind that can be triggered by many things, including abstract knowledge.

They can be triggered by the future

It's trivial regardless of the significance of future events.

(And the matter of accuracy is independent of both these factors.)



Not always. It's also possible to model possible future events based on past and present information, regardless of whether or not equivalent events have ever occurred in the past.

But either way, isn't this exactly what the human mind does?


So?

The human brain evolved to deal with the real world. The ability to abstract from the past which is no longer real and project that into the future is a recent development in human history.We are hypnotized by this ability and project it everywhere, but that is not reality's problem its ours.

They'd be better off generating their own abstractions rather than relying on ours. Independence of thought is important.

A computers "abstractions" will be our abstractions in the same way that a cabinet was not built by the hammer we used to build it.

We wouldn't be programming to feel a specific future. They'd be generating their own expectations of the future and react emotionally to these expectations.

But are you saying computers cannot feel emotion about the actual future, because they can only anticipate possible futures based on existing information derived from past and present experience, and cannot know for certain what the actual future will be?

If so, how are humans any different?

We are not someones abstractions from the past.

 

[Mr. Scott]

No we adapt. We cannot anticipate a future that has not happened, that would be magic.

What future are you anticipating?

I assume your a human

As I said, what we use our conscious for is just advanced and complicated forms of leading the target.

There was a cool study of bees that showed how, with their pin head sized brains, were about to "feel the future."

IIRC the researchers put a series of dishes some distance from a bee hive, one of which contained food they liked, the others did not. Each day the bees learned which one had the food, and the next day, only the dish one step farther from the hive was baited. Each day the bees went to the dish where the food was the previous day, found no food, and discovered by random search the baited dish. You know what happened after several days of this? The bees went not to yesterday's dish, but the dish they expected would be baited next. They learned to lead the target, or, as you say, feel the future. Just like we would, the "got it."

So, do bees need consciousness to do this?

 

[PixyMisa]

As I said, what we use our conscious for is just advanced and complicated forms of leading the target.

There was a cool study of bees that showed how, with their pin head sized brains, were about to "feel the future."

IIRC the researchers put a series of dishes some distance from a bee hive, one of which contained food they liked, the others did not. Each day the bees learned which one had the food, and the next day, only the dish one step farther from the hive was baited. Each day the bees went to the dish where the food was the previous day, found no food, and discovered by random search the baited dish. You know what happened after several days of this? The bees went not to yesterday's dish, but the dish they expected would be baited next. They learned to lead the target, or, as you say, feel the future. Just like we would, the "got it."

So, do bees need consciousness to do this?

Consciousness is not necessary for a relatively simple problem like that, but it would certainly help, and might be the most computationally efficient method.

Do you have a link to the research? That behaviour is interestingly un-insecty.

 

[Mr. Scott]

Consciousness is not necessary for a relatively simple problem like that, but it would certainly help, and might be the most computationally efficient method.

Do you have a link to the research? That behaviour is interestingly un-insecty.

Sorry, Pixy, it was a real long time ago I read it in a magazine, don't remember which.

I wonder if it's linked to a mechanism they may have evolved that anticipated the gradually lengthening and shortening days, sunrise and sunset time changing with the seasons, that they re-purposed for other daily movements.

 

[quarky]

I recommend "Superorganisms" by E.O. Wilson. Ant colonies are truly amazing.

 

[Brian-M]

Evolutionary adaption only has "meaning" based on how a biological entity functions with regard to the future that meets it in the present.
A collection of genes has no "meaning" with regards to evolutionary adaption without interacting with the future that it confronts in the present.

What makes you think that evolutionary adaption has meaning? It just is.

But could you rephrase the bit about "the future that meets it in the present"? I'm don't know what you're trying to say.

No we adapt. We cannot anticipate a future that has not happened, that would be magic.

The future by definition has not happened. But we can anticipate events that may (or may not) happen in the future by extrapolating from past and present events and information. Computers can do this too.

Sure we do, watch the olympics this year.

What does the Olympics have to do with this?

They can be triggered by the future

No they can't. But they can be triggered by our present expectations of future events. The triggered emotions will reflect our expectations of what will happen in the future, regardless of whether or not future events match these expectations.

These expectations exist in the present and are based on past and present information.

The human brain evolved to deal with the real world. The ability to abstract from the past which is no longer real and project that into the future is a recent development in human history.We are hypnotized by this ability and project it everywhere, but that is not reality's problem its ours.

Recent development in human history? Even animals can do this.

For example, a cat or dog might "abstract from the past" that the sound of a can-opener is often followed by the arrival of food, and so at the sound of a can-opener in the present they can "project that into the [near] future" and come running over in anticipation of being fed.

We are not someones abstractions from the past.

You're saying that computers are?