3.4 Levels of Visual Processing

In the previous section, I noted that visual recognition and discrimination

of objects are key to an animal’s survival, and I went on to suggest that

these activities are possible because of the phenomenon of visual modularity

and the mechanism of visual integration. There is more to the story

concerning an animal’s visual cognition of an object than simply visual

modularity and visual integration. Numerous studies indicate that (1)

iconic memory, namely, a form of short-term memory that lasts no more

than a second or so, as well as (2) attention and (3) the synchronous fi ring

of the neurons in the areas relevant to the visual percept, seem to be necessary

for visual cognition. These parts and processes may not comprise the

suffi cient condition for an overall coherently unifi ed visual percept; this is

to say, there may be some element or elements missing. However, the

empirical evidence seems to indicate that such parts and processes are

necessary for a coherent visual scene.

Before discussing these additional conditions necessary for visual cognition,

I want to distinguish the various levels of visual processing, of

which visual cognition is a part. We can distinguish four levels of visual

processing in the visual system. The fi rst level is a noncognitive visual

processing that occurs at the lowest level of the visual hierarchy associated

with the eye, LGN, and primary visual cortex. At this level, the

animal is wholly unaware of the processing, as the brain receives the

disparate pieces of basic information in the visual fi eld concerned with

lines, shapes, distance, depth, color, and so forth, of an object in the

visual fi eld (Gray, 1999; Julesz, 1984; Merikle & Daneman, 1998; Rees,

Kreiman, & Koch, 2002).

Evidence for this level of processing comes from data gathered from

backward masking experiments as well as from patients who suffer from

visual form agnosia, prosopagnosia, visual neglect, blindsight, and color anomia

(Farah, 1984, 1990, 1997; Cowey & Stoerig, 1995; Crick & Koch, 1998;

Marcel, 1983). The patients in these experiments are able to process

visual information, seemingly without being aware or cognizant of that

information. In his blindsight experiments, Weiskrantz (1986, 1988, 1997)

has shown that limited stimulus detection and movement toward objects

is possible in patients who claim they “cannot see,” that is, have no

awareness of the stimulus. Graves & Jones (1992) also draw a distinction

between nonconscious visual processing and visual awareness as a result

of their experiments with blindsight patients. Humphrey (1992) notes

a case of color anomia where a woman was able to process colors without being aware that she was making mistakes in her conscious

reporting of those colors (also see Oxbury, Oxbury, & Humphrey,

1969).

In his backward masking experiments, Eagle (1959) documented interesting

cases where people were shown a quick image of a man wielding a

knife—approximately one-tenth of a second—followed by a longer lasting

image of the same man standing and smiling. They then were asked to

describe the character of the man in the second longer lasting picture.

Interestingly enough, many subjects were unaware that the fi rst picture

occurred and yet still would judge the character of the man in the second

picture according to what was processed by their visual system in the fi rst

picture. These studies seem to indicate that there is noncognitive visual

processing occurring, despite the absence of awareness and/or conscious

visual experience.

The second level of visual processing is a cognitive visual processing that

occurs at a higher level of visual awareness associated with the what and

the where visual unimodal areas. When it is said that an animal visually

perceives what an object looks like or where an object is located, this means

that the animal is cognitively aware of or cognitively attends to that object in

the visual fi eld. We can infer that cognitive visual processing is taking place

from the fact that if the what system is nonfunctioning, a primate still may

be able to distinguish where an object is; conversely, if the where system is

nonfunctioning, a primate still may be able to distinguish what an object

is (Goodale et al., 1994; Goodale & Murphy, 2000; Ungerleider & Haxby,

1994).

The move from noncognitive visual processing to cognitive visual processing

is a move from the purely neurobiological to the psychological

dimension associated with the brain’s activities. I am using words like cognition,

awareness, and perception to refer to similar psychological discriminatory

abilities of an animal. Cognition enables an animal with a complex

nervous system to negotiate environments as optimally as possible. In the

previous chapter, I argued that the components of an organism are emergent

entities nonreducible to the physicochemical parts of which they are

composed, based upon the way in which the components are organized

to do something directly related to the generalized homeostasis of this

hierarchically organized living system (also see Arp, 2008a). The psychological

dimension associated with the brain’s activities can be considered

as another level of emergent phenomena—psychologically emergent

phenomena—added to the neurobiological hierarchy. Cognition is an

emergent phenomenon because the parts and processes associated with it appear to be organized in such a way so as to aid an animal in discriminating

information in environments. However, the kind of end result or end

product of cognition—although similar to other activities in the animal’s

hierarchy in having generalized homeostasis as the goal—is different in

that such a product is a psychological phenomenon that aids in generalized

homeostasis.

There is a huge amount of literature devoted to questions about the

existence of psychological phenomena and whether psychological phenomena

supervene upon or emerge from neurobiological phenomena (for

starters, see Heil, 2004a, 2004b; Stich & Warfi eld, 2003; Chalmers, 1996;

McGinn, 1982; Hasker, 1999; Hatfi eld, 1999; Mesalum, 1998; Kim, 2000;

Lycan, 1995; Searle, 1992; Arp, 2005b, 2007b, 2008d). Working out the

problems associated with these issues constitutes solving several so-called

mind–body problems. Now, no one has been able to give a satisfactory

account of how it is that psychological states—particularly conscious psychological

states—arise from, as well as interact with, the gray matter of

the brain. However, my intuition concerning the emergence of psychological

states is that just as the components at various levels of neurobiological

and biological hierarchies—such as organelles, cells, tissues, and organs—

cannot be reduced to the physicochemical parts of which they are composed,

so too visual cognition, although dependent upon neurobiological

processes, is not reducible to such processes. Again, the main reason why

psychological phenomena are nonreducible to neurobiological phenomena

is the same reason why neurobiological and biological components

are nonreducible to the physicochemical parts of which they are composed,

namely, such components and phenomena emerge as a result of

the way in which they are organized to do something directly related to

generalized homeostasis of the organism. I will have more to say about

this in the next two chapters when I speak more directly about the evolution

of scenario visualization in our species’ psychology.

The third level of visual processing is a cognitive visual processing that

occurs at an even higher level of visual awareness concerned with the

integration of the disparate pieces of visual unimodal information in the

visual unimodal association area. There are times when an animal must

determine both what an object is and where it is located, and this level of

visual processing makes such a determination possible.

The fourth level of visual processing is a conscious cognitive visual processing

that occurs at the highest level of the visual hierarchy associated with

the multimodal areas, frontal areas, and, most probably, the summated

areas of the cerebral cortex. This is the kind of visual processing related to

human visual awareness and visual experience, and evidence for this level

comes from reports made by individuals, as well as from observing human

behavior (Roth, 2000; James, 1890; Chalmers, 1996). As I will make clear

in the next two chapters, one form of conscious cognitive visual processing

is scenario visualization, namely, the selection and integration of visual

images from mental modules, as well as the projection of visual images

into future scenarios for the purposes of negotiating environments.

In this project, I am concerned mostly with the progression from cognitive

visual processing to conscious cognitive visual processing, the relationship

of these processes to one another, and, ultimately, how conscious

cognitive visual processing evolved from cognitive visual processing. This

is so because, as I show, conscious cognitive visual processing—in terms of

what I call scenario visualization—is necessary for vision-related, nonroutine

creative problem solving. Although I will not be able to solve completely

the mind–body problem of how it is that conscious experience can

emerge from and interact with the gray matter of the brain, my hypothesis

concerning scenario visualization is an attempt to explain a part of our

conscious abilities and the reason for its emergence in our species.

We can think of the four levels of processing in relation to the various

species in the animal kingdom. All vertebrate species in the phylum

Chordata with a rudimentary visual system—mammals, birds, reptiles,

amphibians, and fi sh—exhibit noncognitive visual processing of some

kind. These same vertebrate species also exhibit cognitive visual processing

to some degree or another. Besides the numerous studies on apes, monkeys,

dolphins, cats, dogs, birds, and turtles that seem to indicate that they have

cognitive abilities such as awareness and associative learning (for starters,

see Marten & Psarakos, 1994; Byrne, 1995; Pearce, 1997; Stamp Dawkins,

1993; Parker, 1996; Tomasello et al., 1993; Weir, Chappell, & Kacelnik,

2002; Eimer & van Velzen, 2002), I have witnessed my own cat’s abilities

to perceive and recall where her treat is located in the cabinet, as well as

her abilities to expect the treat and go for the treat when I have left the

room. (Sometimes, she actually would open the cabinet where the can of

treats was kept and knock the can to the counter below!) Given the data,

I agree with Roth (2000, p. 95) that “it is fair to assume that all vertebrates

with larger cortexlike structures, particularly those with cortices showing

cross-modality information transfer, have awareness about what is going

on around them.” However, within the order primates, human beings

alone seem to exhibit conscious cognitive visual processing to a full degree,

while the other primates may do so to a lesser degree (cf. Parker, 1996;

Whiten et al., 1999; Mitchell, 1993).

In the previous section, I noted that visual recognition and discrimination

of objects are key to an animal’s survival, and I went on to suggest that

these activities are possible because of the phenomenon of visual modularity

and the mechanism of visual integration. There is more to the story

concerning an animal’s visual cognition of an object than simply visual

modularity and visual integration. Numerous studies indicate that (1)

iconic memory, namely, a form of short-term memory that lasts no more

than a second or so, as well as (2) attention and (3) the synchronous fi ring

of the neurons in the areas relevant to the visual percept, seem to be necessary

for visual cognition. These parts and processes may not comprise the

suffi cient condition for an overall coherently unifi ed visual percept; this is

to say, there may be some element or elements missing. However, the

empirical evidence seems to indicate that such parts and processes are

necessary for a coherent visual scene.

Before discussing these additional conditions necessary for visual cognition,

I want to distinguish the various levels of visual processing, of

which visual cognition is a part. We can distinguish four levels of visual

processing in the visual system. The fi rst level is a noncognitive visual

processing that occurs at the lowest level of the visual hierarchy associated

with the eye, LGN, and primary visual cortex. At this level, the

animal is wholly unaware of the processing, as the brain receives the

disparate pieces of basic information in the visual fi eld concerned with

lines, shapes, distance, depth, color, and so forth, of an object in the

visual fi eld (Gray, 1999; Julesz, 1984; Merikle & Daneman, 1998; Rees,

Kreiman, & Koch, 2002).

Evidence for this level of processing comes from data gathered from

backward masking experiments as well as from patients who suffer from

visual form agnosia, prosopagnosia, visual neglect, blindsight, and color anomia

(Farah, 1984, 1990, 1997; Cowey & Stoerig, 1995; Crick & Koch, 1998;

Marcel, 1983). The patients in these experiments are able to process

visual information, seemingly without being aware or cognizant of that

information. In his blindsight experiments, Weiskrantz (1986, 1988, 1997)

has shown that limited stimulus detection and movement toward objects

is possible in patients who claim they “cannot see,” that is, have no

awareness of the stimulus. Graves & Jones (1992) also draw a distinction

between nonconscious visual processing and visual awareness as a result

of their experiments with blindsight patients. Humphrey (1992) notes

a case of color anomia where a woman was able to process colors without being aware that she was making mistakes in her conscious

reporting of those colors (also see Oxbury, Oxbury, & Humphrey,

1969).

In his backward masking experiments, Eagle (1959) documented interesting

cases where people were shown a quick image of a man wielding a

knife—approximately one-tenth of a second—followed by a longer lasting

image of the same man standing and smiling. They then were asked to

describe the character of the man in the second longer lasting picture.

Interestingly enough, many subjects were unaware that the fi rst picture

occurred and yet still would judge the character of the man in the second

picture according to what was processed by their visual system in the fi rst

picture. These studies seem to indicate that there is noncognitive visual

processing occurring, despite the absence of awareness and/or conscious

visual experience.

The second level of visual processing is a cognitive visual processing that

occurs at a higher level of visual awareness associated with the what and

the where visual unimodal areas. When it is said that an animal visually

perceives what an object looks like or where an object is located, this means

that the animal is cognitively aware of or cognitively attends to that object in

the visual fi eld. We can infer that cognitive visual processing is taking place

from the fact that if the what system is nonfunctioning, a primate still may

be able to distinguish where an object is; conversely, if the where system is

nonfunctioning, a primate still may be able to distinguish what an object

is (Goodale et al., 1994; Goodale & Murphy, 2000; Ungerleider & Haxby,

1994).

The move from noncognitive visual processing to cognitive visual processing

is a move from the purely neurobiological to the psychological

dimension associated with the brain’s activities. I am using words like cognition,

awareness, and perception to refer to similar psychological discriminatory

abilities of an animal. Cognition enables an animal with a complex

nervous system to negotiate environments as optimally as possible. In the

previous chapter, I argued that the components of an organism are emergent

entities nonreducible to the physicochemical parts of which they are

composed, based upon the way in which the components are organized

to do something directly related to the generalized homeostasis of this

hierarchically organized living system (also see Arp, 2008a). The psychological

dimension associated with the brain’s activities can be considered

as another level of emergent phenomena—psychologically emergent

phenomena—added to the neurobiological hierarchy. Cognition is an

emergent phenomenon because the parts and processes associated with it appear to be organized in such a way so as to aid an animal in discriminating

information in environments. However, the kind of end result or end

product of cognition—although similar to other activities in the animal’s

hierarchy in having generalized homeostasis as the goal—is different in

that such a product is a psychological phenomenon that aids in generalized

homeostasis.

There is a huge amount of literature devoted to questions about the

existence of psychological phenomena and whether psychological phenomena

supervene upon or emerge from neurobiological phenomena (for

starters, see Heil, 2004a, 2004b; Stich & Warfi eld, 2003; Chalmers, 1996;

McGinn, 1982; Hasker, 1999; Hatfi eld, 1999; Mesalum, 1998; Kim, 2000;

Lycan, 1995; Searle, 1992; Arp, 2005b, 2007b, 2008d). Working out the

problems associated with these issues constitutes solving several so-called

mind–body problems. Now, no one has been able to give a satisfactory

account of how it is that psychological states—particularly conscious psychological

states—arise from, as well as interact with, the gray matter of

the brain. However, my intuition concerning the emergence of psychological

states is that just as the components at various levels of neurobiological

and biological hierarchies—such as organelles, cells, tissues, and organs—

cannot be reduced to the physicochemical parts of which they are composed,

so too visual cognition, although dependent upon neurobiological

processes, is not reducible to such processes. Again, the main reason why

psychological phenomena are nonreducible to neurobiological phenomena

is the same reason why neurobiological and biological components

are nonreducible to the physicochemical parts of which they are composed,

namely, such components and phenomena emerge as a result of

the way in which they are organized to do something directly related to

generalized homeostasis of the organism. I will have more to say about

this in the next two chapters when I speak more directly about the evolution

of scenario visualization in our species’ psychology.

The third level of visual processing is a cognitive visual processing that

occurs at an even higher level of visual awareness concerned with the

integration of the disparate pieces of visual unimodal information in the

visual unimodal association area. There are times when an animal must

determine both what an object is and where it is located, and this level of

visual processing makes such a determination possible.

The fourth level of visual processing is a conscious cognitive visual processing

that occurs at the highest level of the visual hierarchy associated with

the multimodal areas, frontal areas, and, most probably, the summated

areas of the cerebral cortex. This is the kind of visual processing related to

human visual awareness and visual experience, and evidence for this level

comes from reports made by individuals, as well as from observing human

behavior (Roth, 2000; James, 1890; Chalmers, 1996). As I will make clear

in the next two chapters, one form of conscious cognitive visual processing

is scenario visualization, namely, the selection and integration of visual

images from mental modules, as well as the projection of visual images

into future scenarios for the purposes of negotiating environments.

In this project, I am concerned mostly with the progression from cognitive

visual processing to conscious cognitive visual processing, the relationship

of these processes to one another, and, ultimately, how conscious

cognitive visual processing evolved from cognitive visual processing. This

is so because, as I show, conscious cognitive visual processing—in terms of

what I call scenario visualization—is necessary for vision-related, nonroutine

creative problem solving. Although I will not be able to solve completely

the mind–body problem of how it is that conscious experience can

emerge from and interact with the gray matter of the brain, my hypothesis

concerning scenario visualization is an attempt to explain a part of our

conscious abilities and the reason for its emergence in our species.

We can think of the four levels of processing in relation to the various

species in the animal kingdom. All vertebrate species in the phylum

Chordata with a rudimentary visual system—mammals, birds, reptiles,

amphibians, and fi sh—exhibit noncognitive visual processing of some

kind. These same vertebrate species also exhibit cognitive visual processing

to some degree or another. Besides the numerous studies on apes, monkeys,

dolphins, cats, dogs, birds, and turtles that seem to indicate that they have

cognitive abilities such as awareness and associative learning (for starters,

see Marten & Psarakos, 1994; Byrne, 1995; Pearce, 1997; Stamp Dawkins,

1993; Parker, 1996; Tomasello et al., 1993; Weir, Chappell, & Kacelnik,

2002; Eimer & van Velzen, 2002), I have witnessed my own cat’s abilities

to perceive and recall where her treat is located in the cabinet, as well as

her abilities to expect the treat and go for the treat when I have left the

room. (Sometimes, she actually would open the cabinet where the can of

treats was kept and knock the can to the counter below!) Given the data,

I agree with Roth (2000, p. 95) that “it is fair to assume that all vertebrates

with larger cortexlike structures, particularly those with cortices showing

cross-modality information transfer, have awareness about what is going

on around them.” However, within the order primates, human beings

alone seem to exhibit conscious cognitive visual processing to a full degree,

while the other primates may do so to a lesser degree (cf. Parker, 1996;

Whiten et al., 1999; Mitchell, 1993).