Margaret Livingstone - US grants

For the last five years I have been involved in two, rather different, areas of research. The first is a study of the biochemical processes involved in learning, analyzing learning and memory mutants of Drosophila melanogaster. Drosophila can learn. They exhibit time-associative learning in several behavioral tests: They will selectively avoid odors previously associated with electric shock or will show a preference for odors previously associated with a sucrose reward. Five single gene mutations in Drosophila have been isolated which either abolish learning in these tests or shorten memory. Research in many other systems, both vertebrate and invertebrate, has suggested that modulation of intracellular cyclic AMP levels plays an important role in behavioral plasticity. Results with Drosophila support this idea: Three mutants that do not learn have defects that directly or indirectly alter cyclic AMP metabolism. This project will initially focus on one learning mutant, rutabaga, with a defect in the enzyme adenylate cyclase. Biochemical and genetic studies of this mutant should yield information about the enzyme itself and how it is involved in behavioral plasticity. We will also use specific biochemical selection screens to isolate other mutants in this pathway and characterize them. The second project is a collaboration and represents continuation of work begun in 1979. The immediate objective of these studies is to determine the anatomical connections and physiological significance of certain cytochrome oxidase dense structures in primate areas 17 and 18. We have already shown that cells in the densely staining blobs of area 17 are concerned with color, that they are closely interconnected in 17, and are specifically connected to the thinner of two sets of densely staining stripes in 18. We plan to examine these connections in more detail, using axonal transport methods, and to extend our recent physiological studies to area 18, to learn how the color information is further processed in the thin stripes. We have strong hints that the thicker stripes are concerned with stereoscopic depth perception, and plan to examine this further. One long-range objective is to find out more about the effects of neuromodulators in mammalian central nervous systems. A study on the influence of sleep and arousal in area 17 will be continued and extended, within the next year or so.

DESCRIPTION: The first specific aim concerns the functional organization of disparity tuned cells in monkey visual cortex. The primary approach will be to make long penetrations through V1, V2, and V3/V3A, quantitatively testing disparity sensitivities and correlating those properties with cortical layer and cytochrome oxidase compartments. In V1 the principal aim will be to refine the current view that most disparity sensitive neurons are found in layer 4B by examining deeper layers, which are likely to contain other disparity sensitive neurons, and by examining the distribution relative to the location of cytochrome oxidase stripes. In V2 the experiments will pursue the observation that the cytochrome oxidase stripes are functionally heterogeneous, and will explore whether neurons with different disparity preferences are grouped or clustered in different regions within a stripe. Recordings in V3 will follow up on observations of clustering of different types of disparity preferences by looking for a columnar organization. Besides the recordings, injections of tracers will be made into V2 and V3 to show the organization of connections between regions with established properties. The second aim will address how disparity sensitivity is associated with other response properties in individual neurons.Efforts will focus on color, motion, orientation, and responses to random dot stereograms. The question of the relationship between disparity sensitivity and color, motion and orientation is important, because there are several lines of evidence that suggest that disparity sensitivity may be preferentially associated with motion processing, and less with color sensitivity. Conclusive data on this point will help clarify the extent to which these properties are segregated. The third aim will examine disparity sensitivity in visually naive animals. The final aim is to learn whether squirrel monkeys, which lack ocular dominance columns, show stereopsis in their evoke potentials and optokinetic reflexes.

The ability to assess depth depends on many cues, one of which is stereopsis--the determination of the distance of a object from differences in the images in the two eyes. Like color vision, stereoscopic depth perception is frequently defective in humans, but unlike most forms of color blindness, defects in stereopsis must be caused by abnormal wiring in the central nervous system. To understand these defects will thus require an understanding of the normal central- nervous mechanisms for stereopsis. Furthermore, an understanding of how one particular well-defined calculation is done by the central nervous system should shed light on how, in general, the brain performs complex calculations. Cells selective for differences in the images in the two eyes have been described and characterized in primates, but the receptive-field mechanisms underlying stereopsis have only been addressed in anesthetized cats, and not at all in monkeys. The goal of this study is to fill in this large gap in our understanding of binocular interactions and depth perception. A recently developed technique for mapping receptive-fields in alert fixating macaques will be used to determine the receptive-field organization of a large series of neurons in V1. Stereoscopic depth selectivity in each cell will be compared with its receptive-field organization in the two eyes. This should provide an understanding of how binocular interactions are used to generate depth selectivity.

Mechanisms underlying direction-selectivity were studied in V1 of alert fixating macaque monkeys. Some direction-selective cells showed delayed asymmetric inhibition, some showed as shifting excitatory time-course across the receptive field, and some showed both. Both the direction of the spatial offset of the inhibition and the direction of the shift in excitatory response time-course correlated with the cells' preferred directionality. Part of the mechanism underlying a shifting response time-course may be delayed asymmetric inhibition. The data suggest that asymmetric inhibition is the major determinant for directionality in these cells, though both mechanisms could contribute. Based on this physiology, a simple, single-cell model is proposed, consistent with the known anatomy of some direction-selective cells in layer 6 (Meynert cells).

DESCRIPTION (Adapted from applicant's abstract): Some cells in cat and monkey visual cortex show strong preferences for the direction of stimulus motion. Primary visual cortex is the first stage in the geniculocortical visual pathway where direction selectivity is encountered, so the mechanisms underlying direction selectivity must be found there. This project is a detailed, quantitative receptive-field mapping study of direction-selective cells in macaque V1. These studies will establish the spatiotemporal pattern(s) of excitatory and inhibitory ON and OFF inputs to directional cells. Maps will be generated to look at interactions across directional cells' receptive fields, to find out how inputs from different parts of the receptive field combine spatially and temporally to result in directionality. The data will be analyzed to ask which neuronal models could account for this kind of specificity. Direction selectivity is a fundamental property of visual systems in general. In many other sensory and motor systems various kinds of sequence-dependent specificities have been observed. This study offers the potential to understand the underlying mechanism of one such sequence-specific selectivity. Though this is a basic-science level project, understanding this one kind of temporal information processing may shed light on diseases that are thought to involve defects in temporal processing, such as dyslexia and schizophrenia.

[unreadable] DESCRIPTION (provided by applicant): The project goal is to explore the functional anatomy and mechanisms of face detection and recognition. Faces are a complex yet highly constrained subset of all possible visual shapes. Face-selective regions in the temporal lobe have been identified using functional magnetic resonance imaging (fMRI) as showing increased blood flow in response to visual stimulation with faces as opposed to images of other objects. MRI can therefore be used to target single-unit recording to these face-selective regions. The ability to specifically target fMRI-identified face patches for single-unit recording provides an unprecedented opportunity to record repeatedly from areas of temporal cortex entirely selective for faces. One bilateral set of face patches is in the lower bank of the superior temporal sulcus and likely corresponds to the human fusiform face area (FFA). This laboratory has successfully targeted single-unit recording to this face patch in two subjects, and in this region almost all (98%) of the single units recorded so far have shown strong response selectivity for faces. Single-unit electrophysiology and fMRI will be used in parallel and in conjunction to explore the functional specialization in the face-selective regions of the temporal lobe and the hierarchical organization of face processing, in particular whether distinct regions are dedicated to processing specific aspects of faces, such as emotional expression, view angle, or gaze direction. Single-unit reverse correlation mapping and fMRI will be used to map selectivity to large sets of faces varying in view angle, position, scale, illumination, and expression to explore the development of invariant coding along the face-processing hierarchy. Reverse-correlation mapping will be used to map selectivity for multiple facial features in a parameterized artificial-face space to determine how features are combined in these areas to generate face-selective response properties from simpler features or feature constellations. Face recognition is one of the most important aspects both socially and developmentally of higher-level visual processing. Understanding how the brain carries out such a complex yet specific process as face recognition will yield insights into human cognition as well as into neurological and psychiatric conditions in which face recognition is impaired, such as prosopagnosia, schizophrenia, and autism. [unreadable] [unreadable] [unreadable]