Vision and Blindsight

This paper reflects the research and thoughts of a student at the time the paper was written for a course at Bryn Mawr College. Like other materials on Serendip, it is not intended to be "authoritative" but rather to help others further develop their own explorations. Web links were active as of the time the paper was posted but are not updated.

Contribute Thoughts | Search Serendip for Other Papers | Serendip Home Page

Biology 202
1999 Second Web Reports
On Serendip

Vision and Blindsight
Implications Regarding Consciousness

David Mintzer

Vision-- receiving and interpreting light signals from the environment in order to form an image in one's mind-- is an incredibly complex process. Somehow signals from photoreceptors located in the eye are converted into the conscious experience of sight. Of all the aspects of vision, perhaps the most difficult for us to comprehend scientifically is this notion of consciousness. Somehow the brain interprets light waves hitting the retina so that we are visually aware of our surroundings. While the mechanism of signal transduction from the photoreceptor through the visual cortex has been extensively elucidated, science has difficulty dealing with the phenomena of consciousness and awareness, especially on a reductionist level.

A recent neurobiological approach to understanding consciousness, at least on a perceptual level, has involved the study of the phenomenon of blindsight. Damage to areas of the visual cortex often result in complete or partial blindness. Although the eye itself is undamaged, patients report an inability to detect any light input in part of (or the entire) visual field. However, experiments regularly show that somehow, visual cues are processed. Visual inputs presented to the blind field affect the patient's response to stimulus in the normal visual field. Reaction times to stimuli are affected as well as the interpretation of the stimuli. A visual cues presented in the blind field may suggest a certain interpretation of an ambiguous stimuli. For example, the interpretation of the word "bank", presented as an auditory cue, differs depending on whether the word "river" or "money" is presented to the blind field, even though the patient does not report seeing these cues . Furthermore, patients are often able to respond directly to visual cues in their blind fields. Patients tend to correctly identify shape, color, and motion of inputs. In a forced choice experiment, where the subject is asked to identify certain features of a visual cue, the subject will perform much better than chance even though they feel as if they are randomly guessing (2).

While head trauma or tumors often induce the "psychic" blindness of these patients, a model has been developed in monkeys by removing all or part of the primary visual cortex. These monkeys are able to respond to visual inputs. They can be trained to touch illuminated bulbs rather than unlit ones and identify certain colors and patterns in order to obtain food. This phenomenon is believed to parallel human blindsight because when trained to respond differently according to whether there is a visual cue or not, these monkeys respond as if there were no cue when a visual input is presented to the blind field (1). It is therefore believed that these animals are able to respond to and identify features of a visual cue even though they do not report seeing it.

The phenomenon of blindsight has far reaching implications regarding consciousness, awareness and the "I" function. These studies demonstrate that receiving and interpreting visual inputs is independent of our awareness of that input. Does this imply that there is a separate mechanism of "consciousness" which can be disengaged from our senses? Marcel proposes that this loss of visual consciousness results from the patients' lack of phenomenal representation. Information about contrast, intensity and wavelength are not represented in phenomenal form as borders, brightness and color, respectively. However, there is some evidence that phenomenal vision not caused by external input (mental imagery, dreams, hallucinations, etc) may still occur with patients exhibiting blindsight. Also these patients may lack visual awareness, but sometimes are still aware of a visual cue. They may have a feeling or sensation of its presence without actually "seeing" it (3).

In order to attempt to understand the neurobiology of blindsight and its implications for human consciousness, it is important to understand the brain's role in vision. The visual cortex is divided into 20 distinct areas. The primary visual cortex (also known as V1 or striate cortex) is the area which, when damaged, results in blindsight. The optic nerves transfer information from the photoreceptors of the retina to the dorsal lateral geniculate nucleus. Optic radiations project from the dLGN to the primary visual cortex. The primary visual cortex is made up of six layers, and if spread out flat, would map out, contralaterally, the visual field. The V1 is known to process visual information regarding orientation, spatial frequency, texture, color and retinal disparity (depth perception) Information then proceeds to the visual association cortex where it is integrated into a visual image (4).

So how is visual information processed in the brain on an unconscious level when the primary visual cortex is damaged or destroyed? In fact, there are many divergent pathways which project from the retina, so it is not overly surprising that patients maintain some aspect of vision when one area is destroyed. In monkeys it has been found that when the V1 area is destroyed, ganglion cells send visual information to the retinorecipient nuclei of the midbrain and diencephalon. Projections to the dorsal lateral geniculate nucleus are almost completely destroyed (1). However it is believed that some connections remain since they would be needed to process color information (an ability some blindsight patients display). One of the best supported theories is that projections to the superior colliculus and pulvinar might provide indirect visual input to the extra-striate cortex (5). Recent MRI experiments, comparing aware with unaware vision in blindsight patients (vision in the unaffected field compared to blind field) reveal a direct correlation with activity in certain cortical areas (6). The V5 area is believed to be involved in visual processing motion and speed and could be important regarding blindsight of moving cues.

While these neuronal mechanisms may help explain how patients with damage to areas of the visual cortex are able to process a limited amount of visual input, they do not provide a clear explanation regarding the phenomenon of consciousness (or lack thereof.) With our limited understanding of the brain's functioning, it may help to investigate this aspect from a theoretical viewpoint. The classic sequence input-->pattern generator-->motor symphony helps us understand how certain behavior may arise. While this model certainly helps us interpret external output, we may also look at it in regard to internal output such as thoughts, vision and consciousness. If these phenomena are caused (or at least enabled) by neuronal activity, the sequence input-->pattern generator-->internal output should be valid.

Such a model has actually been developed regarding visual consciousness. Crick and Koch discovered that neurons in the visual cortex exhibit 40 Hz synchronous oscillations when presented with visual input. They theorized that this phenomena underlies visual awareness by integrating all visual information to form one conscious image. It is known that different visual attributes such as shape, color, texture, etc are processed by different areas of the visual cortex. Crick and Koch hypothesize that the 40 Hz oscillations integrate the different neurons relevant to a certain visual input by synchronizing their firing rate. They believe that it is attention which enables consciousness. When someone is attending a certain visual cue, neurons processing all the different information regarding that input synchronize their oscillations and create a coherent image. It may work something like this: "Objects in the visual field generate responses in the visual cortex. Attention selects one of these objects, perhaps by means of a principle of feature saliency. When a salient location is selected, the information encoded about that location in the cortex is activated and sent to the thalamus, a subcortical structure that gates information coming into the cortex. The thalamus can then feed back to the cortex, choose the appropriate neurons to entrain in oscillation. (7)" Hence the neurons processing the incredible amounts of information regarding that object are coordinated and an awareness of the object is formed. It may be possible that when areas of the primary visual cortex are damaged, this process cannot occur. Information reaches the brain, but cannot be organized into a coherent image and brought into consciousness. This theory has not been extensively investigated, and due to the nature of the phenomenon, it may be impossible to confirm. However, it does give us insight into a possible theoretical model which can be grounded in empirical data.

The phenomenon of blindsight raises as many questions as it answers. It complicates and confuses the notion of an "I" function. It is evident that our brain processes visual information which we are not aware of and that this can affect our behavior. While for the moment blindsight remains a somewhat mysterious concept, hopefully further study will help reveal the mechanisms and implications of our own consciousness and awareness.