We're Not Computers 1: I Just Do Eyes
Ever since we invented computers we've started to think of our own brains as computers, dubbing the computers themselves "electronic brains".
Neither is the other. Whilst I have no doubt that one day we will invent thinking machines, sentient intelligences with no appreciable difference from our own minds, at the moment we're barking up the wrong tree. In fact I would go so far as to say we're barking up a tree in the wrong forest. On a different continent.
This is because the way we think about our minds and brains isn't really how they work. Perhaps unsurprising seeing as it's the brains themselves doing this thinking. We imagine we are organic computers and therefore built electronic boxes in our own mental image; how we thought of ourselves at the time. If we'd been able to build computers in an earlier age, say when the heart was imagined to be the seat of consciousness, perhaps our computers would have resembled pumps.
To demonstrate the difference between the way we think we are and the way we actually are let's think about vision. It's the most advanced of our senses and the one with which we identify the most. It's also one we probably imagine we have already duplicated; after all the camera has been around since the middle of the nineteenth century. But the first fundamental error in the way we think about ourselves is right there. A camera is indeed a mechanical eye. However, what we do with the information once it's been detected by our retinas has yet to be duplicated by any machine. We can do eyes but we can't do vision.
Thanks to digital cameras we're all familiar with the concept of megapixels. The more megapixels there are the better the camera. The higher the resolution the clearer the picture.
This isn't how human vision works at all though. There is no inner screen onto which the pictures from the eyes are being projected, and no "resolution". Most of what we think we see is filled in by our brains afterwards; the retinal data is passed through a series of filters each of which tries to guess what it is looking at. This is part of The Question Machine, something that will be familiar to the longer term readers of this blog. If for some reason this machine failed you'd find yourself "looking at" a meaningless pattern of colours and shapes.
The human field of vision is surprisingly small. Whilst we may think we can see 180 degrees in front of us, most of this is appalling resolution and it's only around the fovea that we can see anything in detail. The eye flicks across the peripheries as well to pick up more detail but in general what you think you can see out of the corner of your eye is only a guess.
The strange thing though is that we're not aware of our field of view being a small circle of sharp surrounded by a field of increasing blurred. The brain guesses what the rest of your field of vision is and tells you so in no uncertain terms. As a result, the small circle of sharpness doesn't feel like an island of clarity in a sea of fog but more a focus of attention.
You can get an idea of how appalling peripheral vision is by using a simple Playing Card Test as described by Daniel Dennett in Consciousness Explained. Pick a card, any card. Don't look at it. Instead pick it up and hold it at arm's length with the value pointing away from your eyes. Move your straightened arm over to one side and only then turn the card around.
Don't look directly at it. Keep looking straight ahead. You can see there's something there. You know it's a playing card so you can probably see a playing card. You can't see the value though.
Start slowly moving your arm - still straightened - round to the front. It takes ages before you can even tell whether it's a red, black or picture card, let alone the value. You may be shocked to find that the card is almost directly in front of you before you can tell what the value is.
This is why something new appearing in the corner of your eye can be so alarming. A sudden movement could be anything and if it moves too fast for you to catch when you turn your head you might find yourself making up all sorts of stuff. Usually it's best to assume the worst which might explain why people are always seeing monsters out of the corner of their eyes rather than harmless fluffy marshmallow like creatures.
A digital camera that worked to this resolution probably wouldn't sell very well, but whilst a digital camera is a more efficient eye it's what happens in the brain that counts. So although we will be able to improve ourselves by upgrading our peripherals it may be a while before we're able to transfer our consciousnesses to artificial brains, because just as the vision isn't photography, likewise thinking isn't processing.
We are not computers (yet) and until we start making computers that use guessing to make sense of the world around them, then computers will not be us either.
In part two I will look more closely at how the brain works and compare it to the machine on which you're reading this
Neither is the other. Whilst I have no doubt that one day we will invent thinking machines, sentient intelligences with no appreciable difference from our own minds, at the moment we're barking up the wrong tree. In fact I would go so far as to say we're barking up a tree in the wrong forest. On a different continent.
This is because the way we think about our minds and brains isn't really how they work. Perhaps unsurprising seeing as it's the brains themselves doing this thinking. We imagine we are organic computers and therefore built electronic boxes in our own mental image; how we thought of ourselves at the time. If we'd been able to build computers in an earlier age, say when the heart was imagined to be the seat of consciousness, perhaps our computers would have resembled pumps.
To demonstrate the difference between the way we think we are and the way we actually are let's think about vision. It's the most advanced of our senses and the one with which we identify the most. It's also one we probably imagine we have already duplicated; after all the camera has been around since the middle of the nineteenth century. But the first fundamental error in the way we think about ourselves is right there. A camera is indeed a mechanical eye. However, what we do with the information once it's been detected by our retinas has yet to be duplicated by any machine. We can do eyes but we can't do vision.
Thanks to digital cameras we're all familiar with the concept of megapixels. The more megapixels there are the better the camera. The higher the resolution the clearer the picture.
This isn't how human vision works at all though. There is no inner screen onto which the pictures from the eyes are being projected, and no "resolution". Most of what we think we see is filled in by our brains afterwards; the retinal data is passed through a series of filters each of which tries to guess what it is looking at. This is part of The Question Machine, something that will be familiar to the longer term readers of this blog. If for some reason this machine failed you'd find yourself "looking at" a meaningless pattern of colours and shapes.
The human field of vision is surprisingly small. Whilst we may think we can see 180 degrees in front of us, most of this is appalling resolution and it's only around the fovea that we can see anything in detail. The eye flicks across the peripheries as well to pick up more detail but in general what you think you can see out of the corner of your eye is only a guess.
The strange thing though is that we're not aware of our field of view being a small circle of sharp surrounded by a field of increasing blurred. The brain guesses what the rest of your field of vision is and tells you so in no uncertain terms. As a result, the small circle of sharpness doesn't feel like an island of clarity in a sea of fog but more a focus of attention.
You can get an idea of how appalling peripheral vision is by using a simple Playing Card Test as described by Daniel Dennett in Consciousness Explained. Pick a card, any card. Don't look at it. Instead pick it up and hold it at arm's length with the value pointing away from your eyes. Move your straightened arm over to one side and only then turn the card around.
Don't look directly at it. Keep looking straight ahead. You can see there's something there. You know it's a playing card so you can probably see a playing card. You can't see the value though.
Start slowly moving your arm - still straightened - round to the front. It takes ages before you can even tell whether it's a red, black or picture card, let alone the value. You may be shocked to find that the card is almost directly in front of you before you can tell what the value is.
This is why something new appearing in the corner of your eye can be so alarming. A sudden movement could be anything and if it moves too fast for you to catch when you turn your head you might find yourself making up all sorts of stuff. Usually it's best to assume the worst which might explain why people are always seeing monsters out of the corner of their eyes rather than harmless fluffy marshmallow like creatures.
A digital camera that worked to this resolution probably wouldn't sell very well, but whilst a digital camera is a more efficient eye it's what happens in the brain that counts. So although we will be able to improve ourselves by upgrading our peripherals it may be a while before we're able to transfer our consciousnesses to artificial brains, because just as the vision isn't photography, likewise thinking isn't processing.
We are not computers (yet) and until we start making computers that use guessing to make sense of the world around them, then computers will not be us either.
In part two I will look more closely at how the brain works and compare it to the machine on which you're reading this
Comments