Processing of auditory information in insects

Timescale-invariant representation of acoustic communication signals by a bursting neuron

Acoustic communication often involves complex sound motifs in which the relative durations of individual elements, but not their absolute durations, convey meaning. Decoding such signals requires an explicit or implicit calculation of the ratios between time intervals. Using grasshopper communication as a model, we demonstrate how this seemingly difficult computation can be solved in real time by a small set of auditory neurons. One of these cells, an ascending interneuron, generates bursts of action potentials in response to the rhythmic syllable-pause structure of grasshopper calls. Our data show that these bursts are preferentially triggered at syllable onset; the number of spikes within the burst is linearly correlated with the duration of the preceding pause. Integrating the number of spikes over a fixed time window therefore leads to a total spike count that reflects the characteristic syllable-to-pause ratio of the species while being invariant to playing back the call faster or slower. Such a timescale-invariant recognition is essential under natural conditions, because grasshoppers do not thermoregulate; the call of a sender sitting in the shade will be slower than that of a grasshopper in the sun. Our results show that timescale-invariant stimulus recognition can be implemented at the single-cell level without directly calculating the ratio between pulse and interpulse durations.

The origin of adaptation in the auditory pathway of locusts is specific to cell type and function

We investigated the origin of spike frequency adaptation within a layered sensory network: the auditory pathway of locusts. Spike frequency adaptation as observed in an individual neuron may arise because of intrinsic or presynaptic adaptation mechanisms. To separate the contribution of different mechanisms, we recorded from the same cell during acoustic and intracellular current stimulation. We studied three identified neuron types that are representative for each network layer and participate in processing auditory patterns and localizing sound sources. By comparing current and acoustic stimulation, three distinct patterns of the distribution of adaptation mechanisms within the sensory network emerged: (1) balanced influence of both intrinsic and presynaptic adaptation mechanisms in an interneuron that summates over several receptor afferents (TN1), (2) predominantly inhibiting input as the source for spike frequency adaptation in a cell that transmits both pattern representation and directional information (BSN1), (3) primarily intrinsic, spike-triggered adaptation currents within an interneuron coding exclusively for direction (AN2). The time courses of spike frequency adaptation differed significantly between the cells types. Using the adaptation time constants, we were able to predict signal transmission properties for the different cells. We conclude that the adaptation mechanisms differ greatly among interneurons within this sensory pathway and are a function of their role in information processing.

A complex mechanism of call recognition in the katydid Neoconocephalus affinis (Orthoptera: Tettigoniidae)

Acoustic pattern recognition is important for bringing together males and females in many insect species. We used phonotaxis experiments on a walking compensator to study call recognition in the katydid Neoconocephalus affinis, a species with a double-pulsed call and an atypically slow pulse rate for the genus. Call recognition in this species is unusual because females require the presence of two alternating pulse amplitudes in the signal. A Fourier analysis of the stimulus-envelopes revealed that females respond only when both the first and second harmonics of the AM spectrum are of similar amplitude. The second harmonic is generated by the amplitude difference between the two pulses making up a pulse-pair. Females respond to double pulses that have been merged into a single pulse only if this amplitude modulation is preserved. Further experiments suggest that females use a resonance mechanism to recognize the pulse rate of the call, supporting a neural model of rate recognition in which periodic oscillations in membrane potential are used to filter the pulse rate of the signal. Our results illustrate how a reduction in pulse rate extends the opportunities for females to evaluate fine-scale temporal properties of calls, and provide further evidence for the importance of oscillatory membrane properties in temporal processing. The results are discussed with regard to evolutionary changes in call recognition mechanisms within the genus.

Auditory sensitivity of an acoustic parasitoid (Emblemasoma sp., Sarcophagidae, Diptera) and the calling behavior of potential hosts

Using field broadcasts of model male calling songs, we tested whether Tibicen pruinosa and T. chloromera (Hemiptera: Cicadidae) are candidate hosts for acoustic parasitoid flies. The model calling song of T. pruinosa attracted 90% of the flies (Sarcophagidae: Emblemasoma sp.; all larvapositing females) when broadcast simultaneously with the model T. chloromera song, a phonotactic bias reconfirmed in single song playbacks. In paired broadcasts of model T. pruinosa songs with different relative amplitudes (3 dB or 6 dB), significantly more flies were attracted to the more powerful song, a result consistent with the responses predicted by a model proposed by Forrest and Raspet [1994]. Using intracellular recordings and dye injections, we characterized the sensitivity of auditory units in sound-trapped flies. Intracellular recordings from six auditory units (5 interneurons, 1 afferent) revealed best sensitivity for frequencies near 3-4 kHz, matching the predominant spectral components of the calling songs of both species of cicada. Interestingly, although flies could be attracted to T. pruinosa broadcasts throughout the day, hourly censuses of singing males revealed that calling occurred exclusively at dusk. Furthermore, the duration of the dusk chorus in T. pruinosa was significantly shorter than the midday chorus of the less attractive song of T. chloromera. We propose that the tight temporal aggregation of the dusk chorus time could function to reduce risk from attracted parasitoids.

Influence of sound pressure level on the processing of amplitude modulations by auditory neurons of the locust

Typical features of natural sounds are amplitude changes at different time scales. In many species, amplitude modulations constitute decisive cues to recognize communication signals. Since these signals should be recognizable over a broad intensity range, we investigated how the encoding of amplitude modulations by auditory neurons depends on sound pressure level. Identified neurons that represent different processing stages in the locusts' auditory pathway were stimulated with sinusoidal modulations of a broad band noise carrier, at different intensities, and characteristic parameters of modulation transfer functions (MTFs) were determined. The corner frequencies of temporal MTFs turned out to be independent of intensity for all neurons except one. Furthermore, for none of the neurons investigated corner frequencies were significantly correlated with spike rate, indicating a remarkable intensity invariance of the upper limits of temporal resolution. The shape of the tMTFs changed with increasing intensity from a low-pass to a band-pass for receptors and local neurons, while no consistent change was observed for ascending neurons. The best modulation frequency depended on intensity and spike rate, especially for receptors and local neurons. Remarkably, the adaptation state of some neurons turned out to be independent of the spike rate during the modulation part of the stimulus.

The role of frequency, phase and time for processing of amplitude modulated signals by grasshoppers

Acoustic signals consist of pressure changes over time and can thus be analyzed in the frequency- or in the time-domain. With behavioural experiments we investigated which frequency components (FC) are necessary for the recognition of the periodic envelope of the conspecific song by females of the grasshopper Chorthippus biguttulus. Further, we determined up to which frequency component phase information is required which would indicate processing in the time domain. Responses of females revealed that signals composed of FC between 10 and 50 Hz are sufficient for recognition of the song envelope. A systematic reduction in the number of FC showed that no single frequency component was required; signals without the fundamental frequency were still highly attractive and only three FC may be sufficient for song recognition. Phase changes for frequencies up to 40 Hz strongly changed the attractiveness of song signals but only little at 50 Hz. Females were also tested with rectangular signals in which pause duration was varied. Evidently, and despite the high attractiveness of song signals with a "missing fundamental", females evaluated the attractiveness of signals in the time-domain, since the selectivity for pause duration predicted the responses to signals composed from FC well.

Auditory discrimination of amplitude modulations based on metric distances of spike trains

Sound envelope cues play a crucial role for the recognition and discrimination of communication signals in diverse taxa, such as vertebrates and arthropods. Using a classification based on metric similarities of spike trains we investigate how well amplitude modulations (AMs) of sound signals can be distinguished at three levels of the locust's auditory pathway: receptors and local and ascending neurons. The spike train metric has the advantage of providing information about the necessary evaluation time window and about the optimal temporal resolution of processing, thereby yielding clues to possible coding principles. It further allows one to disentangle the respective contributions of spike count and spike timing to the fidelity of discrimination. These results are compared with the traditional paradigm using modulation transfer functions. Spike trains of receptors and two primary-like local interneurons enable an excellent discrimination of different AM frequencies, up to about 150 Hz. In these neurons discriminability depends almost completely on the timing of spikes, which must be evaluated with a temporal resolution of <5 ms. Even short spike-train segments of 150 ms, equivalent to five to eight spikes, suffice for a high (70%) discrimination performance. For the third level of processing, the ascending interneurons, the overall discrimination accuracy is reduced. Spike count differences become more important for the discrimination whereas the exact timing of spikes contributes less. This shift in temporal resolution does not primarily depend on the investigated stimulus space. Rather it appears to reflect a transformation of how amplitude modulations are represented at more central stages of processing.

Resonant neurons and bushcricket behaviour

The resonant properties of the intrinsic dynamics of single neurons could play a direct role in behaviour. One plausible role is in the recognition of temporal patterns, such as that seen in the auditory communication systems of Orthoptera. Recent behavioural data from bushcrickets suggests that this behaviour has interesting resonance properties, but the underlying mechanism is unknown. Here we show that a very simple and general model for neural resonance could directly account for the different behavioural responses of bushcrickets to different song patterns.

Comprehensive classification of the auditory sensory projections in the brain of the fruit fly Drosophila melanogaster

We established a comprehensive projection map of the auditory receptor cells (Johnston's organ neurons: JONs) from the antennae to the primary auditory center of the Drosophila brain. We found 477 +/- 24 cell bodies of JONs, which are arranged like a "bottomless bowl" within the auditory organ. The target of the JONs in the brain comprises five spatially segregated zones, each of which is contributed by bundles of JON axons that gradually branch out from the antennal nerve. Four zones are confined in the antennal mechanosensory and motor center, whereas one zone further extends over parts of the ventrolateral protocerebrum and the subesophageal ganglion. Single-cell labeling with the FLP-out technique revealed that most JONs innervate only a single zone, indicating that JONs can be categorized into five groups according to their target zones. Within each zone, JONs innervate various combinations of subareas. We classified these five zones into 19 subareas according to the branching patterns and terminal distributions of single JON axons. The groups of JONs that innervate particular zones or subareas of the primary auditory center have their cell bodies in characteristic locations of the Johnston's organ in the antenna, e.g., in concentric rings or in paired clusters. Such structural organization suggests that each JON group, and hence each zone of the primary auditory center, might sense different aspects of sensory signals.

Tuning the drum: the mechanical basis for frequency discrimination in a Mediterranean cicada

Cicadas are known to use sound to find a mate. While the mechanism employed by male cicadas to generate loud calling songs has been described in detail,little information exists to explain how their ears work. Using microscanning laser Doppler vibrometry, the tympanal vibrations in the cicada Cicadatra atra are measured in response to acoustic playbacks. The topographically accurate optical measurements reveal the vibrational behaviour of the anatomically complex tympanal membrane. Notably, the tympanal ridge, a distinct structural element of the tympanum that is a link to the receptor cells, undergoes mechanical vibrations reminiscent of a travelling wave. In effect, the frequency for which the maximum deflection amplitude is observed regularly decreases from the apex to the base of the ridge. It is also shown that whilst female ears are mechanically tuned to the male's song, the male's tympanum is only partially tuned to its own song. This study establishes the presence of a peripheral auditory mechanism that can potentially process auditory frequency analysis. In view of the importance of acoustic signalling in cicadas, this unconventional tympanal mechanism may be employed in the context of species recognition and sexual selection.

Flying in tune: sexual recognition in mosquitoes

Mosquitoes hear with their antennae, which in most species are sexually dimorphic. Johnston, who discovered the mosquito auditory organ at the base of the antenna 150 years ago, speculated that audition was involved with mating behaviour. Indeed, male mosquitoes are attracted to female flight tones. The male auditory organ has been proposed to act as an acoustic filter for female flight tones, but female auditory behavior is unknown. We show, for the first time, interactive auditory behavior between males and females that leads to sexual recognition. Individual males and females both respond to pure tones by altering wing-beat frequency. Behavioral auditory tuning curves, based on minimum threshold sound levels that elicit a change in wing-beat frequency to pure tones, are sharper than the mechanical tuning of the antennae, with males being more sensitive than females. We flew opposite-sex pairs of tethered Toxorhynchites brevipalpis and found that each mosquito alters its wing-beat frequency in response to the flight tone of the other, so that within seconds their flight-tone frequencies are closely matched, if not completely synchronized. The flight tones of same-sex pairs may converge in frequency but eventually diverge dramatically.

Spectral selectivity during phonotaxis: a comparative study in Neoconocephalus (Orthoptera: Tettigoniidae)

The calls of male Neoconocephalus have most energy concentrated in a relatively narrow low-frequency band. In N. robustus this low-frequency band is centered around 7 kHz, whereas calls of N. nebrascensis and N. bivocatus have center frequencies close to 10 kHz. The importance of the position of the low-frequency band for female phonotaxis in these three species was determined using a walking compensator. Female N. robustus showed significant phonotaxis towards call frequencies from 5 to 10 kHz, and spectral selectivity towards higher frequencies did not change with stimulus amplitude. Significant responses in N. nebrascensis and N. bivocatus occurred at significantly higher frequency ranges than in N. robustus. In these species,spectral selectivity changed with stimulus amplitude; at 68 dB sound pressure level (SPL), upper cut-off frequency was significantly lower than at 80 dB SPL in both species. Adding a higher harmonic to the conspecific carrier frequency had a strong inhibitory effect on phonotaxis in N. robustus: at higher relative amplitudes of the harmonic, phonotaxis was completely suppressed. Adding a higher harmonic to the conspecific carrier frequency had a much weaker but significant inhibitory effect in N. nebrascensisand little, if any, effect in N. bivocatus. The processing of song spectrum in the sensory system is discussed with regard to the differences in spectral selectivity among the three species. The sharp spectral selectivity of N. robustus is interpreted as an adaptation for species isolation.

Synaptic ultrastructure of Drosophila Johnston's organ axon terminals as revealed by an enhancer trap

The role of auditory circuitry is to decipher relevant information from acoustic signals. Acoustic parameters used by different insect species vary widely. All these auditory systems, however, share a common transducer: tympanal organs as well as the Drosophila flagellar ears use chordotonal organs as the auditory mechanoreceptors. We here describe the central neural projections of the Drosophila Johnston's organ (JO). These neurons, which represent the antennal auditory organ, terminate in the antennomechanosensory center. To ensure correct identification of these terminals we made use of a beta-galactosidase-expressing transgene that labels JO neurons specifically. Analysis of these projection pathways shows that parallel JO fibers display extensive contacts, including putative gap junctions. We find that the synaptic boutons show both chemical synaptic structures as well as putative gap junctions, indicating mixed synapses, and belong largely to the divergent type, with multiple small postsynaptic processes. The ultrastructure of JO fibers and synapses may indicate an ability to process temporally discretized acoustic information.

Pulse-rate recognition in an insect: evidence of a role for oscillatory neurons

Various mechanisms have been proposed as the neural basis for pulse-rate recognition in insects and anurans, including models employing high- and low-pass filters, autocorrelation, and neural resonance. We used the katydid Tettigonia cantans to test these models by measuring female responsiveness on a walking compensator to stimuli varying in temporal pattern. Each model predicts secondary responses to certain stimuli other than the standard conspecific pulse rate. Females responded strongly to stimuli with a pulse-rate equal to half the standard rate, but not to stimuli with double the standard rate. When every second pulse or interval was varied in length, females responded only when the resulting stimuli were rhythmic with respect to the period of the standard signal. These results provide evidence rejecting the use of either high-/low-pass filter networks or autocorrelation mechanisms. We suggest that rate recognition in this species relies on the resonant properties of neurons involved in signal recognition. According to this model, signals with a pulse rate equal to the resonant frequency of the neurons stimulate the female to respond. The results are discussed with regard to both neural and evolutionary implications of resonance as a mechanism for signal recognition.

Mechanisms underlying phonotactic steering in the cricket Gryllus bimaculatus revealed with a fast trackball system

Phonotactic steering behaviour of the cricket G. bimaculatus was analysed with a new highly sensitive trackball system providing a spatial and temporal resolution of 127 microm and 0.3 ms, respectively. Orientation to artificial calling songs started at 45 dB SPL, it increased up to 75 dB SPL and then saturated. When exposed to two identical patterns of different intensity, crickets significantly steered towards the louder sound pattern, whenever the intensity difference was greater than 1 dB. Bilateral latency differences in sound presentation did not always cause clear orientation towards the leading side. The overall walking direction depended on the number of sound pulses perceived from the left or right side with the animals turning towards the side providing the larger number of pulses. The recordings demonstrated rapid changes in walking direction performed even during a chirp. These rapid steering responses occurred with a latency of 55-60 ms, well before the central nervous system had time to evaluate the temporal structure of a whole chirp. When every other sound pulse was presented from opposite directions, the crickets followed the temporal pattern of sound presentation and rapidly steered towards the left and right side. Steering towards individual sound pulses does not agree with the proposal that crickets analyse the quality of sound patterns and then steer towards the better pattern. Rather, these experiments suggest that fast steering to single sound pulses determines the lateral deviation of the animals and that complex auditory orientation emerges from this simple mechanism of auditory steering.

Increase of neuronal response variability at higher processing levels as revealed by simultaneous recordings

A key problem for neuronal information processing is the variability of spike trains, something that is likely to constrain the encoding of sensory signals. We measured interspike-interval variability (coefficient of variation) as well as spike-count variability (Fano factor) in the metathoracic auditory system of locusts. We performed simultaneous intracellular recordings at the first three processing levels to establish identical physiological conditions. This allows us to assess whether variability is generated anew or is reduced during synaptic transmission and processing. Both the interspike-interval variability as well as the spike-count variability revealed similar trends and showed an increase from the periphery to higher processing levels. This result was confirmed by single-cell recordings. A comparison of ascending interneurons coding for sound direction and those encoding sound patterns showed that the latter respond more reliably to repeated stimulus presentations. In general, the variability of spiking responses was much lower than expected from a Poisson process. Furthermore, we observed a strong dependence of variability on the spike rate, which differed at the three levels investigated. The differences in spike rates account for most of the differences in variability observed between processing levels. For auditory receptors, we found a good agreement between the Fano factor and the squared coefficient of variation, suggesting similarities to a renewal process of spike generation at the periphery. At the level of interneurons, the Fano factor was lower than the squared coefficient of variation; this indicates a higher reliability than expected from the interspike-interval distribution.

Variability of spike trains and the processing of temporal patterns of acoustic signals-problems, constraints, and solutions

Object recognition and classification by sensory pathways is rooted in spike trains provided by sensory neurons. Nervous systems had to evolve mechanisms to extract information about relevant object properties, and to separate these from spurious features. In this review, problems caused by spike train variability and counterstrategies are exemplified for the processing of acoustic signals in orthopteran insects. Due to size limitations of their nervous system we expect to find solutions that are stripped to the computational basics. A key feature of auditory systems is temporal resolution, which is likely limited by spike train variability. Basic strategies to reduce such variability are to integrate over time, or to average across several neurons. The first strategy is constrained by its possible interference with temporal resolution. Grasshoppers do not seem to explore temporal integration much, in spite of the repetitive structure of their songs, which invites for 'multiple looks' at the signal. The benefits of averaging across neurons depend on uncorrelated responses, a factor that may be crucial for the performance and evolution of small nervous systems. In spite of spike train variability the temporal information necessary for the recognition of conspecifics is preserved to a remarkable degree in the auditory pathway.