Cerebellum-like structures in fish appear to act as adaptive sensory processors, in which learned predictions about sensory input are generated and subtracted from actual sensory input, allowing unpredicted inputs to stand out. Pairing sensory input with centrally originating predictive signals, such as corollary discharge signals linked to motor commands, results in neural responses to the predictive signals alone that are 'negative images' of the previously paired sensory responses. Adding these 'negative images' to actual sensory inputs minimizes the neural response to predictable sensory features. At the cellular level, sensory input is relayed to the basal region of Purkinje-like cells, whereas predictive signals are relayed by parallel fibres to the apical dendrites of the same cells. The generation of negative images could be explained by plasticity at parallel fibre synapses. We show here that such plasticity exists in the electrosensory lobe of mormyrid electric fish and that it has the necessary properties for such a model: it is reversible, anti-hebbian (excitatory postsynaptic potentials (EPSPs) are depressed after pairing with a postsynaptic spike) and tightly dependent on the sequence of pre- and postsynaptic events, with depression occurring only if the postsynaptic spike follows EPSP onset within 60 ms.

Distance determination in animals can be achieved by visual or non-visual cues. Weakly electric fish use active electrolocation for orientation in the dark. By perceiving self-produced electric signals with epidermal electroreceptors, fish can detect, locate and analyse nearby objects. Distance discrimination, however, was thought to be hardly possible because it was assumed that confusing ambiguity could arise with objects of unknown sizes and materials. Here we show that during electrolocation electric fish can measure the distance of most objects accurately, independently of size, shape and material. Measurements of the 'electric image' projected onto the skin surface during electrolocation revealed only one parameter combination that was unambiguously related to object distance: the ratio between maximal image slope and maximal image amplitude. However, slope-to-amplitude ratios for spheres were always smaller than those for other objects. As predicted, these objects were erroneously judged by the fish to be further away than all other objects at an identical distance. Our results suggest a novel mechanism for depth perception that can be achieved with a single, stationary two-dimensional array of detectors.

The electrosensory lobe (ELL) of mormyrid electric fish is one of several cerebellum-like sensory structures in fish that remove predictable features of the sensory inflow. This adaptive process obeys anti-Hebbian rules and appears to be mediated by associative depression at the synapses between parallel fibers and Purkinje-like cells of ELL. We show here that there is also a nonassociative potentiation at this synapse that depends only on the repeated occurrence of the EPSP. The depression can be reversed by the potentiation and vice versa. Finally, we show that the associative depression requires NMDA receptor activation, changes in postsynaptic calcium, and the occurrence of a postsynaptic dendritic spike within a few milliseconds following EPSP onset.

The electrosensory lobe (ELL) of mormyrid electric fish is the first stage in the central processing of sensory input from electroreceptors. The responses of cells in ELL to electrosensory input are strongly affected by corollary discharge signals associated with the motor command that drives the electric organ discharge (EOD). This study used intracellular recording and staining to describe the physiology of three major cell types in the mormyrid ELL: the medium ganglion cell, the large ganglion cell, and the large fusiform cell. The medium ganglion cell is a Purkinje-like interneuron, whereas the large ganglion and large fusiform cells are efferent neurons that convey electrosensory information to higher stages of the system. Clear differences were observed among the three cell types. Medium ganglion cells showed two types of spikes, a small narrow spike and a large broad spike that were probably of axonal and dendro-somatic origin, respectively, whereas the large ganglion and large fusiform cells showed only large narrow spikes. Most of the medium ganglion cells and all of the large ganglion cells were inhibited by electrosensory stimuli in the center of their receptive fields, whereas the large fusiform cells were excited by such stimuli.Responses to the EOD corollary discharge were different in the three cell types, and these responses underwent plastic changes after a few minutes of pairing with an electrosensory stimulus. Plastic changes were also observed in medium and large ganglion cells after the corollary discharge was paired with depolarizing, intracellular current pulses.

1. This is the first of a series of papers on the electrosensory lobe and closely associated structures in electric fish of the family Mormyridae. The study describes the neuronal responses to sensory stimuli and to corollary discharge signals associated with the motor command that drives the electric organ discharge (EOD). The study is focused on the regions of the electrosensory lobe where primary afferent fibers from mormyromast electroreceptors terminate. 2. This first paper of the series describes the field potentials in the caudal lobe of the cerebellum and in the electrosensory lobe. It also describes the different types of unit activity in the caudal lobe of the cerebellum. Granule cells of the caudal lobe of the cerebellum provide the parallel fibers for most of the molecular layer of the electrosensory lobe. Determination of the input and responses of these cells is therefore an important part of the effort to understand the electrosensory lobe. 3. Corollary discharge field potentials evoked by the EOD motor command are very prominent in the caudal lobe of the cerebellum and in the electrosensory lobe. The potentials indicate that corollary discharge excitation affects first the granule cells of the caudal lobe and then, a few milliseconds later, the deeper cellular layers of the electrosensory lobe. The prominence and complexity of the field potentials indicate that corollary discharge signals have an important and varied role in the processing of electrosensory information by the mormyrid electrosensory lobe. 4. The field potentials evoked by electrosensory stimuli suggest that direct primary afferent excitation is limited to the granule and intermediate layers of the electrosensory lobe, as is indicated also by anatomic studies. 5. Proprioceptive units are the most common type of unit recorded in the granule cell region of the caudal lobe of the cerebellum (eminentia granularis posterior). These units have a regular discharge rate that changes tonically in response to slight bending of the trunk, bending of the tail, or bending of individual fins. Proprioceptive input will have a strong effect on the molecular layer of the electrosensory lobe and will thus modulate the responses of electrosensory lobe cells to electrosensory stimuli. Such proprioceptive input to the electrosensory lobe would allow the expected effects of body position changes to be accounted for in the processing of electrosensory information. 6. Units with stereotyped, short-latency corollary discharge bursts to the EOD motor command were the next most common type of unit in the eminentia granularis posterior. These corollary discharge units were not affected by sensory stimuli.(ABSTRACT TRUNCATED AT 400 WORDS)

Knollenorgan electroreceptors in mormyrid fish are concerned with electrocommunication, i.e., with detecting electric organ discharges (EODs) of other electric fish. But knollenorgan electroreceptors are also activated by the fish's own EOD. Potential interference by such self-stimulation is blocked by an inhibitory corollary discharge in the nucleus of the electrosensory lateral line lobe (NELL), the first central relay of the knollenorgan pathway. This study used intracellular recording and staining to examine the mechanism of the corollary inhibition and the specializations in anatomy and physiology that permit the accurate relaying of temporal information about the EODs of other fish. Several events are recorded inside primary knollenorgan afferents in addition to a large orthodromic action potential. The additional events include small orthodromic electronic epsps, postsynaptic action potentials, and a corollary discharge inhibitory postsynaptic potential (ipsp) associated with the EOD motor command. These additional events are also recorded in NELL cells and almost certainly originate there. Electrical coupling between afferents and cells makes it possible to observe the events inside primary afferents. The corollary discharge ipsp in the cell is associated with a conductance increase and inverts rapidly when recorded with chloride-containing electrodes, supporting a hypothesis of GABA mediation. The ipsp lasts longer in cells than in afferents. Each electrotonic excitatory postsynaptic potential (epsp) is probably caused by a single primary afferent, and any one of several epsps in a given cell seems capable of eliciting a postsynaptic spike in that cell. The epsps follow stimulation rates as high as 500/sec with minimal variability. No lateral inhibition is observed in NELL. These and other properties indicate that the knollenorgan pathway is specialized for temporal information rather than spatial or intensity information.

SUMMARY1. Synaptic transmission and neuronal morphology were studied in the nucleus tractus solitarius and in the dorsal vagal motor nucleus (solitary complex), in coronal brain-stem slices of rat or cat, superfused in vitro.2. Electrical stimulation of afferent fibres of the solitary tract evoked two different types of post-synaptic response recorded intracellularly in different solitary complex neurones. Labelling with horseradish peroxidase showed that these two sorts of orthodromically evoked responses were correlated with different post-synaptic neuronal morphologies.3. The majority of recorded neurones (n = 93) showed a prolonged reduction in excitability following the initial solitary-tract-evoked excitatory post-synaptic potential (e.p.s.p.). A smaller number of neurones (n = 53) showed a prolonged increase in excitability following solitary tract stimulation. In no case did the solitary tract stimulation induce a burst of action potentials at high frequency.4. The time-to-peak and the half-width of the initial solitary-tract-evoked e.p.s.p. were shorter in neurones with prolonged increased excitability than in those with prolonged reduced excitability. In neurones with prolonged reduced excitability, this e.p.s.p. was followed by a hyperpolarization lasting 60-100 ms. The latency of this inhibitory post-synaptic potential (i.p.s.p.) was 3-5 ms longer than that of the initial e.p.s.p. and its reversal potential was 10 mV more negative than the reversal potential of the response measured following application of y-aminobutyric acid or glycine. In neurones with prolonged increased excitability, at a membrane potential of -40 to -50 mV, the initial solitary tract e.p.s.p. was followed by a prolonged depolarization lasting 100-400 ms.5. Background synaptic activity was high in neurones with prolonged increased excitability, consisting of unitary e.p.s.p.s with an amplitude of more than 0-8 mV. This activity was increased for a period of 300-800 ms following solitary tract stimulation. Spontaneous excitatory potentials of more than 0-5 mV were not seen in neurones with prolonged reduced excitability. In these neurones, after intracellular injection of choride ions, reversed unitary i

Synaptic plasticity occurs in several regions of the vertebrate brain and is believed to mediate the storage of behaviorally significant information during learning. Synaptic plasticity is well demonstrated in most cases, but the behavioral meaning of the relevant neural signals and the behavioral role of the plasticity are uncertain. In this paper we describe a case of synaptic plasticity which involves identifiable sensory and motor signals and which appears to mediate the storage of an image of past sensory input. Corollary discharge signals associated with the motor command that drives the electric organ are prominent in the electrosensory lobe of mormyrid electric fish. Some of these corollary discharge signals elicit a negative image or representation of the electrosensory input pattern that has followed recent motor commands. When the temporal and spatial pattern of sensory input changes, the corollary discharge effect also changes in a corresponding manner. The cellular mechanisms by which the corollary discharge-evoked representation is stored were investigated by intracellular recording from cells of the electrosensory lobe and pairing intracellular current pulses with the corollary discharge signal. The results indicate that the representation of recent sensory input is stored by means of anti-Hebbian plasticity at the synapses between corollary discharge-conveying fibers and cells of the electrosensory lobe. The results also suggest that dendritic spikes and plasticity at inhibitory synapses are involved in the phenomenon.Many sensory regions of the brain are affected by signals of central origin that are associated with motor commands (1-3). These signals are known as efference copy (4) or corollary discharge (5) signals and prepare the sensory regions for the (re)afferent (4) input that follows a commanded motor action. Optimal interaction between reafferent and corollary discharge inputs requires an approximate match in the timing and spatial distribution of the two signals. This requirement for matching, and the likelihood of variation in reafferent input, suggests that some corollary discharge effects may be plastic. Plastic corollary discharge effects are in fact present in the mormyrid electrosensory lobe (6-8).The electrosensory lobe is the termination site for afferent fibers from the three classes of electroreceptors in mormyrid fish: knollenorgans, ampullary organs, and mormyromasts (9). This paper is concerned only with the regions that receive ampullary and mormyromast input. Ampullary afferents project to the ventrolateral zone of the electrosensory lobe cortex, and mormyromast afferents project to the medial and dorsolateral zones (Fig. 1A). The projections are somatotopically organized. The histological structure of the cortex is like that of the cerebellum and contains Purkinje-like cells (9, 10) with cell bodies in the deeper layers and apical dendrites in the molecular layer.Each electric organ discharge (EOD) evokes reafferent responses in ampullary and mormyromast fibers ...