4.4 The Evolution of the Visual System

So, how did this wiring come into existence? And how, ultimately, did it

form into the complex wiring that comprises the human visual system,

eventually enabling humans to solve vision-related, nonroutine problems

creatively with the use of scenario visualization?

As noted earlier, organisms fi rst developed a light/dark sensitivity area.

We have already spoken about the euglena’s eyespot, but hydras and

worms also have light/dark sensitivity cells located in their outer epidermal

layers. Eventually, a pinhole eyespot formed, which enabled even more

complex processing of visual information. Of course, in order for this

eyespot to have occurred, other parts of a rudimentary nervous system had

to be in place, like the mass of neurons localized in one area, such as the

head.

For example, the nautilus—one of the so-called living fossils, because it

morphologically has not changed much since the Cambrian explosion—is

representative of this next step in the evolution of the eye. The evolution

of the visual system likely began with a general light sensitivity around the ventricle of the forebrain. These forebrain neurons were transformed

in the course of evolution into an assemblage of neurons capable of analyzing

strictly visual information, rather than other information carried to

this area by other input neurons. The processed visual information was

transferred through relay neurons (the future ganglion cells) to the contralateral

part of the brain. In subsequent generations, this primordial

retina evaginated (formed a cup-like hole) and eventually transformed into

the lateral eyes of vertebrates (Cronly-Dillon, 1991).

Once this rudimentary eye developed, it took different directions in

terms of its formation among the various species that have vision. There

are all types of eyes falling into two broad categories: the compound eyes

of insects, crabs, and arthropods that are composed of hundreds of cylindrical

elements consisting of a lens, eight photoreceptor cells, and a sleeve

of opaque pigment cells, and the vertebrate eyes of fi sh, reptiles, and

mammals that all build upon the basic fl atworm eye having a retina, ganglion

cells, and projections to a brain. This divergence further ratifi es the

evolutionary principles of genetic variability and the natural selection of

the fi ttest traits given a specifi c environment. The kind of eye that fortuitously

developed, and then fl ourished, in the sand shark population under

the sea will be different from the eye that fortuitously developed, and then

fl ourished, in the black widow spider population on land, and so on.

As noted already, through investigation of the fossil endocasts of insectivores,

Jerison (1973, 1991) has offered a convincing case that the ancestor

of the primate was a small, nocturnal creature that prowled through trees

and bushes on the lookout for insects, as some small primates do today.

The eyes of this creature were close set and forward facing, indicating that

it was a predator. Predators like owls, cats, and primates have close set,

forward facing eyes with stereotopic vision, enabling them to more easily

identify depths and distances while hunting their prey. Animals like rabbits

and squirrels—the prey of predators like owls, cats, and some primates—

have eyes on the sides of their heads with panoramic vision, enabling them

to more easily identify predators from a variety of angles. The eyes of this

insectivore would have had an abundance of rods and fewer cones to

enable maneuverability in the dark. Insectivores had to forage on the

ground and in the dark in order to steer clear of bigger reptiles, amphibians,

and dinosaurs that would have preyed on them during that time.

However, when most of the dinosaurs and other larger predators were

wiped out by an asteroid at the close of the Cretaceous, this allowed for

the possibility of insectivores to occupy different niches. Eventually, some

of the descendents of these creatures did their foraging during the day,

The Evolution of the Visual System and Scenario Visualization 107

while others took to life in the trees. Still others, like our primate ancestors,

did both. By 20 mya, Proconsul had taken to life in the trees of Africa, most

likely because such a life off of the ground made it easier to avoid predators

lurking on the ground, as well as pounce upon prey that happened to be

on the ground. This life in the trees is evidenced by their money-like and

ape-like skeletons as well as their diet, which indicate a lifestyle not unlike

that of present-day monkeys and apes (Groves, 2002; Hartwig, 2002;

McGrew, 2004; Byrne & Whiten, 1988; Allman, 1982, 2000).

Life during the daytime and in the trees meant that a visual system, at

the very least, would need to have evolved more cones, the capacity to

distinguish colors, more memory-storage capacities, and facial recognition.

More cones would be needed to be able to handle the brighter light of

daytime living. A capacity to distinguish colors would be needed to determine

what foods one could eat, as well as distinguish foreground from

background while maneuvering through the trees. Rudimentary forms of

long-term and short-term memory would be needed for a mental map of

the area where this creature lived, as well as for distinguishing friend from

foe. Finally, facial recognition would be needed to distinguish friend from

foe, as well as to communicate various messages like “Leave me alone,”

“Back off,” “It’s o.k. to touch me,” and the like to other members of one’s

social group or to enemies. On top of all of this, arboreal life required

extremities more fi t for grasping, jumping, launching, and other sorts of

bodily maneuvering (Barton, 1998; Fleagle, 1999; Byrne, 1995, 2001; Rilling

& Insel, 1998; Allman, 1977).

In order to do all of this, the connections between eye and brain, as well

as those connections between other sensorimotor systems and brain, would

need to be more complex. As Jerison (1973, 1997) and Kaas (1996) point

out, the visual system of this early arboreal ancestor would have had a

brain very similar to that of a lemur or a monkey. This being the case, at

least three features of the brain and visual system would have needed to

be in place in Proconsul.

First, referring back to MacLean’s model, the reptilian, the paleomammalian,

and part of the neomammalian brain would have to be present in

Proconsul, since the amount of processing that such arboreal ancestors had

to do required a brain larger than that of reptiles. The amount of processing

that the visual system does alone would constitute a bigger brain.

However, arboreal social life entailed the interaction of sensory modalities

in combination with memory, emotion, planning, and motor function. All

of this processing required a bigger brain, more synaptic connections, and

more types of neural connections. In addition, the system must have been

a hierarchical organization of processes and subsystems so that the right

data could be selected and information could be integrated as Proconsul

went about the business of negotiating its environment.

Second, in order to organize and coordinate information from the specialized

visual modules and each of the other sensory modalities, integrative

or association areas and systems had to evolve. Research by Kaas (1987,

1993, 1995, 1996), Bear et al. (2001), Uylings & van Eden (1990), Roberts,

Robbins, & Weiskrantz (1998), and Hofman (2001) has shown that it is

primarily the association areas of the brain that increase in the evolution

of the mammal. The association areas show a marked increase when we

compare the brains of rats, cats, macaque monkeys, and humans, in that

order. On inspection, the visual, auditory, and sensorimotor areas of a rat

basically butt up against one another with very little of the cortex—some

15%–20%—devoted to the association of these areas. Some 40%–50% of

the frontal, parietal, and temporal lobes of the macaque monkey are

devoted to association of sensorimotor information. Even more cortex—

some 50%–60%—is devoted to association of sensorimotor information in

humans.

Third, the visual areas would have needed to be linked up with the motor

areas, so that visual information could be communicated to the brain and

then translated into voluntary motion. The motor areas of primates receive

inputs from the thalamic nuclei that relay information from the basal telencephalon

and the cerebellum, and they send outputs to motor control

neurons in the brain and spinal cord. The motor area is situated next to

the somatosensory area, and both are located in the frontal lobe near the

central sulcus at the parietofrontal junction of the brain. The sensorimotor

area—along with the auditory and visual areas—send projections to association

areas in the parietal and temporal cortices. From there, the projections

are sent to the multimodal areas located in the parietotemporal,

prefrontal, and limbic cortices (Kandel et al., 2000). These connections

would have had to be in place in the brain of Proconsul so that it could

have acted on visual information it received.

In the last chapter, I mentioned the idea put forward by the Gestalt

theorists and William James that the mind and the world are “something

of a mutual fi t.” This idea should be kept in mind here as we investigate

the evolution of visual modules, along with the evolution of other sensory

modules and mechanisms of selectivity and integration. Events in the

physical environment are composed of materials that the human brain,

complete with its specialized modules, has evolved to perceive and discriminate.

These materials take forms ranging from specifi c chemicals, to

The Evolution of the Visual System and Scenario Visualization 109

mechanical energy, to electromagnetic radiation (light) and are discriminated

by the different sensory modalities that are specifi cally attuned to

these stimuli: taste and smell are specialized for processing concentrations

of different chemicals, touch is specialized for processing mechanical

deformations of the skin, hearing is specialized for the intake of sound

waves, and vision is specialized for the intake of electromagnetic radiation.

Thus, given the environment surrounding an animal, it makes sense that

the CNS is a hierarchically organized system composed of specialized

modules, as well as mechanisms of selectivity and integration, in order to

process the myriad pieces of environmental information.

So, how did this wiring come into existence? And how, ultimately, did it

form into the complex wiring that comprises the human visual system,

eventually enabling humans to solve vision-related, nonroutine problems

creatively with the use of scenario visualization?

As noted earlier, organisms fi rst developed a light/dark sensitivity area.

We have already spoken about the euglena’s eyespot, but hydras and

worms also have light/dark sensitivity cells located in their outer epidermal

layers. Eventually, a pinhole eyespot formed, which enabled even more

complex processing of visual information. Of course, in order for this

eyespot to have occurred, other parts of a rudimentary nervous system had

to be in place, like the mass of neurons localized in one area, such as the

head.

For example, the nautilus—one of the so-called living fossils, because it

morphologically has not changed much since the Cambrian explosion—is

representative of this next step in the evolution of the eye. The evolution

of the visual system likely began with a general light sensitivity around the ventricle of the forebrain. These forebrain neurons were transformed

in the course of evolution into an assemblage of neurons capable of analyzing

strictly visual information, rather than other information carried to

this area by other input neurons. The processed visual information was

transferred through relay neurons (the future ganglion cells) to the contralateral

part of the brain. In subsequent generations, this primordial

retina evaginated (formed a cup-like hole) and eventually transformed into

the lateral eyes of vertebrates (Cronly-Dillon, 1991).

Once this rudimentary eye developed, it took different directions in

terms of its formation among the various species that have vision. There

are all types of eyes falling into two broad categories: the compound eyes

of insects, crabs, and arthropods that are composed of hundreds of cylindrical

elements consisting of a lens, eight photoreceptor cells, and a sleeve

of opaque pigment cells, and the vertebrate eyes of fi sh, reptiles, and

mammals that all build upon the basic fl atworm eye having a retina, ganglion

cells, and projections to a brain. This divergence further ratifi es the

evolutionary principles of genetic variability and the natural selection of

the fi ttest traits given a specifi c environment. The kind of eye that fortuitously

developed, and then fl ourished, in the sand shark population under

the sea will be different from the eye that fortuitously developed, and then

fl ourished, in the black widow spider population on land, and so on.

As noted already, through investigation of the fossil endocasts of insectivores,

Jerison (1973, 1991) has offered a convincing case that the ancestor

of the primate was a small, nocturnal creature that prowled through trees

and bushes on the lookout for insects, as some small primates do today.

The eyes of this creature were close set and forward facing, indicating that

it was a predator. Predators like owls, cats, and primates have close set,

forward facing eyes with stereotopic vision, enabling them to more easily

identify depths and distances while hunting their prey. Animals like rabbits

and squirrels—the prey of predators like owls, cats, and some primates—

have eyes on the sides of their heads with panoramic vision, enabling them

to more easily identify predators from a variety of angles. The eyes of this

insectivore would have had an abundance of rods and fewer cones to

enable maneuverability in the dark. Insectivores had to forage on the

ground and in the dark in order to steer clear of bigger reptiles, amphibians,

and dinosaurs that would have preyed on them during that time.

However, when most of the dinosaurs and other larger predators were

wiped out by an asteroid at the close of the Cretaceous, this allowed for

the possibility of insectivores to occupy different niches. Eventually, some

of the descendents of these creatures did their foraging during the day,

The Evolution of the Visual System and Scenario Visualization 107

while others took to life in the trees. Still others, like our primate ancestors,

did both. By 20 mya, Proconsul had taken to life in the trees of Africa, most

likely because such a life off of the ground made it easier to avoid predators

lurking on the ground, as well as pounce upon prey that happened to be

on the ground. This life in the trees is evidenced by their money-like and

ape-like skeletons as well as their diet, which indicate a lifestyle not unlike

that of present-day monkeys and apes (Groves, 2002; Hartwig, 2002;

McGrew, 2004; Byrne & Whiten, 1988; Allman, 1982, 2000).

Life during the daytime and in the trees meant that a visual system, at

the very least, would need to have evolved more cones, the capacity to

distinguish colors, more memory-storage capacities, and facial recognition.

More cones would be needed to be able to handle the brighter light of

daytime living. A capacity to distinguish colors would be needed to determine

what foods one could eat, as well as distinguish foreground from

background while maneuvering through the trees. Rudimentary forms of

long-term and short-term memory would be needed for a mental map of

the area where this creature lived, as well as for distinguishing friend from

foe. Finally, facial recognition would be needed to distinguish friend from

foe, as well as to communicate various messages like “Leave me alone,”

“Back off,” “It’s o.k. to touch me,” and the like to other members of one’s

social group or to enemies. On top of all of this, arboreal life required

extremities more fi t for grasping, jumping, launching, and other sorts of

bodily maneuvering (Barton, 1998; Fleagle, 1999; Byrne, 1995, 2001; Rilling

& Insel, 1998; Allman, 1977).

In order to do all of this, the connections between eye and brain, as well

as those connections between other sensorimotor systems and brain, would

need to be more complex. As Jerison (1973, 1997) and Kaas (1996) point

out, the visual system of this early arboreal ancestor would have had a

brain very similar to that of a lemur or a monkey. This being the case, at

least three features of the brain and visual system would have needed to

be in place in Proconsul.

First, referring back to MacLean’s model, the reptilian, the paleomammalian,

and part of the neomammalian brain would have to be present in

Proconsul, since the amount of processing that such arboreal ancestors had

to do required a brain larger than that of reptiles. The amount of processing

that the visual system does alone would constitute a bigger brain.

However, arboreal social life entailed the interaction of sensory modalities

in combination with memory, emotion, planning, and motor function. All

of this processing required a bigger brain, more synaptic connections, and

more types of neural connections. In addition, the system must have been

a hierarchical organization of processes and subsystems so that the right

data could be selected and information could be integrated as Proconsul

went about the business of negotiating its environment.

Second, in order to organize and coordinate information from the specialized

visual modules and each of the other sensory modalities, integrative

or association areas and systems had to evolve. Research by Kaas (1987,

1993, 1995, 1996), Bear et al. (2001), Uylings & van Eden (1990), Roberts,

Robbins, & Weiskrantz (1998), and Hofman (2001) has shown that it is

primarily the association areas of the brain that increase in the evolution

of the mammal. The association areas show a marked increase when we

compare the brains of rats, cats, macaque monkeys, and humans, in that

order. On inspection, the visual, auditory, and sensorimotor areas of a rat

basically butt up against one another with very little of the cortex—some

15%–20%—devoted to the association of these areas. Some 40%–50% of

the frontal, parietal, and temporal lobes of the macaque monkey are

devoted to association of sensorimotor information. Even more cortex—

some 50%–60%—is devoted to association of sensorimotor information in

humans.

Third, the visual areas would have needed to be linked up with the motor

areas, so that visual information could be communicated to the brain and

then translated into voluntary motion. The motor areas of primates receive

inputs from the thalamic nuclei that relay information from the basal telencephalon

and the cerebellum, and they send outputs to motor control

neurons in the brain and spinal cord. The motor area is situated next to

the somatosensory area, and both are located in the frontal lobe near the

central sulcus at the parietofrontal junction of the brain. The sensorimotor

area—along with the auditory and visual areas—send projections to association

areas in the parietal and temporal cortices. From there, the projections

are sent to the multimodal areas located in the parietotemporal,

prefrontal, and limbic cortices (Kandel et al., 2000). These connections

would have had to be in place in the brain of Proconsul so that it could

have acted on visual information it received.

In the last chapter, I mentioned the idea put forward by the Gestalt

theorists and William James that the mind and the world are “something

of a mutual fi t.” This idea should be kept in mind here as we investigate

the evolution of visual modules, along with the evolution of other sensory

modules and mechanisms of selectivity and integration. Events in the

physical environment are composed of materials that the human brain,

complete with its specialized modules, has evolved to perceive and discriminate.

These materials take forms ranging from specifi c chemicals, to

The Evolution of the Visual System and Scenario Visualization 109

mechanical energy, to electromagnetic radiation (light) and are discriminated

by the different sensory modalities that are specifi cally attuned to

these stimuli: taste and smell are specialized for processing concentrations

of different chemicals, touch is specialized for processing mechanical

deformations of the skin, hearing is specialized for the intake of sound

waves, and vision is specialized for the intake of electromagnetic radiation.

Thus, given the environment surrounding an animal, it makes sense that

the CNS is a hierarchically organized system composed of specialized

modules, as well as mechanisms of selectivity and integration, in order to

process the myriad pieces of environmental information.