ABSTRACT
In recent years the concept of the neural organization of visual perception has undergone rapid changes. Based on neurophysiological studies in primates, it has been shown that visual information is mediated not only by the geniculostriate pathway (the “first” visual system) but also by the midbrain (superior colliculus) and the thalamus (nc. pulvinar, the so-called second visual system). Considerable spared visual capacity has been found in monkeys blinded after striate cortex lesions; such animals are able to detect and to localize visual stimuli. The monkeys were not able, however, to identify these stimuli.
By using methods that depend upon forced-choice procedures (analogous to those used with monkeys), rather than on verbal responses, it could be shown that brain-damaged patients have similar capacities.
These patients had lost large parts of their binocular visual field after damage (mostly of cerebrovascular origin) to the geniculostriate pathway. The lesions had been verified by computerized tomography.
The patients were able to localize light targets presented briefly in their perimetrically blind hemifield by finger pointing or by shifting their gaze towards them. When forced to localize these targets repeatedly in different sessions, they showed a marked improvement in their localization accuracy.
In further experiments, the phasic electrodermal response was used as an indicator for the registration of a light target presented in the hemianopic field region. The experiments showed a significant correlation between the occurrence of the phasic electrodermal response and the presentation of a light stimulus in the perimetrically blind field. This response—a component of the orienting reaction—does, therefore, not depend on the intact geniculostriate visual system.
696Hemianopic patients were also able to indicate the presence of a light target in their “blind” hemifield by voluntary motor responses such as lid closure and bar pressing after they underwent specific practice. In contrast, they were unable to indicate the presence of a light target by a verbal response. This observation suggests that the mechanisms for language are not associated with the neural substrate underlying the detection and localization of light stimuli in the absence of the geniculostriate pathway.
It is interesting to note that patients could localize light targets within their hemianopic field quite accurately after they were trained to indicate the presence of the light stimulus. It is concluded that both detection and localization of visual stimuli are subserved by the same or related neural mechanisms. These capacities in an otherwise blind field region can only be demonstrated, however, when the patient is forced to use them, even though he never can “see” any light stimulus, and does not have any conscious experience of it.
These observations suggest that patients possess some visual capacities in their perimetrically blind field region. These capacities are similar to those observed in destriated monkeys. Both humans and animals need specific practice in order to make use of these capacities. It seems plausible that they are mediated by the “second” visual system, since the geniculostriate pathway was completely damaged in the patients tested.
This suggestion could be strengthened if one were able to demonstrate the reverse effect after lesions had damaged these phylogenetically older retino-tectal pathways. One would expect that such damage would not lead to visual field loss, but would disturb the detection of visual stimuli and would also affect orienting reactions towards them. In monkeys, such effects were observed after tectal or pulvinar lesions. In patients, lesions affecting only the superior colliculus or the nc. pulvinar are relatively scarce. Single case studies of patients suffering a tectal or pulvinar lesion are, therefore, of high value. Reports on a few cases support the suggestion mentioned above. These patients have difficulties detecting light stimuli in the periphery of the visual hemifield contralateral to the lesion when presented simultaneously in either half-field. They also cannot easily shift attention to the side contralateral to the lesion, or shift their gaze towards targets presented there.
Summarizing the observations obtained in both monkeys and humans after damage to the “first” or to the “second” visual system, one might conclude that the detection and localization of visual stimuli is subserved by extrageniculostriate neural mechanisms, whereas the identification of these stimuli depends crucially on the geniculostriate pathway.
For visual perception both the “first” and the “second” visual systems are needed. The detection of visual events and the shift of gaze towards them could be termed prerequisites for the analysis of these events.
