topics: |  vision | cognitive |

A fascinating look at the many ways in which vision failures tell us about the integration between how we process images and derive conscious information from them vs. how we act on them.

Excerpts

[patient Dee has lost partial vision after a CO poisoning attack]

As Mrs Fletcher passed a cup to her daughter, Dee said something rather startling. 'You know what's peculiar, Mum?' she said. 'I can see the tiny hairs on the back of your hand quite clearly!' This surprising remark led her mother to think that perhaps Dee's sight was on the road to a full recovery. But her pleasure was shortlived, when Dee added that despite seeing those fine details, she could not make out the shape of her mother's hand as a whole. In fact it soon became apparent that Dee was completely lost when it came to the shape and form of things around her. Unless an object had a distinctive color, or visual texture or grain, she had no idea what it was.

Inability to see shapes

though she could not use shape to tell one object from another, she could
still use their surface detail and color. ... in formal testing she could not
only name colors correctly but was also able to make fine discriminations
between different shades of the same color. So she might say that an object
was made of red plastic or out of shiny metal — but at the same time she
could only guess at its shape or function. p.7

Her difficulty in telling even horizontal from vertical lines shows just how
extreme this deficit was. 8

We discovered that Dee even had problems in separating an object from its
background — a basic first step for the brain in working out what an object
is. Dee said that objects seemed to 'run into each other', so that two
adjacent objects of a similar color such as a knife and fork will often look
to her like a single indefinable 'blob'. 9

Dee had difficulty naming even the simplest geometrical shapes, like a
triangle, a square, an oblong or a diamond. We began by showing her line
drawings of shapes, or filled-in black shapes on a white background. But she
was no better when we showed her shapes that differed from their backgrounds
in color instead of brightness. In other words, although she could see the
colors all right, she couldn't make out the edges between them.  Neither
could she recognize a shape made up of random dots where the dots making up
the shape were textured differently from the background. Nor could she see
'shape from motion' where a patch of dots is moved against a background of
stationary dots. 10

Drawings: copying vs. memory

Dee was able to recognize none of the three drawings on the left. In fact
as the middle column shows, she could not even make recognizable copies of
the drawings.  ... on another occasion, when asked to draw (for example) an
apple from memory, she produced reasonable renditions, as shown in the
right-hand column. Dee was presumably able to do this because she still has
memories of what objects like apples look like.  Yet when she was later
shown her own drawings from memory, she had no idea what they were.

Also, despite her copying problems, Dee can draw pictures of many common
objects from memory. For example, when asked to 'draw an apple' or 'draw a
house', she does this quite well. Her drawings are by no means perfect, but
then it is almost as if she is drawing with her eyes closed, because she does
not appreciate visually what she is drawing.

Imagery

Suppose you are asked to say whether a particular animal has a tail that is
longer than its body.You will probably do this by conjuring up a visual image
of the animal. Examining this image allows you to say that a mouse, for
example, has a tail longer than its body, while a cow does not. Dee can do
this just as well as most people, and unfailingly comes up with the right
answers. She can even do things involving more complex mental
operations. Take the following case: 'Think of the capital letter D; now
imagine that it has been rotated flat-side down; now put it on top of the
capital letter V; what does it look like?' Most people will say 'an ice-cream
cone' — and so does Dee.

So Dee can imagine things that her brain damage prevents her from
seeing. This must mean that manipulating images in the mind's eye does not
depend on exactly the same parts of the brain that allow us to see things out
there in the world. After all, if visual imagination did depend on those
brain structures, then Dee should not have been able to imagine things at
all — at least visually.

she often reports experiencing a full visual world in her dreams, as rich
in people, objects, and scenes as her dreams used to be before the
accident.  Waking up from dreams like this, especially in the early years,
was a depressing experience for her. Remembering her dream as she gazed
around the bedroom, she was cruelly reminded of the visual world she had
lost.

Visual agnosia

Dee's basic problem is in recognizing shapes. In cases such as hers,
where brain damage causes a disturbance in people's ability to
recognize things, the disorder is known as 'agnosia'. This term
was coined in the late nineteenth century by a then little-known
neurologist named Sigmund Freud. He borrowed two elements
from the ancient Greek (a = not, and gnosis = knowledge), in
order to convey the idea that patients of this kind have a problem
in making sense of what they see. 12


   Efron's rectangles: [Dee was shown a square, with some other rectangle of
   the same area but differing aspect ratios]. She was asked to say whether
   the two shapes were the same or different. Her answers - 9/20 or 11/20
   correct.  upto 17/20.  She sometimes even made mistakes when we used the
   most elongated rectangle, despite taking a long time to decide.

Paul Efron:  tested on Mr. S -patient similar to Dee -CO poisoning, could
recog colours, but no shape --> VISUAL FORM AGNOSIA.

Efron's test: level of difficulty could be scaled, so that he could compare
the degree of deficit in different patients.
The rectangles could be distinguished only by attending to their relative
dimensions, not to their overall size.

2 Doing without seeing

[Was testing her with familiar objects.]
When we held up a pencil, we were not surprised that she couldn't tell us
what it was, even though she could tell us it was yellow.

In fact, she had no idea whether we were holding it horizontally or
vertically. But then something quite extraordinary happened.  Before we knew
it, Dee had reached out and taken the pencil.  [This required her to form the
hand so that it knew how the pencil was oriented etc.]

Yet it was no fluke: when we took the pencil back and asked her to do it
again, she always grabbed it perfectly, no matter whether we held the pencil
horizontally, vertically, or obliquely.

Dee's ability to perform this simple act presented a real paradox.  How could
she see the location, orientation, and shape of the pencil well enough to
posture her hand correctly as she reached out to grasp it, while at the same
time she couldn't tell us what she saw? She certainly could not have grasped
the pencil accurately without using vision.

At a picnic

The contrast between what she could perceive and what she could actually do
with her sense of vision could not have struck us more forcibly than it did
one day when a group of us went out on a picnic while visiting her in
Italy.We had spent the morning at her home carrying out a series of visual
tests, recording one failure after another.

To lighten the gloom, Carlo suggested that we all go for a picnic in the
Italian Alps, to a popular spot not far from their home.

To reach the meadow, we had to walk along a half-mile trail through a dense
pine forest. The footpath was steep and uneven. Yet Dee had no trouble at
all. She walked confidently and unhesitatingly, without stumbling, tripping
over a root, or colliding with the branches of the trees that hung over the
path.

We eventually arrived at the meadow and began to unpack the picnic
hamper. Here Dee displayed once more how apparently normal her visual
behavior was. She reached out to take things that were passed to her with the
same confidence and skill as someone with completely normal sight. No-one
would ever have guessed that she could not see the difference between a knife
and a fork, or recognize the faces of her companions.

Refining a test first described by Marie-Thérèse Perenin and Alain Vighetto
(see Chapter 3), we set up a simple piece of apparatus where we could ask Dee
to 'post' a card into an open slot — like a mailbox, but with the added feature
that the slot could be presented at different orientations, not just at the
horizontal.

img/goodale-milner-2004_visual-agnosia-but-card-insertion-action-ability
In the 'matching' task, we asked her to tell us the orientation by simply
lifting the card up and turning it to match the orientation of the slot, but
without making a reaching movement toward the slot. Here we were not asking
her to use words to tell us what she saw, but to use a hand movement to show
us what she saw.

In the 'posting' task, she was asked to reach out and 'post' the
card into the slot.

Dee had no problem with the posting task, but performed almost randomly
on the matching task.

Further, askied her to imagine a slot at different orientations. Once she
had done this, she had no difficulty rotating the card to show us the
orientation she had been asked to imagine. It was only when she had to look
at a real slot and match its orientation that her deficit appeared.

The semantics of action

When you pick up a knife, you usually pick it up by the handle, not the
blade. When you pick up a screwdriver you do the same thing even though,
unlike the knife, there is no danger of cutting yourself. In other words,
many objects, especially tools, elicit 'use-appropriate' postures. Even
when the screwdriver is positioned with its handle pointing away from you,
you will typically turn your hand right around in a slightly awkward
fashion and grasp it by the handle as if you were about to use it.

When preoccupied with a memory task, however, you may pick up the screwdriver
by its shaft, using a grasp which is well-shaped but unrelated to tool
use. Dee, who cannot recognize screwdrivers and other tools, shows a similar
tendency to pick them up efficiently but inappropriately when the handle is
facing away.

img/goodale-milner-2004_2-1-visual-agnosia-grasping-horiz-vertical-pencil-correctly
	Dee was offered a pencil either vertically or horizontally. Though
	she could only guess whether it was V or H, she always grasped it
	perfectly.


Marie-Thérèse Perenin and Alain Vighetto discovered that patients with 'optic
ataxia' not only have problems reaching to point to something accurately, but
also tend to direct their hand at the wrong angle when trying to pass it
through a slot. The same patients, however, often have no problem describing
the orientation of the slot in words.

img/goodale-milner-2004_visual-agnostic-vs-controls-reaching-for-square-or-elongated

Dee and two healthy control subjects picked up blocks placed in different
orientations on a table in front of them. The lines connect the points where
the index finger and thumb first made contact with the block. (The results at
the different orientations are all shown together on a standard drawing of
the block.) Just like the control subjects, when Dee reached out to pick up
the block that was nearly square, she was almost — though not quite — as likely
to pick it up lengthwise as widthwise. But with more elongated blocks, she
and the control subjects were progressively less likely to do that. None of
subjects ever tried to pick up the most elongated block lengthwise. In short,
Dee was able to take both the orientation and the shape of the block into
account in planning her movement, just like people with normal vision. From
Carey, D.P., Harvey, M., & Milner (1996).  Visuomotor sensitivity for shape
and orientation in a patient with visual form agnosia. Neuropsychologia, 34,
329–337 (Figure 3).

3 : Optical Ataxia: When vision for action fails

Balint's syndrome:
cases where a patient had a specific problem in translating vision into
action. Later work has gone on to show that at least some of these patients
show remarkably intact visual perception — despite having profound difficulties
performing even simple visually guided movements. In short, the clinical
picture they present is the mirror image of Dee Fletcher’s.


The Hungarian neurologist Rudolph Bálint was the first to document
a patient with this kind of problem, in 1909.The patient was a
middle-aged man who suffered a massive stroke to both sides of the
brain in a region called the parietal lobe (see Figure 3.1). Although
the man complained of problems with his eyesight, he certainly
was not agnosic in the way that Lissauer’s and Freud’s patients were.
He could recognize objects and people, and could even read. He did
tend to ignore objects on his left side and had some difficulty moving
his eyes from one object to another. But his big problem was
not a failure to recognize objects, but rather an inability to reach out
and pick them up. Instead of reaching directly toward an object, he
would grope in its general direction much like a blind man, often
missing it by a few inches. Unlike a blind man, however, he could
see the object perfectly well; he just couldn’t guide his hand toward
it. Bálint coined the term ‘optic ataxia’ (optische Ataxie) to refer to this
problem in visually guided reaching.

it turned out that the patient showed the problem only when he used his right
hand.  When he used his left hand to reach for the same object, his reaches
were pretty accurate.This means that there could not have been a general
problem in seeing where something was. In other words, this was not a
visuospatial deficit. After further testing, Bálint discovered that the man’s
reaching difficulty was not a purely motor problem either — some kind of
general difficulty in moving his right arm correctly. He deduced this from
asking the patient to point to different parts of his own body using his
right hand with his eyes closed: there was no problem.

What has gone wrong in optic ataxia?

It was not in fact until the 1980s that the true nature of optic ataxia
became apparent, in large part through the work of the French neurologists
Marie-Thérèse Perenin and Alain Vighetto.  They made detailed video
recordings of patients with optic ataxia in a number of different visuomotor
tests. Like Bálint, they observed that the patients made errors in reaching
toward target objects placed in different spatial locations. Nevertheless,
the patients were able to give accurate verbal descriptions of the relative
location of the very objects to which they could not direct their hand.

img/goodale-milner-2004_optical-ataxia-patients
    patients with ‘optic ataxia’ not only have problems reaching to point to
    something accurately, but also tend to direct their hand at the wrong
    angle when trying to pass it through a slot.

Of course when their hand made contact with
the disk they could correct themselves using touch...

Marc Jeannerod went on to show that the well-regulated patterns of movement that
typify the normal person’s reaching and grasping behavior were severely
disrupted in patients with optic ataxia.  Instead of first opening the hand
during the early part of the reach, and then gradually closing it as it moved
toward the target object, the optic ataxic patient would keep the hand widely
opened throughout the movement, much as a person would do if reaching
blindfolded toward the object


Vision evolved only because it somehow improved an animal’s fitness — in other
words, improved its ability to survive and reproduce. Natural selection, the
differential survival of individuals in a population, ultimately depends on
what animals do with the vision they have, not on what they experience. It
must have been the case therefore that vision began, in the mists of
evolutionary time, as a way of guiding an organism’s behavior. It was the
practical effectiveness of our ancestors’ behavior that shaped the ways our
eyes and brains evolved.There was never any selection pressure for internal
‘picture shows’—only for what vision could do in the service of external
action.This is not to say that visual thinking, visual knowledge, and even
visual experience did not arise through natural selection. But the only way
this could have happened is through the benefits these mental processes have
for behavior. 39

The origins of vision

A single-cell organism like the Euglena, which uses light as a source of
energy, changes its pattern of swimming according to the different levels of
illumination it encounters in its watery world. Such behavior keeps Euglena
in regions of the environment where an important resource, sunlight, is
available. But although this behavior is controlled by light, no one would
seriously argue that the Euglena ‘sees’ the light or that it has some sort of
internal model of the outside world. The simplest and most obvious way to
understand this behavior is that it works as a simple reflex, translating
light levels into changes in the rate and direction of swimming.

Of course, a mechanism of this sort, although activated by light, is far less
complicated than the visual systems of multicellular organisms.  But even in
complex organisms like vertebrates, many aspects of vision can be understood
entirely as systems for controlling movement, without reference to perceptual
experience or to any general-purpose representation of the outside world.

Vertebrates have a broad range of different visually guided behaviors. What
is surprising is that these different patterns of activity are governed by
quite independent visual control systems.  The neurobiologist David Ingle,
for example, showed during the 1970s that when frogs catch prey they use a
quite separate visuomotor control module from the one that guides them around
      visual obstacles blocking their path. These modules run on parallel tracks
from the eye right through the brain to the motor output systems that execute
the behavior.

Ingle demonstrated the existence of these modules by taking advantage of the
fact that nerves in the frog’s brain, unlike those in the mammalian brain,
can regenerate new connections when damaged. In his experiments, he was able
to ‘rewire’ the visuomotor module for prey catching by first removing a
structure called the optic tectum on one side. The optic nerves that brought
information from the eye to the optic tectum on the damaged side of the brain
were severed by this surgery. A few weeks later, however, the cut nerves
re-grew, but finding their normal destination missing, crossed back over and
connected with the remaining optic tectum on the other side of the brain. As
a result, when these ‘rewired’ frogs were later tested with artificial prey
objects, they turned and snapped their tongue to catch the prey — but in the
opposite direction (see Figure 4.1). This

img/goodale-milner-2004_4-1_frog-rewired-tectum-snaps-opposite-but-jumps-right
   The dissociation between prey-catching behavior and visually-guided
   barrier avoidance in a ‘rewired’ frog.
   Left: when "fly" shown opposite the missing optic
   tectum (circles), frog snaps at mirror-image
   point (crosses).
   Right: But escape jumping behaviour remains intact -
   directions in which the ‘rewired’ frog jumped in response to a
   gentle touch from behind in the presence of a barrier, always cleared the
   barrier successfully

This is because only the eye’s projections to the optic tectum were in fact
rewired: the other projections, including those supporting barrier avoidance
behavior, remained correctly hooked up. From Ingle, D.J. (1973).Two visual
systems in the frog. Science, 181, 1053–1055 (Figures 1 & 2).

BOX: Routes from the eye to the brain

Neurons in the retina send information to a number of distinct target areas
in the brain. The two largest pathways from the eye to the brain in humans
and other mammals are the ones projecting to the superior colliculus (SC) and
the dorsal part of the lateral geniculate nucleus in the thalamus (LGNd). The
pathway to the SC is a much more ancient system (in the evolutionary sense)
and is the most prominent pathway in other vertebrates such as amphibians,
reptiles, and birds.The SC (or optic tectum, as it is called in non-mammalian
animals) is a layered structure forming the roof (Latin: tectum) of the
midbrain. It is interconnected with a large number of other brain structures,
including motor nuclei in the brainstem and spinal cord. It also sends inputs
to a number of different sites in the cerebral cortex. The SC appears to play
an essential role in the control of the rapid eye and head movements that
animals make toward important or interesting objects in their visual world.

The pathway to the LGNd is the most prominent visual pathway in humans and
other higher mammals. Neurons in the primate LGNd project in turn to the
cerebral cortex, with almost all of the fibers ending up in the primary
visual area, or striate cortex (often nowadays termed area V1) in the
occipital lobe. This set of projections and its cortical elaborations
probably constitute the best-studied neural system in the whole of
neuroscience. Scientists’ fascination with the so-called ‘geniculo– striate’
pathway is related to the fact that our subjective experience of the world
depends on the integrity of this projection system (see the section on
‘Blindsight’ in Chapter 5).

Although the projections to the SC and LGNd are the most prominent visual
pathways in the human brain, there are a number of other retinal pathways
that are not nearly so well studied as the first two. One of the earliest
pathways to leave the optic nerve consists of a small bundle of fibers that
project to the so-called suprachiasmatic nucleus (SCN).The visual inputs to
the SCN are important for synchronizing our biorhythms with the day–night
cycle.

There are also projections to the ventral portion of the lateral geniculate
nucleus (LGNv), the pulvinar nucleus and various pretectal nuclei, and a set
of three nuclei in the brainstem known collectively as the nuclei of the
accessory optic tract (AOT). The different functions of these various
projections are not yet well understood — although they appear to play a
critical role in the mediation of a number of ‘automatic’ reactions to visual
stimuli.The AOT have been implicated in the visual control of posture and
certain aspects of locomotion, and have been shown to be sensitive to the
optic flow on the retina that is created as we move through the world. The
AOT also plays an important role in controlling the alternating fast and slow
eye movements that we make when looking at a large visual stimulus, such as a
train, passing before our eyes. Retinal projections to one area in the
pretectum are thought to be part of the circuitry controlling the pupillary
light reflex — the constriction of the pupil as we move into a brightly lit
environment such as that found on the beach or the ski slopes.There is also
some evidence from studies in amphibians and lower mammals that certain
pretectal nuclei play a role in visually guided obstacle avoidance during
locomotion. However almost nothing is known about the functions of the other
pretectal nuclei, the ventral part of the lateral geniculate nucleus, or the
pulvinar.

The fact that each part of the animal’s behavioral repertoire has its own
separate visual control system refutes the common assumption that all
behavior is controlled by a single general-purpose representation of the
visual world. Instead, it seems, vision evolved, not as a single system that
allowed organisms to ‘see’ the world, but as an expanding collection of
relatively independent visuomotor modules.

Vision for perception

Of course, in complex animals such as humans and other primates, such as
monkeys, vision has evolved beyond a set of discrete visuomotor modules. Much
of our own behavior is certainly not rigidly bound by our sensory input. Even
frogs can learn to some degree from their previous visual encounters with the
world — but humans and other higher primates can use their previous visual
experience and knowledge of the visual world in much more flexible ways so as
to guide what they do in the future.We can internally rehearse different
courses of action, for example, often using visual imagery in doing so,
before deciding what to do.

In other words, vision can serve action not just in the here and now, but
also ‘off-line’—at other times and in other places.To do this, the visual
brain creates a rich and detailed representation of the visual scene that the
animal is looking at. We do not know what animals experience, but in humans
at least, these perceptual representations are normally conscious.We
experience them, and thereby we can communicate them to others. The visual
mechanisms that generate these representations are quite different from the
simple visuomotor modules of amphibians described earlier, and appear to have
arisen more recently in evolutionary time.  Rather than being linked directly
to specific motor outputs, these new mechanisms create a perceptual
representation that can be used for many different purposes. Moreover, as we
mentioned in Chapter 1, our perception of the world is not slavishly driven
by the pattern of light on the eye but is also shaped by our memories,
emotions, and expectations. Visuomotor mechanisms may be driven largely
bottom-up but perception has an important top-down component as well. The
memories that affect our perception in this top-down way are themselves built
up from previous perceptions. As a result of all this two-way traffic,
perception and memory literally blend into one another. After all, we have
visual experiences in our dreams, and these must be generated entirely by
top-down processes derived from memory.

It is important to bear in mind that when people talk about what they ‘see’,
they are talking only about the products of their perceptual system. Yet
until recently researchers on vision have seen no need to go further than
perceptual reports when gathering their data. In fact, a very important
tradition in visual research, called psychophysics, depends entirely on what
people report about what they can and cannot see. It has always been assumed
that this is all there is to vision. Admittedly, psychophysics, which was
founded by the nineteenth-century German physicist turned philosopher, Gustav
Fechner, has told us a great deal about the capacities and limits of the
perceptual system. But it has told us nothing about how vision controls the
skilled movements that we make.
... because the visuomotor machinery governing our actions is subconscious.

Vision for action
Alongside the evolution of perceptual systems in the brains of
higher mammals such as humans, the visuomotor systems in turn
have become progressively more complex. The main reason for
this is that the movements we make have themselves become more
complex. In our primate ancestors, one of the great landmarks in
evolution was the emergence of the prehensile hand — a device that
is capable of grasping objects and manipulating them with great
dexterity. But just as the development of any sophisticated piece of
machinery, such as an industrial robot, needs an equally sophisticated
computer to control it, the evolution of the primate hand
would have been useless without the coevolution of an equally
intricate control system. The control of eye movements too has
become more sophisticated and has become closely linked with
the control of our hand movements. All of these changes, in other
words, were accompanied by the evolution of new brain circuitry.

Many of these new control systems in the brain have strong links to and from
the basic modules in those older parts of the brain that were already present
in simpler vertebrates like frogs and toads.

[saccadic] eye movements in primates,
such as monkeys and humans.

head and eye movements in rodents are controlled by the same basic structures
(the optic tectum, or superior colliculus) that control prey-catching in
frogs.  These same structures retain a central role in the machinery that
programs head and eye movements in primates.

At first sight this may seem a puzzle — why didn’t nature devise totally new
systems from the ground up? In his book Evolving Brains, the American
neurobiologist John Allman tells the story of how, on a visit to a power
generation plant during the 1970s, he was struck by the side by side
coexistence of several control systems for the generators dating from
different periods in the life of the plant. There were pneumatic controls and
a system of controls based on vacuum tube technology, along with several
generations of computer-based control systems. All of these systems were
being used to control the processes of electrical generation at the
plant. When he asked the reason for this strange mix, he was told that the
demand for power had always been too great for the plant ever to be shut
down. As Allman points out:

	The brain has evolved in the same manner as the control systems in
	this power plant. The brain, like the power plant, can never be shut
	down and fundamentally reconfigured, even between generations. All
	the old control systems must remain in place, and new ones with
	additional capacities are added and integrated in such a way as to
	enhance survival.

img/ventral-and-dorsal-streams-in-monkey
Two streams theory - Ungerleider and Mishkin’s (1982)]
ventral stream receives most of its visual input from the
primary visual cortex (V1), which in turn receives its input from the lateral
geniculate nucleus (LGNd) of the thalamus. The dorsal stream also receives input
from V1, but in addition gets a substantial input from the superior colliculus (SC)
via the pulvinar (Pulv), another nucleus in the thalamus. From Milner, A.D. &
Goodale, M.A. (1995).Visual Brain in Action, Oxford University Press (Figure
Old World monkey. The
3.1).

img/goodale-milner-2004_4-4_huber-wiesel-V1-responses-to-edges

img/goodale-milner-2004_4-6_neuroms-specific-to-grasp-shape
4.6 neuron in area AIP when a monkey looks at and then grasps six different
kinds of solid shapes. As the graphs below each shape show, the neuron
responds best when the monkey grasps a vertically-oriented square plate. The
neuron begins to fire when the monkey is first shown the object (marked ‘fix’
for fixation) and continues to fire after the monkey has grasped it (marked
‘hold’). From Murata, A., Gallese,V., Luppino, G., Kaseda, M., & Sakata,
H. (2000). Selectivity for the shape, size, and orientation of objects for
grasping in neurons of monkey parietal AIP. Journal of Neurophysiology, 83,
2580–2601 (Figure 4).

img/goodale-milner-2004_6-3-virtual-mirror-grasping-setiup

img/goodale-milner-2004_6-4-perception-comparative-doesnt-affect-reach
Visual displays shows target block accompanied by another.
If the other is bigger, the target seems smaller and vice versa.
However, this perception does not affect the reach opening.
6.4

Ebbinghaus illusion

img/goodale-milner-2004_6-6-ebbinghaus-illusion-doesnt-affect-reach-B

img/goodale-milner-2004_6-7-ebbinghaus-illusion-grasp-aperture-is-not-fooled
   graph showing the changing distance between the finger and thumb (grip
   aperture) as a typical subject reaches out to pick up two disks, one of
   which is physically larger than the other.  Even though the subject
   believed that the two disks were the same size, their visuomotor system
   was not fooled by the illusion. In other words, they scaled their grip
   aperture to the real size of the objects.

TODO: From Aglioti, S., DeSouza, J., & Goodale, M.A.  (1995). Size-contrast
illusions deceive the eyes but not the hand. Current Biology, 5(6), 679–685
(Figure 4).



the illusion is evident even in a matching task in which they opened their
index finger and thumb to match the perceived diameter of one of the disks.

but their grip aperture completely resisted the illusion and was instead
tailored to the real size of the disk when they reached out to pick it up

img/goodale-milner-2004_6-8-other-illusions-that-do-not-fool-grasping
   6.8 Three other pictorial illusions that have been found to have little
   effect on the visuomotor system.  The well-known Müller–Lyer illusion is
   shown at the top.  The two lines between the arrowheads appear to be
   different lengths. In the horizontal–vertical illusion (bottom left) the
   vertical line appears to be longer than the horizontal. In the Ponzo (or
   railway-lines) illusion, the upper bar appears to be longer than the lower
   bar. In all of these illusions, the pairs of lines (or bars) are actually
   identical in length. When these lines are replaced by solid rods or bars
   and subjects asked to pick them up end to end, although the illusion is
   still present perceptually, it has little effect on the in-flight scaling
   of their grasp.

img/goodale-milner-2004_6-9_rod-frame-illusion-doesnt-affect-but-tilt-does
   6.9 The ‘rod and frame illusion’ (top) and the ‘simultaneous tilt
   illusion’ (bottom). In both illusions we see the central line or stripes
   to be tilted in opposite directions according the tilt of the frame (top)
   or striped background (bottom). The reasons for this, however, are
   different for the two illusions. The rod and frame illusion depends on the
   same kinds of perceptual mechanisms as the pictorial illusions already
   mentioned, and like those affects perception but not action. The
   simultaneous tilt illusion, however, is probably the result of local
   effects within the primary visual cortex, and therefore would be passed on
   to both visual streams. As a consequence this illusion affects both
   perception and action.

From preface

We published two papers on this work in early 1991; but while they were in
press, we became aware of a quite independent investigation of single neurones
in the posterior parietal cortex of the monkey by a group of Japanese
physiologists. This showed that not only was the machinery present in this
part of the brain for organizing such simple visuomotor acts as watching and
reaching toward objects shown to the animal, but also for visually controlling
the monkey’s grasp to match the particular sizes and shapes of objects.We
were well aware of the pioneering and highly influential work of Mishkin,
Ungerleider, and their colleagues at N.I.H., in which they first identified and
then systematically explored the two major streams of visual processing in the
cerebral cortex.

[why are the japanese researchers not mentioned while N.I.H. work is referred
by name?]
(possibly [H. Sakata etal 1992])
Hand-movement-related neurons of the posterior parietal cortex of the
monkey: their role in the visual guidance of hand movements.
in Control of Arm Movement in Space, R. Caminiti (ed)

Ungerleider, L.G. and Brody, B.A. (1977).
Extrapersonal spatial orientation: the role of posterior parietal,
anterior frontal, and inferotemporal cortex. Exp. Neurol., 56, 265–80.