It is important to consider only tasks that MUST be performed consciously. Many activities, like driving and changing gear, become so practiced that the consciousness does not need to participate. Driving is quite different in the early stages, those first attempts at clutch and brake control were much harder, took longer and took up all of the resources of the conscious. Once the skill is learned most of the donkey work is delegated to the unconscious. Once you’ve learned to drive you can drive AND listen to the radio. The conscious has been freed up for other things.
It’s also worth remembering that in the case of emergency those practiced unconscious reflexes are in charge. They must be, not only is the conscious limited in capacity, or bandwidth, it’s also lagging behind the unconscious in time, and way behind reality.
Experiments described here seem to show that the conscious lags behind perception by around half a second. We might expect some delays while objects in the field of view are evaluated and decisions are made, but consciousness, whatever it is, even lags behind when everything that is going on is internal, in our heads.
What this means is that any decision we might make is not revealed to the consciousness until around half a second after we’ve decided to do it. The decision takes place, and conscious recognition of the decision takes place around 500mS later.
This doesn’t mean that we are not creatures of free will, just that our recognition of it lags behind by around half a second. But we’ve grown accustomed to it, our unconscious mind works much faster, so it’s the unconscious that drives the car, thank goodness, the conscious just can’t keep up.
What we call consciousness, just seems to be a recognition of what we have decided, it doesn’t seem to be an essential part of the process.
In my rather facetious attack on HDTV, I intentionally left out a key point. The actual experience of reality, even the reality of watching a movie, is very different to our memories. Memories, no matter how cherished, can’t equal the richness of the actual experience. The conscious, when editing and storing our memories must leave out much of the experience. What is felt during the actual deed, as it’s happening, will be more intense than anything we can possibly recall.
The richness of the actual experience are what those 20 million bits per second of HDTV are trying to reproduce.
So is the conscious involved in editing and laying down memories? I think it must be, massive amounts of information, events that happen to us are edited out and lost. Why should they be remembered? All the automatic learned skills that we use, day to day, serve no useful purpose by being recalled. What took place on the journey, road conditions, stops for petrol, etc are rarely worth memory space once you've claimed your expenses. If, on the other hand, you stop for petrol and in doing so meet the love of your life, you’ll remember it for ever. But, it is only in retrospect that the true importance of an event can be recognised. If that love turns out only to be a passing fancy, over a longer time that recollection too may fade.
This is what Daniel Dennett describes as the Orwellian model, described here, where history, in our memories, is re-written in the light of current events and priorities.
Intentionally editing memories, BTW, which can mean just removing painful ones, as in the movie Eternal Sunshine of the Spotless Mind, seems to be getting a little closer, as related here.
I still have problems with this. Firstly, I question how they divide 'conscious' from 'unconscious' tasks. Most of the time when I'm driving, I'm aware of shifting gear. Therefore I would call this conscious. I would accept that reading the writing on number-plates and road signs is mostly unconscious, however, the meaning is apprehended from many signs simultaneously. This meaning would have to have many bits, so I still think 20 bps is too low.
ReplyDeleteFurthermore, a fast typist can type at 100 words per minute. Say 6 characters per word gives 600 characters a minute. lets say 4 bits per chracter, as we don't need the full 8-bits to encode what's on a typewriter keyboard, gives 2400 bits per minute. Divide by 60, gives 40 bps, double what has been claimed.
However, typing speeds are far lower than reading or speaking speeds. Average reading is 250-300 wpm. With rapid reading techniques 800 wpm can be reached, though comprehension drops off. 400 wpm is the limit for 100% comprehension, apparently.
However, I'm sure the claim will be made that all these activities are unconscious. Indeed, I suspect that any high-speed activity will automatically be considered unconscious. I think there is a circular argument going on here. How are they dividing activities between the unconscious and the conscious?
If I can read at 400 wpm and maintain comprehension, then I must clearly be achieving more than 20 bpm, I think? There may be some delay in my comprehension, but that is a matter of latency, not bandwidth, my throughput will still be higher than 20 bps?
A better test would be to ask what can be done with 20 bps in the computing world? I think very few tasks of the form humans do every day could be done with such a low bitrate. Indeed, I still fail to understand how I can bee seeing a complex image, with many differently shaped and colored items in it, if I only had 20 bps. 20 bps is ONE PIXEL A SECOND? If we assume that the human mind has an 'image resolution' of 640x300 pixels (I'm quite sure its much higher) then this would mean that one 'frame' would take 4200 minutes to load?
I think the notion of proof reading gives a clue. Reading for comprehension, to someone fluent and practiced in the language is largely supported by the unconcious.
ReplyDeleteProof reading, where every character in a text must be considered for correctness, is a much slower activity. The conciousness is fully engaged, because this form of reading is so different from that which we normally practice.
Deciphering an image IS a largely unconcious act, if we are familair with the image content. Preliminary, unconcious parts of the brain recognise information and extract dimensions, and such, to create an internal model which the concious brain can deal with. But these unconcious parts don't understand what it is they are looking at, anymore than the camera in my mobile phone can.
I'm still not buying it. The best that can be said is that, for certain tasks, the bit-rate is 20 bits per second, and I think even that is highly suspect.
ReplyDelete> But these unconcious parts don't understand what
> it is they are looking at, anymore than the camera
> in my mobile phone can.
Very true. But if you can only be conscious of 20 bits in a second, that means you can only be aware of 20 on/off values?
20-bits is less than the bits required to express the color of a single pixel on a modern computer system. If this 20-bits statement were true, then the human mind would only have the capacity to be conscious of one thing per second: a color. You've have insufficient bandwidth to be conscious of anything other than a single color. However, at any one time I'm conscious of many colors and shapes, and if any of them change, I know it fairly quickly.
If you take it to its extreme, you can see how impossible this 20-bit claim for consciousness is. If we slowed it down to 1-bit a second (which is not a vast change) then we'd only be able to be aware of one value per second, say light/dark. Nothing else. There just wouldn't be the bandwidth for it. Unconsciously we might be processing much more data, but our conscious would only be able to process one on/off value per second.
The twenty-bit statement is saying that this is all we can process in a twenty-ith of a second, and that in one second we can process 20 on/off values. If our experience was this limited, we wouldn't be able to do very much at all.
Even the earliest microprocessors like the intel 8080 and zilog z80 (the latter was inside the zx81) were capable of 500,000 instructions per second. An instruction is a lot more than a bit. Now, granted, these were much faster than a human being for arithmetic tasks, but I do not believe it would be at all possible to simulate human consciousness on a zx81, even with huge amounts of memory?
Indeed, it's a very interesting question to ask what you *can* do at 20 bps? I suspect not very much at all.
How concious are you of the fine graduations of the blue of the sky? Unless you are painting it, I'd say not very.
ReplyDeleteHow long does a painter taking matching the colour of the sky? How long he takes is a measure of how concious, he is, of that sky colour, and rather more concious, I would say, than a non-painter.
The time he takes matching the colour is indicative of the bandwidth of conciousness.
>How concious are you of the fine graduations of the
ReplyDelete>blue of the sky? Unless you are painting it, I'd say
>not very.
I agree with you, but right now I am conscious of three shades of grey, two of blue, white, and green, because these are some radically different colors in my visual field on my computer screen. 20-bits would mean I could only distinguish one of these a second.
If someone walks in the room, and says something, I am conscious of what they said, how they said it, how they look, what sex they are etc, etc, etc. I do not think you can say that I know these things only subconciously. I'm not conscious of the complex parsing of all this information, that is done unconsciously, but what I am conscious of in this situation, is vastly greater than 20-bits per second. At twenty bits per second I could only perceive the color of their jacket, nothing else, no sound, no meaning, nor the color of their trousers or skin. There just isn't the bandwidth to apprehend such things at 20 bits a second.
Okay, lets say that the human visual system actually works at only 8 bits (255 colors) most of the time, and only ups to higher bitrates when we focus our attention. This would still mean I could apprehend only 3 colors when someone walked in the room. Not three colors and something else, like shape, sound, the persons expression, their sex, no, just literally three colors, maybe the color of their face, jacket, and trousers. I wouldn't percieve face, jacket, and trousers as being those things, that's just not supportable at this bit rate.
Human beings are constantly aware of vast amounts of information. I would say that if they are aware of it, then it's conscious. They are not aware of the various processing steps that, say, resolve a complex pattern of colors into a face, but they are aware of the face, the expression on it, how they feel towards it, what sounds are coming out of it, etc, etc, etc. This requires hugely more than 20 bps?