Mel Goodale is a psychologist with a neat trick. In his lab at the University of Western Ontario in London, Ontario, he has a selection of circular blocks shaped like poker chips, which he likes to lay out on a table to create an optical illusion. The configuration must be precise: he puts one chip inside a ring of chips that are much bigger than it is, and another, of identical diameter, inside a ring of much smaller chips. When asked to judge the relative sizes of the two central chips, people always say that the one surrounded by the smaller circles is larger than the other. But even though their conscious brains fall for the illusion every time, their bodies are not so easily fooled: Goodale has videotaped evidence showing that when people reach out to grasp the chips, their hands open equally wide for either one.
From this one simple experiment flow at least three ego-deflating insights. Your unconscious brain and body â alias the Zombie â know things about the world that you donât. What you see âout thereâ is often illusion, not reality. And to cap it all, You Yourself are not fully in charge of your own perceptions and actions: you have no choice about seeing the optical illusion, yet your hands open unconsciously to the proper size to grasp the chips. The Zombie within, not the conscious self, is running the show.
Advertisement
And as psychologists like Goodale are discovering, this metaphorical Zombie knows things â and can do things â that the conscious self is not merely too slow to deal with, but too gullible or simply too preoccupied. Its true strong suit â at least so far as researchers have learnt so far â lies in using vision to control movement. But the real story doesnât begin with poker-chip illusions in labs. It begins in the north of Italy, with a broken water heater and a woman known to science as DF, whom we shall call Diane.
Taking a shower one morning in her home, Diane collapsed from carbon monoxide poisoning. The accident left her with a damaged visual cortex and an inability to consciously perceive shapes or objects. âItâs a very difficult thing to get your mind around,â admits David Milner, the neuropsychologist who was called in to test Dianeâs vision when she came to live in Scotland after the accident. âYou have to imagine what it would be like to see that an object is green but not see it as an object.â
Diane cannot say how big an object is or which side of a rectangle is longer. She cannot even tell vertical lines from horizontal or oblique ones. But, extraordinarily, she can grasp objects cleanly at almost every try.
At Milnerâs lab at the University of St Andrews in Fife, and later in Goodaleâs lab in Canada, more extensive testing showed that Diane could twist her hand properly while fitting it in a narrow slot, and she could open her grip to the right width when picking up objects of different sizes. But paradoxically, she could not use her hand to match the orientation of the slot or open her grip to indicate the size of an object unless she actually performed the task.
So, here is a woman who is blind to objects and yet can reach out and grasp a pencil in a single, fluid and accurate movement, almost as though she has some kind of extrasensory perception. Whatâs going on?
Forget common sense notions about what vision involves. What patients like Diane prove is that seeing cannot simply involve scanning a mental picture of some kind, and vision cannot consist of just one perceptual skill. Other patients have the opposite impairment from Diane â they can perceive objects clearly but have lost the ability to guide their movements properly. To Milner and Goodale, this suggested that the brain must contain two pathways for processing visual information, one for action and one for conscious perception and description.
âVision for action is very different from vision for perception,â says Goodale. The conscious perceptual system â the one Yourself uses â describes objects so the mind can remember, compare, and think about them. To be useful, the perceptions must be abstract so we recognise the same object seen from different angles or in different light. This necessarily involves a lot of assumptions about the raw data. âIf you see a small elephant on your retina, you assume itâs a large elephant some distance away because you have all those assumptions about elephants being big,â says Goodale.
By contrast, the visuomotor system â the one the Zombie uses â needs to know not what the object is, but exactly how big it is and where it is located relative to the body. âItâs there to guide you without your having to take time out to think about things. If youâre a monkey jumping from branch to branch in a tree, itâs pretty important to get your arm in the right place at the right time, or youâre going to die,â says Milner.
With two such different needs, one might expect two different visual systems to have evolved in the brain. If so, Milner and Goodale reasoned, it should be possible to tease the two apart in people without brain damage
Sure enough, experiments have done exactly that. Neuroanatomists have known for many years that the visual systemâs tangled web of interconnecting brain regions separates roughly into two pathways, a âhigh roadâ that leads to the rear of the parietal lobe of the cortex and a âlow roadâ to the bottom of the temporal lobe. Some evidence suggests that the upper path carries mostly visuomotor information while the lower path carries mostly conscious perception. Of course, âcross-talkâ between the two pathways is bound to occur. âYou pick up a screwdriver differently from the way you pick up a stick to put in the fire,â notes Goodale. âThat means that perceptual information must access the motor programs.â
Goodale hoped that his experiment with the poker-chip shaped blocks and the optical illusion would turn up evidence of the two visual systems in normal people â and it did, in spades. The volunteers were fooled into thinking the central blocks were different in size, while the visuomotor system knew they were the same. And when Goodale used central blocks that looked identical to the eye but really werenât, the volunteersâ hands knew the difference. The motor system avoids the optical illusion because it relies on binocular vision to calculate size and distance from first principles, Goodale thinks. Indeed, with one eye covered volunteers lost their ability to ignore the illusion and grasp correctly.
Virtually real
To test in more detail exactly which cues guide grasping, Goodale is developing a simulator that uses virtual reality to offer visual cues of his own design â cues that do not necessarily conform to real-world laws of physics, such as impossibly-shaped shadows. The system has proven to be tricky to build because the visuomotor Zombie is such a stickler for realism that it refuses to play along unless the object can be felt as well as seen. âThe first time they grab it and their hand goes through it, the visuomotor system gives up and they revert to a pantomime mode, pretending to grasp,â says Goodale.
As a result, he has had to use a system of screens and mirrors so that subjects grasp a real, unseen object but see Goodaleâs virtual object in the same apparent location. The system is now working, and Goodale is beginning experiments in which he presents a small visual object together with a large tactile one to see whether observers can adjust their grasp when eyes and hands tell different stories.
Just as Dianeâs unconscious visuomotor pathway was useless for describing objects, Goodale found that the canât-fool-me Zombie within normal brains only works for purely physical tasks like reaching and grasping. When he asked the volunteers instead to open their fingers the correct width without actually reaching for the object â that is, to consciously match their hand to their perception â each one fell under the spell of the illusion. The same thing happened if volunteers had to wait in the dark more than two seconds before they began their reach. The motor Zombie, Goodale concludes, canât remember any further back than that and must resort to memories stored by easily duped conscious perception.
Grasping is not the only skill where the Zombie proves less gullible than Yourself. People tend to grossly overestimate the steepness of hills, says Dennis Proffitt, a psychologist at the University of Virginia in Charlottesville. For example, most people standing at the bottom of a hill with an incline of only 5 degrees will guess its steepness at 20 degrees, Proffitt found. But if asked to tilt their hand to match the steepness of the hill, they do it accurately, even without looking at their hand. Once again, the Zombie knows things that Yourself does not.
Tough climb
But here thereâs method in Yourselfâs madness. The volunteers in Proffittâs experiments judged hills to be even steeper if they wore a 30-kilo backpack or had just completed an hourâs exhausting run. Likewise, older volunteers and couch potatoes saw hills as steeper than younger people and trained athletes. In other words, your conscious selfâs overestimate of the hill is an accurate measure of how hard you will have to work to get your physical self to the top â conscious perception is a useful fiction that misrepresents reality in our interests. âYouâre seeing not just the hill, but your physiological potential,â says Proffitt. âThe more physically fit you are, the shallower the apparent terrain you live in.â
Out of shape
The conscious perception of shape, too, is prone to distortion. Studies by Jack Loomis of the University of California at Santa Barbara show that people looking at a three-dimensional figure usually underestimate its depth relative to its width. One easy illustration he gives is how much more prominent a personâs nose looks in profile compared with face-on. The same holds true for estimating distances and dimensions receding from you compared with going across your line of vision. The two table tops pictured (see diagram) have identical dimensions, but you wouldnât think so just by looking at them.FIG-mg21505603.jpg
âThese are very dramatic effects. Itâs very clear that we do not perceive the shapes of three-dimensional objects correctly,â says Loomis. Yet when Loomis and his colleagues asked volunteers to close their eyes and walk to a target a few metres away then turn and walk blindly to a second target, they walked the latter distance with equal accuracy in any direction. âThey donât show any tendency to walk a shorter distance in the depth interval than in the width interval,â says Loomis.
However, Loomis argues, this does not prove that accurate, unconscious sensations guide walking, while conscious perceptions slip into error. It may be that we merely perceive locations more accurately than we perceive dimensions, and use the former to guide our walking. âItâs very clear that when you look at something, close your eyes, and then move through space, your action is based on what you initially perceived. I think itâs a conscious perception,â he says.
Edited version
Other evidence suggests that the motor system â which often needs to act fast â may have no choice but to do without the rich detail and flexibility of conscious thought. For example, Yves Rossetti, a neuroscientist at the French National Institute for Health and Medical Research in Lyon, asked volunteers to point as quickly as possible to a dot when it appeared on a video screen, but not to follow the dot with their finger if it jumped to a second location. When the jump came soon after the dotâs appearance, he found that the volunteers could not stop their finger from following the dot. âWe can hear them swearing in each trial, because they can see the hand going where it shouldnât,â says Rossetti. The conscious effort to stop the finger kicked in too late to override the motor systemâs pointing, he concluded.
In classic experiments from the 1970s, Benjamin Libet of the University of California at San Francisco provided direct evidence that conscious perception is slow. Libet found that an electrode had to stimulate the sensory cortex for at least half a second before the patient became aware of any sensation, suggesting that conscious experiences take that long to develop. The delay, Libet thinks, may allow time for the conscious mind to edit and interpret the raw data. The various sensations that arise from a single event, for example, are likely to arrive at the brain at slightly different times, since touch sensations must travel all the way from the fingertips, while sight and sound have shorter paths. The curing time for conscious perception would allow the brain to gather up these disparate sensations and bind them into one perception. âBut during that half-second, all sorts of unconscious things can go on,â says Libet. âYou see your carâs going to hit something and you stamp on the brake within 100 to 150 milliseconds, well before youâre consciously aware of whatâs going on.â
This tension between the need for quick response and the need for sophisticated perception and decision-making is also important in robotics, says Goodale. At one extreme, robots with every move hardwired in can react very quickly to familiar objects and events, but are paralysed by novelty. At the other extreme, robots where a remote human operator controls every move can cope with new situations, but the operator often finds movements difficult to coordinate quickly from a distance. Many robotics designers opt for a two-stage solution to the problem: a remote human pilot makes strategic decisions, then hands control to the robot for the individual, stereotyped tasks.
And that, Goodale thinks, is exactly the solution evolution has adopted in the brainâs two visual systems. Yourself, the conscious mind, cranks through all its rich perceptual detail to create and make sense of the big picture. It identifies objects â this magazine, for example â and chooses the next action to be executed.
Damage to the parietal lobe on the right side of the brain often produces a condition known as neglect, in which patients lose awareness of the left half of objects. For example, they will eat food from only the right side of their plate. Strictly speaking, such people are not blind. Instead, they seem to suffer from an inability to focus their attention to the left if there are any competing stimuli on the right â as there always are in real life.
But the brain still processes information in the neglected field, experiments show. If people are shown the word âcatâ in their neglected field and asked to free-associate on the word âanimalâ shown to their good field, they are more likely to come up with âcatâ as their response.
Who decides?
The Zombie may do much more than direct the bodyâs motor machinery. A provocative experiment that was reported in the 1980s by Benjamin Libet of the University of California at San Francisco suggests that the Zombie may be in charge of the brainâs decision-making, too.
Libet asked human volunteers to make hand movements whenever they felt like doing so as he measured the electrical activity in their brains. He found that brain impulses associated with a movement began about 350 milliseconds before the subjects reported any conscious awareness of their intent to make the movement. That is, the âvoluntaryâ action did not originate consciously.
The conscious mind was not powerless, however. It could still veto the proposed movement during a window of 150 to 200 milliseconds between conscious awareness of the intention and its execution, he found.
This suggests that our conscious minds may not have free will, but rather âfree wonâtâ, according to the neurologist and psychologist Vilayanar Ramachandran of the University of California, San Diego.
Patients who suffer brain damage to the lower lateral region of the cortex, around the ear, sometimes lose the ability to identify colours or even tell the difference between two contrasting colours of the same brightness. But people with this disorder, known as cerebral achromatopsia, can still see the boundary between two abutting colours even though the two look identical, Charles Heywood of the University of Durham and his colleagues have found.
This suggests that the brain uses colour information in two different ways: to assign specific colours to particular objects (which would be evolutionarily useful in deciding whether a piece of fruit was ripe, for example) and to pick a figure out from its background (helpful, for example, in spotting a lion ready to pounce).
Brain damage that causes achromatopsia wipes out the first of these abilities in the conscious self but leaves the Zombie still able to perform the second, thinks Robert Kentridge, a psychologist at the University of Durham.
People unlucky enough to lose all feeling on one side of their body after a stroke display an ability called ânumbsenseâ. One patient, studied in detail by Yves Rossetti, a neuroscientist at the French National Institute for Health and Medical Research in Lyon, could not feel a pencil tip on his right arm and could not say when or where he had been touched.
But if forced to guess by pointing at the location with his intact left hand, he got closer than he would by guessing. Again, the Zombie within knows more than the conscious self-but only for physical movements.
Smell the difference
The unconscious mind can sometimes respond to stimuli that are too weak to make the leap into conscious awareness. For example, people forced to guess which of two odours is present can do so correctly even when the odour is too faint to detect consciously, according to experiments by Tomas Radil, a neuroscientist at the Institute of Physiology of the Academy of Sciences in Prague.
And many researchers have found that a word flashed before the eyes too briefly for conscious notice can change how a person responds to a second, similar word or word fragment presented consciously â a phenomenon known as priming.
But there is no good evidence that subliminal perceptions can have any lasting effect on a personâs behaviour, says Philip Merikle of the University of Waterloo in London, Ontario. A famous account from the 1950s, claiming that moviegoers bought more popcorn and soft drinks in response to subliminal messages embedded in the movie, has never been verified and was probably a complete fabrication, says Merikle.
In part, this may be because the stimuli psychologists usually use in such studies-random words or pictures, for example-are so unimportant the mind doesnât bother to retain them for more than a few seconds. More emotionally charged stimuli might produce more lingering effects, he thinks.
Some experiments using such stimuli are now under way. Ray Dolan of University College London and his colleagues flashed a picture of an angry face while using positron emission tomography to measure volunteersâ brain activity-especially in the amygdala, a brain region important in handling emotional stimuli. Photos flashed too quickly for conscious awareness triggered activity in the right amygdala, while photos presented slowly enough to be detected consciously activated the left amygdala, they found.
While this does not prove that unconscious perceptions are closely allied to the right brain and conscious ones to the left, it at least suggests that such a division is worth investigating.
