VISUAL MEMORY, BRIGHTNESS AND FLUX IN

The study of simple visual discriminations reveals fundamental properties of learning and memory in the nervous system. Physical measures of the light source include energy emitted (flux) or reflected (reflectance) from the stimulus per unit area. Heinrich Klüver (1942) called it brightness if the total amount of light was measured over the whole stimulus source (including contours and edges), and density of luminous flux if measured over a unit area of the stimulus. He concluded that visually decorticated monkeys could solve a luminous-flux problem but not a brightness problem.

Using this definition, a brightness discrimination includes both light intensity and light contrast on edges that contribute to pattern perception. A flux discrimination, on the other hand, pertains to differences in light intensity per unit area. This intensity can be for the whole stimulus (total flux) or for parts of the stimulus (local flux). For example, a horizontal-versus-vertical-stripes pattern problem can be equated for total flux and for total brightness (same number of contrasting contours) and yet the edges of the stimulus cards have differences in local flux. Discrimination tasks for intensity that are used for testing animals, whether black and white cards with edges or black and white alleys with an edge that creates a contrast (the wall separating the alleys), normally are not purely flux discriminations but are brightness discriminations that include elements of pattern discrimination. Discriminations more purely restricted to flux can be achieved by use of contact lenses that diffuse light and obscure edges, by successive discrimination problems in which both alleys being lit means to go one way or both alleys being dark means to go the other way, or by a shuttle box in which the subject sits in the middle of the apparatus, one alley is lit, and the alley 180 degrees away is not. The distinguishing feature between a flux and a brightness discrimination is that the latter has contours created by contrasting levels of flux within the same stimulus.

In the simplest visual discrimination a subject distinguishes between a black and a white stimulus card. With this task Karl Lashley (1935) studied the role of cerebral neocortex in his search for the physical manifestation of memory, for the engram. Rats without any neocortex cannot discriminate visual patterns (Lavond and Dewberry, 1980). However, rudimentary visual functions remain in that decorticated rats can see moving objects, can detect the deep side of a visual cliff, and can learn or relearn the brightness discrimination.

Importantly, the results of brightness discrimination training illustrate the property of behavioral recovery of function following brain injury. Lashley trained rats on a brightness discrimination and then systematically removed fractions of the cerebral neo-cortex. After removal of visual neocortex there was no evidence for retention of the brightness discrimination. With continued training, however, the rats reacquired the discrimination, taking as many trials as initially required to learn. One possible conclusion would be that the lesion destroyed the memory and, once it was removed, a new memory could be reestablished with the same effort. However, this is a somewhat fortuitous result in that it is an outcome of the training criterion used, because posteriorly decorticated subjects perform better when trained with less stringent criteria and worse when trained to successively more stringent criteria (Spear and Braun, 1969). This indicates that normal learning and learning by posteriorly decorticated subjects are by different mechanisms.

The recovery of brightness discriminations is consistent with Lashley's previous observations on maze learning (1929) where he found equipotentiality (all parts of the neocortex have the same capacity for supporting memory) and mass action (large areas of neo-cortex contribute to the memory). There is one caveat, however, in that he confined these properties to visual cortex for visual discriminations. Lashley did not think that the anterior neocortex participated in visual discriminations as it does for maze learning. However, more recent work supports the generalization of visual function to the entire neocortex (Cloud, Meyer, and Meyer, 1982).

Bauer and Cooper (1964) suggested that memory was not stored in the neocortex at all, but that visual cortex was necessary for seeing the stimulus as a pattern discrimination (a brightness discrimination). They suggested that learning after a visual cortical lesion was by using a different stimulus feature (i.e., a flux discrimination). However, there is evidence that both brightness and flux are learned simultaneously. Meyer and Meyer showed that a neural stimulant facilitates recovery of the flux discrimination but not a pattern discrimination (see Meyer and Meyer, 1977, for review). They suggest that the role of the neo-cortex is to add context in facilitating access to subcortically established engrams rather than to act as a store for memories. LeVere and Morlock supported this conclusion using a behavioral interference test, a simultaneous brightness discrimination (lit versus dark alleys, 1973), and a successive flux discrimination (both alleys lit means go one way, both dark means go the other way, 1974). Rats were initially trained to one habit, then had the visual neocortex removed. If the cortical lesion actually destroyed the memory, then it should not matter whether the subjects were trained to the same habit or to the opposite habit (go the other way). LeVere and Morlock found that training to the opposite habit took substantially longer to learn, indicating that the old memory still existed and interfered with learning the opposite habit.

As was true in Lashley's time, no one has yet localized the memory for a flux discrimination. Lesions of visual subcortical structures (superior colliculus, lateral geniculate, pretectal area, accessory optic nuclei) or of the limbic system (septum, hippocampus, amygdala), in combination with visual decortication, do not prevent relearning. The fault probably lies within Walter Hunter's criticisms (1930) that there is not enough experimental control over the instrumental training task used in this research. The more general question is whether the neocortex is involved in any memory. Squire (1987) reviews the best evidence for cortical memory, which can also be interpreted as suggesting that the cortex is necessary for perception of the stimuli but not for memory itself. Clearly, these are issues that are not resolved and continue to be of interest.

Bibliography

Bauer, J. H., and Cooper, R. M. (1964). Effects of posterior cortical lesions on performance of a brightness discrimination task. Journal of Comparative and Physiological Psychology 58, 84-93.

Cloud, M. D., Meyer, D. R., and Meyer, P. M. (1982). Induction of recoveries from injuries to the cortex: Dissociation of equipotential and regionally specific mechanisms. Physiological Psychology 10, 66-73.

Hunter, W. S. (1930). A consideration of Lashley's theory of the equipotentiality of cerebral action. Journal of Genetic Psychology 3, 455-468.

Klüver, H. (1942). Functional significance of the geniculo-striate system. In H. Klüver, ed., Visual mechanisms. Lancaster, PA: Jaques Cattell Press.

Lashley, K. S. (1929). Brain mechanisms and intelligence: A quantitative study of injuries to the brain. Chicago: University of Chicago Press.

—— (1935). The mechanism of vision: XII. Nervous structures concerned in the acquisition and retention of habits based on reactions to light. Comparative Psychology Monographs 11, 43-79.

Lavond, D. G., and Dewberry, R. G. (1980). Visual form perception is a function of the visual cortex: II. The rotated horizontal-vertical and oblique-stripes pattern problems. Physiological Psychology 8, 1-8.

LeVere, T. E., and Morlock, G. W. (1973). Nature of visual recovery following posterior neodecortication in the hooded rat. Journal of Comparative and Physiological Psychology 83, 62-67.

Meyer, D. R., and Meyer, P. M. (1977). Dynamics and bases of recoveries of functions after injuries to the cerebral cortex. Physiological Psychology 5, 133-165.

Spear, P. D., and Braun, J. J. (1969). Nonequivalence of normal and posteriorly neodecorticated rats on two brightness discrimination problems. Journal of Comparative and Physiological Psychology 67, 235-239.

Squire, L. R. (1987). Memory and the brain. New York: Oxford University Press.

DavidLavond