ABSTRACT Understanding how individuals detect and recognize signals emitted by conspecifics is fundamental to discussions of animal communication. The species pair Gymnotus omarorum and Brachyhypopomus gauderio , found in syntopy in Uruguay, emit species-specific electric organ discharge (EOD) that can be sensed by both species. The aim of this study was to unveil whether either of these species is able to identify a conspecific EOD, and to investigate distinctive recognition signal features. We designed a forced-choice experiment using a natural behavior (i.e. tracking electric field lines towards their source) in which each fish had to choose between a conspecific and a heterospecific electric field. We found a clear pattern of preference for a conspecific waveform even when pulses were played within 1 Hz of the same rate. By manipulating the time course of the explored signals, we found that the signal features for preference between conspecific and heterospecific waveforms were embedded in the time course of the signals. This study provides evidence that pulse Gymnotiformes can recognize a conspecific exclusively through species-specific electrosensory signals. It also suggests that the key signal features for species differentiation are probably encoded by burst coder electroreceptors. Given these results, and because receptors are sharply tuned to amplitude spectra and also tuned to phase spectra, we extend the electric color hypothesis used in the evaluation of objects to apply to communication signals.

ABSTRACT As in most sensory systems, electrosensory images in weakly electric fish are encoded in two parallel pathways, fast and slow. From work on wave-type electric fish, these fast and slow pathways are thought to encode the time and amplitude of electrosensory signals, respectively. The present study focuses on the primary afferents giving origin to the slow path of the pulse-type weakly electric fish Gymnotus omarorum . We found that burst duration coders respond with a high-frequency train of spikes to each electric organ discharge. They also show high sensitivity to phase-frequency distortions of the self-generated local electric field. We explored this sensitivity by manipulating the longitudinal impedance of a probe cylinder to modulate the stimulus waveform, while extracellularly recording isolated primary afferents. Resistive loads only affect the amplitude of the re-afferent signals without distorting the waveform. Capacitive loads cause large waveform distortions aside from amplitude changes. Stepping from a resistive to a capacitive load in such a way that the stimulus waveform was distorted, without changing its total energy, caused strong changes in latency, inter-spike interval and number of spikes of primary afferent responses. These burst parameters are well correlated suggesting that they may contribute synergistically in driving downstream neurons. This correlation also suggests that each receptor encodes a single parameter in the stimulus waveform. The finding of waveform distortion sensitivity is relevant because it may contribute to: (a) enhance electroreceptive range in the peripheral ‘electrosensory field’, (b) a better identification of living prey at the ‘foveal electrosensory field’ and (c) detect the presence and orientation of conspecifics. Our results also suggest a revision of the classical view of amplitude and time encoding by fast and slow pathways in pulse-type electric fish.

ABSTRACT Sensory systems play a key role in social behavior by mediating the detection and analysis of communication signals. In mormyrid fishes, electric signals are processed within a dedicated sensory pathway, providing a unique opportunity to relate sensory biology to social behavior. Evolutionary changes within this pathway led to new perceptual abilities that have been linked to increased rates of signal evolution and species diversification in a lineage called ‘clade A’. Previous field observations suggest that clade-A species tend to be solitary and territorial, whereas non-clade-A species tend to be clustered in high densities suggestive of schooling or shoaling. To explore behavioral differences between species in these lineages in greater detail, I studied population densities, social interactions, and electric signaling in two mormyrid species, Gnathonemus victoriae (clade A) and Petrocephalus degeni (non-clade A), from Lwamunda Swamp, Uganda. Petrocephalus degeni was found at higher population densities, but intraspecific diversity in electric signal waveform was greater in G. victoriae . In the laboratory, G. victoriae exhibited strong shelter-seeking behavior and competition for shelter, whereas P. degeni were more likely to abandon shelter in the presence of conspecifics as well as electric mimics of signaling conspecifics. In other words, P. degeni exhibited social affiliation whereas G. victoriae exhibited social competition. Further, P. degeni showed correlated electric signaling behavior whereas G. victoriae showed anti-correlated signaling behavior. These findings extend previous reports of social spacing, territoriality, and habitat preference among mormyrid species, suggesting that evolutionary divergence in electrosensory processing relates to differences in social behavior.

Summary Weakly electric gymnotiform and mormyrid fish generate and detect weak electric fields to image their worlds and communicate. These multi-purpose electric signals are generated by electrocytes, the specialized electric organ (EO) cells that produce the electric organ discharge (EOD). Just over 50 years ago the first experimental analyses of electrocyte physiology demonstrated that the EOD is produced and shaped by the timing and waveform of electrocyte action potentials (APs). Electrocytes of some species generate a single AP from a distinct region of excitable membrane, and this AP waveform determines EOD waveform. In other species, electrocytes possess two independent regions of excitable membrane that generate asynchronous APs with different waveforms, thereby increasing EOD complexity. Signal complexity is further enhanced in some gymnotiforms by the spatio-temporal activation of distinct EO regions with different electrocyte properties. For many mormyrids, additional EOD waveform components are produced by APs that propagate along stalks that connect postsynaptic regions to the main body of the electrocyte. I review here the history of research on electrocyte physiology in weakly electric fish, as well as recent discoveries of key phenomena not anticipated during early work in this field. Recent areas of investigation include the regulation of electrocyte activity by steroid and peptide hormones, the molecular evolution of electrocyte ion channels, and the evolutionary selection of ion channels expressed in excitable cells. These emerging research areas have generated renewed interest in electrocyte function and clear future directions for research addressing a broad range of new and important questions.

Summary Electroreception is an ancient vertebrate sense with a fascinating evolutionary history involving multiple losses as well as independent evolution at least twice within teleosts. We review the phylogenetic distribution of electroreception and the morphology and innervation of electroreceptors in different vertebrate groups. We summarise recent work from our laboratory that has confirmed the homology of ampullary electroreceptors in non-teleost jawed vertebrates by showing, in conjunction with previously published work, that these are derived embryonically from lateral line placodes. Finally, we review hypotheses to explain the distribution of electroreception within teleosts, including the hypothesis that teleost ampullary and tuberous electroreceptors evolved via the modification of mechanosensory hair cells in lateral line neuromasts. We conclude that further experimental work on teleost electroreceptor development is needed to test such hypotheses.

Summary A complete understanding of animal signal evolution necessitates analyses of both the proximate (e.g. anatomical and physiological) mechanisms of signal generation and reception, and the ultimate (i.e. evolutionary) mechanisms underlying adaptation and diversification. Here we summarize the results of a synthetic study of electric diversity in the species-rich neotropical electric fish genus Gymnotus . Our study integrates two research directions. The first examines the proximate causes of diversity in the electric organ discharge (EOD) – which is the carrier of both the communication and electrolocation signal of electric fishes – via descriptions of the intrinsic properties of electrocytes, electrocyte innervation, electric organ anatomy and the neural coordination of the discharge (among other parameters). The second seeks to understand the ultimate causes of signal diversity – via a continent-wide survey of species diversity, species-level phylogenetic reconstructions and field-recorded head-to-tail EOD (ht-EOD) waveforms (a common procedure for characterizing the communication component of electric fish EODs). At the proximate level, a comparative morpho-functional survey of electric organ anatomy and the electromotive force pattern of the EOD for 11 species (representing most major clades) revealed four distinct groups of species, each corresponding to a discrete area of the phylogeny of the genus and to a distinct type of ht-EOD waveform. At the ultimate level, our analyses (which emphasize the ht-EOD) allowed us to conclude that selective forces from the abiotic environment have had minimal impact on the communication component of the EOD. In contrast, selective forces of a biotic nature – imposed by electroreceptive predators, reproductive interference from heterospecific congeners, and sexual selection – may be important sources of diversifying selection on Gymnotus signals.

SUMMARY Peripheral filtering is a fundamental mechanism for establishing frequency tuning in sensory systems. By contrast, detection of temporal features, such as duration, is generally thought to result from temporal coding in the periphery, followed by an analysis of peripheral response times within the central nervous system. We investigated how peripheral filtering properties affect the coding of stimulus duration in the electrosensory system of mormyrid fishes using behavioral and electrophysiological measures of duration tuning. We recorded from individual knollenorgans, the electrosensory receptors that mediate communication, and found correlated variation in frequency tuning and duration tuning, as predicted by a simple circuit model. In response to relatively high intensity stimuli, knollenorgans responded reliably with fixed latency spikes, consistent with a temporal code for stimulus duration. At near-threshold intensities, however, both the reliability and the temporal precision of responses decreased. Evoked potential recordings from the midbrain, as well as behavioral responses to electrosensory stimulation, revealed changes in sensitivity across the range of durations associated with the greatest variability in receptor sensitivity. Further, this range overlapped with the natural range of variation in species-specific communication signals, suggesting that peripheral duration tuning affects the coding of behaviorally relevant stimuli. We measured knollenorgan, midbrain and behavioral responses to natural communication signals and found that each of them were duration dependent. We conclude that at relatively low intensities for which temporal coding is ineffective, diversity among sensory receptors establishes a population code, in which duration is reflected in the population of responding knollenorgans.

SUMMARY Previous work has shown that animals alter their locomotor behavior to increase sensing volumes. However, an animal’s own movement also determines the spatial and temporal dynamics of sensory feedback. Because each sensory modality has unique spatiotemporal properties, movement has differential and potentially independent effects on each sensory system. Here we show that weakly electric fish dramatically adjust their locomotor behavior in relation to changes of modality-specific information in a task in which increasing sensory volume is irrelevant. We varied sensory information during a refuge-tracking task by changing illumination (vision) and conductivity (electroreception). The gain between refuge movement stimuli and fish tracking responses was functionally identical across all sensory conditions. However, there was a significant increase in the tracking error in the dark (no visual cues). This was a result of spontaneous whole-body oscillations (0.1 to 1 Hz) produced by the fish. These movements were costly: in the dark, fish swam over three times further when tracking and produced more net positive mechanical work. The magnitudes of these oscillations increased as electrosensory salience was degraded via increases in conductivity. In addition, tail bending (1.5 to 2.35 Hz), which has been reported to enhance electrosensory perception, occurred only during trials in the dark. These data show that both categories of movements – whole-body oscillations and tail bends – actively shape the spatiotemporal dynamics of electrosensory feedback.

SUMMARY Brown ghost knife fish ( Apteronotus leptorhynchus ) can briefly increase their electric organ discharge (EOD) frequency to produce electrocommunication signals termed chirps. The chirp rate increases when fish are presented with conspecific fish or high-frequency (700–1100 Hz) electric signals that mimic conspecific fish. We examined whether A. leptorhynchus also chirps in response to artificial low-frequency electric signals and to heterospecific electric fish whose EOD contains low-frequency components. Fish chirped at rates above background when presented with low-frequency (10–300 Hz) sine-wave stimuli; at 30 and 150 Hz, the threshold amplitude for response was 1 mV cm –1 . Low-frequency (30 Hz) stimuli also potentiated the chirp response to high-frequency (∼900 Hz) stimuli. Fish increased their chirp rate when presented with two heterospecific electric fish, Sternopygus macrurus and Brachyhypopomus gauderio , but did not respond to the presence of the non-electric fish Carassius auratus . Fish chirped to low-frequency (150 Hz) signals that mimic those of S. macrurus and to EOD playbacks of B. gauderio . The response to the B. gauderio playback was reduced when the low-frequency component (<150 Hz) was experimentally filtered out. Thus, A. leptorhynchus appears to chirp specifically to the electric signals of heterospecific electric fish, and the low-frequency components of heterospecific EODs significantly influence chirp rate. These results raise the possibility that chirps function to communicate to conspecifics about the presence of a heterospecific fish or to communicate directly to heterospecific fish.

SUMMARY The five sensory modalities of humans are also found in a wide range of invertebrates. Other vertebrates have evolved additional special senses, such as the magnetic sense, which are also found in some invertebrates. However, there remain a few sensory abilities that curiously appear to be found in either vertebrates or invertebrates, but not both. For example, electrosensitivity — the ability to detect electric fields in water — which should benefit vertebrates and invertebrates alike, is apparently only used by vertebrates. However, recent reports suggest that some invertebrates could have an electric sense. Here we examine that possibility further and demonstrate a behavioural threshold to low-level electrical fields in two freshwater invertebrates. The responses are not low enough for them to detect the Earth's magnetic field as some other electroreceptive species can do, but sufficiently low for them to use in navigation or prey and predator detection. This finding challenges the current view of the sensory world of aquatic invertebrates and has implications for the evolution of this ability.

SUMMARY Elasmobranch fishes (sharks, skates and rays) possess highly sensitive electrosensory systems, which enable them to detect weak electric fields such as those produced by potential prey organisms. Different species have unique electrosensory pore numbers, densities and distributions. Functional differences in detection capabilities resulting from these structural differences are largely unknown. Stingrays and other batoid fishes have eyes positioned on the opposite side of the body from the mouth. Furthermore, they often feed on buried prey, which can be located non-visually using the electrosensory system. In the present study we test functional predictions based on structural differences in three stingray species ( Urobatis halleri, Pteroplatytrygon violacea and Myliobatis californica )with differing electrosensory system morphology. We compare detection capabilities based upon behavioral responses to dipole electric signals(5.3–9.6 μA). Species with greater ventral pore numbers and densities were predicted to demonstrate enhanced electrosensory capabilities. Electric field intensities at orientation were similar among these species, although they differed in response type and orientation pathway. Minimum voltage gradients eliciting feeding responses were well below 1 nVcm –1 for all species regardless of pore number and density.

SUMMARY Paddlefish use their peculiar rostrum to detect minute electric fields from their main prey, small water fleas. Electroreceptors over the rostrum and head sense these fields and send the information into a single hindbrain area, the dorsal octavolateral nucleus (DON). From there, information is sent to various midbrain structures, including the tectum. The response properties of primary afferent fibers and DON units has been well investigated, but nothing is known about electrosensory units in the midbrain. Here we recorded the responses of single units in the midbrain tectum and DON to uniform electric fields. Tectal units exhibited little spontaneous activity and responded to sine waves with a few, well phase-locked spikes. Phase locking was still significant at amplitudes one order of magnitude lower than in the DON. If stimulated with sinusoidal electric fields of different frequencies, phase locking in DON units decreased proportionally with frequency whereas the response of tectal units depended little on frequency. This is in agreement with behavioral studies showing that relevant frequencies range from DC to ca 20 Hz.

SUMMARY Learning to predict the component in the sensory information resulting from the organism's own activity enables it to respond appropriately to unexpected stimuli. For example, the elasmobranch dorsal octavolateral nucleus (DON) can apparently extract the unexpected component (i.e. generated by nearby organisms) from the incoming electrosensory signals. Here we introduce a novel and unique experimental approach that combines the advantages of in vitro preparations with the integrity of in vivo conditions. In such an experimental system one can study, under control conditions, the cellular and network mechanisms that underlie cancellation of expected sensory inputs. Using extracellular and intracellular recordings we compared the dynamics and spatiotemporal organization of the electrosensory afferent nerve and parallel fiber inputs to the DON. The afferent nerve has a low threshold and high conduction velocity; a stimulus that recruits a small number of fibers is sufficient to activate the principal neurons. The excitatory postsynaptic potential in the principal cells evoked by afferent nerve fibers has fast kinetics that efficiently reach the threshold for action potential. In contrast, the parallel fibers have low conduction velocity, high threshold and extensive convergence on the principal neurons of the DON. The excitatory postsynaptic response has slow kinetics that provides a wide time window for integration of inputs. The highly efficient connection between the afferent nerve and the principal neurons in the DON indicates that filtration occurring in the DON cannot be mediated simply by summation of the parallel fibers' signals with the afferent sensory signals. Hence we propose that the filtering may be mediated via secondary neurons that adjust the principal neurons'sensitivity to afferent inputs.

SUMMARY One difficulty in understanding the brain is that of linking the structure of the neurons with their computational roles in neural circuits. In this paper we address this subject in a relative simple system, the fast electrosensory pathway of an electric fish, where sensory images are coded by the relative latency of a volley of single spikes. The main input to this path is a stream of discrete electric images resulting from the modulation of a self-generated carrier by the environment. At the second order cell level, a window of low responsiveness, reducing potential interference from other stimuli, follows activation of the path. In the present study, we further characterize the input–output relationship at the second order neurons by recording field potentials, and ascertain its cellular basis using in vitro whole cell patch recordings. The field potentials from freely behaving, socially interacting fish were obtained from chronically implanted fish restrained in a mesh pen. In addition, at the end of some experiments the fish was curarized and the fast electrosensory path responses to artificial stimuli were further explored. These in vivo approaches showed that larger stimuli cause larger and longer windows of low responsiveness. The simple spherical geometry of the second order cells allowed us to unveil the membrane mechanisms underlying this phenomenon in vitro . These spherical cells respond with a single spike at the onset of current steps of any amplitude and duration, showing inward and outward rectification, and a long refractory period. We postulate that a low-threshold K + conductance generates the outward rectification. The most parsimonious interpretation of our data indicates that slow deactivation of this conductance causes the long refractory period. These non-linear properties of the membrane explain the single spiking profile of spherical cells and the low-responsiveness window observed in vivo . Since the electric organ discharges are emitted at intervals slightly longer than the duration of the low-responsiveness window,we propose that the described cellular mechanisms allow fish streaming self-generated images.

SUMMARY Most biological sensory systems benefit from multiple sensors. Elasmobranchs (sharks, skates and rays) possess an array of electroreceptive organs that facilitate prey location, mate location and navigation. Here, the perceived electrosensory landscape for an elasmobranch approaching prey is mathematically modeled. The voltages that develop simultaneously in dozens of separate sensing organs are calculated using electrodynamics. These voltages lead directly to firing rate modifications in the primary afferent nerves. The canals connecting the sense organs to an elasmobranch's surface exhibit great variation of location and orientation. Here, the voltages arising in the sense organs are found to depend strongly on the geometrical distribution of the corresponding canals. Two applications for the modeling technique are explored: an analysis of observed elasmobranch prey-capture behavior and an analysis of morphological optimization. For the former, results in specific predator-prey scenarios are compared with behavioral observations, supporting the approach algorithm suggested by A. Kalmijn. For the latter, electrosensory performance is contrasted for two geometrical models of multiple sense organs,a rounded head and a hammer-shaped head.

Elasmobranch fishes have an electroreceptive system which they use for prey detection and orientation. Sensory inputs in this system are corrupted by a form of reafference generated by the animal's own ventilation. However, we show here that in the carpet shark, Cephaloscylium isabella , as in two previously studied batoid species, this ventilatory ‘noise’ is reduced by sensory processing within the medullary nucleus of the electrosensory system. It has been proposed that the noise cancellation is achieved by a common mode rejection mechanism. One prediction of this hypothesis is that secondary neurons within the medullary nucleus should have both excitatory and inhibitory components to their receptive fields. This prediction is experimentally verified here. Projection neurons of the medullary nucleus in the carpet shark typically have a focal excitatory, and a diffuse inhibitory, receptive field organization including a component of contralateral inhibition. This result provides strong support for the hypothesis that ventilatory suppression in the elasmobranch electrosensory system is achieved by a common mode mechanism. Note: Department of Biology, Wesleyan University, Middletown, CT 06457, USA. Present address: Department of Zoology, University of Auckland, Auckland, New Zealand.

The electroreceptors of elasmobranchs are strongly modulated by the fish's own ventilation but this source of potential interference is suppressed within the medulla. The mechanism for the suppression is thought to be based on the common mode nature of the ventilatory noise, i.e. it is of the same amplitude and phase for all of the electroreceptors, compared with environmental electric fields which affect the receptors differentially. Evidence for the common mode suppression hypothesis is provided here in skates by the observation that the response to an artificial common mode stimulus that is independent of ventilation and delivered through an electrode inserted into the animal's gut is also suppressed by the medullary neurons; the extent to which a particular neuron suppresses the responses to the gut stimulus and to ventilation is similar. In addition, a potential modulation of 5–150μV is measured between the skate's interior and the sea water during ventilation and this appears to be responsible for the self-stimulation. By passing d.c. or sinusoidal currents through the gut electrode it is demonstrated that this ventilatory potential is due to the variable shunting of a standing d.c. potential across the fish's skin by the opening and closing of the mouth and gill slits during ventilation. Osmoregulatory ion-pumping appears to contribute to the production of the d.c. potential. Note: Present address: Department of Biology, Wesleyan University, Middletown, CT 06457, USA. Present address: Department of Zoology, University of Auckland, Auckland, New Zealand.

Eigenmannia lineata is a tropical South American gymnotiform fish that is both electrogenic and electroreceptive. Its electric organ generates a constant-frequency wave discharge (EOD) that is sexually dimorphic in waveform and harmonic content. Eigenmannia discriminates digitally synthesized, natural male and female EOD waveforms of identical frequency and amplitude. Experiments were devised to investigate behaviourally the sensory mechanism of this discrimination. Both the content in higher harmonics (as seen in an amplitude spectrum) and the waveform (phase spectrum) could provide the cues Eigenmannia uses for discrimination. Five Eigenmannia were trained to discriminate artificially generated stimuli composed of the fundamental frequency, f 1 , and its harmonic, f 2 , of a frequency twice that of f 1 and of weaker intensity (−13 to −3dB re f 1 ; a similar variation is found among the EODs of different Eigenmannia individuals). The rewarded stimulus, S+ (with a strong f 2 intensity of −3dB re f 1 ), remained constant throughout the experiments. The family of negative (S-) stimuli all had a phase difference of 0°, that is, no phase difference, between their harmonics, while the S+ stimulus had a phase difference of 90°. Therefore, all S- stimuli differed from the S+ stimulus in waveform, while one S- stimulus had an amplitude spectrum identical to that of the S+ stimulus. All fish discriminated any S- signal tested from the S+ stimulus, including the S- stimulus with an identical amplitude spectrum, both when the test signals had identical energy contents (that is, slightly different peak-to-peak amplitudes) and when they had identical peak-to-peak amplitudes (hence, slightly different energy contents). These results arc evidence for a true waveform (time domain) sensitivity of Eigenmannia . A sensory mechanism is proposed that enables Eigenmannia to assess the P/N ratio of a wave signal, that is, the duration ratio of positive (P) and negative (N) half-waves between zero-crossings of a stimulus cycle (the S+ and all S- stimuli, as well as female and male EODs, differ in this regard).