Walter Heiligenberg - US grants

This research involves the study sensory information processing in the electrosensory system of gymnotiform fish. Electrosensory information is used in social communication, such as the jamming avoidance response (JAR), as well as for the detection and identification of objects. By intracellular labelling of physiologically identified neurons, the P.I. can explore morphological features and their significance for particular tasks in the processing of information. By tracing anatomical projections of identified classes of neurons, the routes and way stations of pathways along which specific forms of information are processed can be determined. Research efforts will be concentrated upon connections from the torus semicircularis and tectum opticum of the midbrain to the complex of the nucleus electrosensorius (NE) in the pretectum. Neurons in the torus and the tectum have more general response properties and are less sensitive to specific stimulus features than are neurons in the NE. The higher response specificity and sensitivity of neurons in the NE appears to result from extensive neuronal convergence.//

We intend to continue our studies on sensory information processing and the control of simple behavioral responses in electric fish. We have chosen the electric sense as a model system for the exploration of basic neuronal design features which also appear to be involved in more complex sensory systems, such as audition and vision in mammals. Of particular interest are ordered representations of stimulus spaces in laminated neuronal networks which map certain stimulus variables along particular axes within the plane of lamination and process different aspects of the stimulus complexity by specialized neurons located in different layers. By advancing through successively higher-order central nervous stations of stimulus processing, we intend to reach the motor output level and thus to provide a complete neuronal theory for the control of a simple behavioral response. For the Jamming Avoidance Response in Eigenmannia, this goal is now within reach. This proposal focuses upon three projects: 1. The significance of representations of afferent information in multiple maps, such as the projections of identical primary electrosensory afferents to three separate, somatotopically ordered maps in the hindbrain. This research involves intracellular recording and labelling of all available cell types to study their dynamic properties under various stimulus regimes and their anatomical architecture. Differences between the three maps should yield clues for their functional specialization. 2. The significance of descending recurrent input from midbrain and cerebellum to all three electrosensory maps in the hindbrain. The effects of small lesions upon dynamic properties of identified cell types as well as upon responses of the whole animal will be studied. 3. The organization of the medullary electric organ pacemaker. Although not a strictly laminated structure, this nucleus consists of two loosely segregated types of electrically coupled, rhythmically discharging cells which control the electric organ. To investigate causes for the extreme stability of its rhythm, physiological experiments will be conducted on intact preparations, on excised pacemakers maintained in a slice chamber and on pacemaker cells grown in tissue culture. These studies will involve dual intracellular recording and labelling of coupled cells and anatomical investigation of their connections.

Due to its relative simplicity, the electrosensory system is ideally suited for the integration of behavioral and cellular approaches and, therefore, has led to most detailed explanations of stimulus perception and motor performance at the single-cell level. The electric sense shares basic principles in the coding of sensory information with more advanced sensory modalities, such as vision and audition in birds and mammals, and thus provides a convenient model system for studying neuronal mechanisms of information processing in general. Most significantly, some behavioral responses of electric fish are so robust that they remain intact in physiological preparations, thus allowing simultaneous studies at the behavioral and cellular level. The current project will continue to explore the structural and functional organization of the torus semicircularis, a laminated midbrain structure homologous to the inferior colliculus of mammals, which processes temporal and spatial aspects of electrosensory information. By using quartz glass pipettes (prepared by the new 'laser puller' of the Sutter Company), even very small cells, with soma diameters in the range of 5 micra, can now be penetrated and labelled intracellularly. This offers the opportunity of labelling and recording from neurons that could barely be explored in the past. One of these small-neuron types resides in lamina 6 of the torus and computes temporal disparities in the arrival of two inputs, comparable to the processing of interaural time differences in auditory systems. The temporal resolution of these neurons appears to be in the microsecond range. In addition to studies in largely intact fish, we intend to explore functional and structural properties of the torus in slice preparations. Due to the anatomical organization of projections to the torus, one could stimulate lamina 6 of the torus in a near natural manner and explore the physiological properties of small cells under the enhanced mechanical stability of a slice preparation.

A key question in neuroscience is how sensory information is processed to control adaptive behavioral responses. An excellent system in which to study this issue is found in fish called weakly electric fish. These fish have an electric organ like that in the electric eel, but much weaker, and produce periodic discharges of particular waveforms. These fish also have electrosensory systems that detect the electric field produced by the fish itself or by other nearby electric fish. The electrosensory system is used for orientation and the pattern of signals is used for communication. The brain of these animals is now known to process the signals using certain kinds of spatial maps, comparable to those in the visual system, and using certain kinds of temporal analysis, comparable to processes in the auditory system. This project has its focus on a particular group of nerve cells, or nuclei, in the brain, located at an interface between the spatial representation of the midbrain, and the nuclei that produce the electric signalling behavior such as the prepacemaker nucleus, which modulate the discharge pattern of the electric organ. Anatomical and physiological techniques are used to label physiologically identified nerve cells to trace their functional connections in this circuit. Behaviorally relevant stimulus paradigms will be used to test the response properties of individual neurons along pathways involved in control of specific behavioral responses. Clarification of this circuitry for integrating complex sensory input into a discrete behavioral output in this model system will provide major advances for neuroscience, with importance to sensory systems, motor systems, integration, and behavior.