Search

The responses of 230 single neurons in the inferior colliculus of the horseshoebat to single tones have been studied, emphasizing systematic analysis of the effective frequency bands, dynamic properties and the time course of responses. Distribution of the units' best excitatory frequencies (BEF) is: low frequency neurons 23% (BEF 3-65 kHz); FM-frequency neurons 25% (BEF 65-81 kHz, i.e., frequencies occurring in the FM-part of the bat's echo signal); filter neurons 45% (BEF 81-88 kHz, i.e., frequencies occurring in the stabilized CF-part of the bat's echo=reference frequency (RF)); high frequency neurons about 7% (BEF > 88 kHz). Tuning curves show conventional shapes (Fig. 1), apart from those of filter neurons, which are extremely narrow. Accordingly, Q-values (BEF divided by the bandwidth of the tuning curve at 10 dB above threshold) are 80-450 in filter neurons (Fig. 2). Response patterns (Fig. 3) are similar to those of Nucleus cochlearis units (transient, sustained, negative and complex responders) with an increased percentage of complex responders up to 38% and a decreased number of transient responders. All types of spike-count functions are found (Fig. 4); nonmonotonic ones dominating. Maximal spike counts are not at the BEF but a few kHz below. Distinct upper thresholds, especially at the BEF of filter neurons (Fig. 5) lead to abrupt changes in activity by slightly shifting stimulus frequency or intensity. The hallmark of inferior colliculus neurons is inhibition, disclosed by distinct inhibitory areas enfolding and overlapping excitatory ones (Figs. 3 and 5). Duration of inhibition varies with stimulus frequency, but is largely independent of stimulus duration (Fig. 6), whereas rebound of inhibition depends on stimulus duration building up periodic rebound activities, if stimulus duration is lengthened. In addition, there are neurons responding only periodically, regardless of stimulus frequency and intensity (Fig. 7). Inhibition is discussed in terms of improving the neuronal signal/spontaneous noise ratio and altering responsiveness of neurons after stimulation, so that these neurons may be suited to time processing in the acoustic pathway. [ABSTRACT FROM AUTHOR]

For echolocation, Rhinolophus ferrumequinum emits orientation sounds, each of which consists of a long constant-frequency (CF) component and short frequency-modulated (FM) components. The CF component is about 83 kHz and is used for Doppler-shift compensation. In this bat, single auditory nerve fibers and cochlear nuclear neurons tuned at about 83 kHz show low threshold and very sharp filter characteristics. The slopes of their tuning curves ranged between 1,000 and 3,500 dB/octave and their Q-10 dB values were between 20 and 400, 140 on the average (Figs. 3-5). The peripheral auditory system is apparently specialized for the reception and fine frequency analysis of the CF component in orientation sounds and Doppler-shift compensated echoes. This specialization is not due to suppression or inhibition comparable to lateral inhibition, but due to the mechanical specialization of the cochlea. Peripheral auditory neurons with the best frequency between 77 and 87 kHz showed not only on-responses, but also off-responses to tonal stimuli (Figs. 1, 2, and 6). The off-responses with a latency comparable to that of N-off were not due to a rebound from either suppression or inhibition, but probably due to a mechanical transient occurring in the cochlea at the cessation of a tone burst. [ABSTRACT FROM AUTHOR]

The tonotopic organization of the inferior colliculus (IC) in two echolocating bats, Hipposideros speoris and Megaderma lyra, was studied by multiunit recordings. In Hipposideros speoris frequencies below the range of the echolocation signals (i.e. below 120 kHz) are compressed into a dorsolateral cap about 400-600 Μm thick. Within this region, neuronal sheets of about 4-5 Μm thickness represent a 1 kHz-band. In contrast, the frequencies of the echolocation signals (120-140 kHz) are overrepresented and occupy the central and ventral parts of the IC (Fig. 3). In this region, neuronal sheets of about 80 Μm thickness represent a 1 kHz-band. The largest 1 kHz-slabs (400-600 Μm) represent frequencies of the pure tone components of the echolocation signals (130-140 kHz). The frequency of the pure tone echolocation component is specific for any given individual and always part of the overrepresented frequency range but did not necessarily coincide with the BF of the thickest isofrequency slab. Thus hipposiderid bats have an auditory fovea (Fig. 10). In the IC of Megaderma lyra the complete range of audible frequencies, from a few kHz to 110 kHz, is represented in fairly equal proportions (Fig. 7). On the average, a neuronal sheet of 30 Μm thickness is dedicated to a 1 kHz-band, however, frequencies below 20 kHz, i.e. below the range of the echolocation signals, are overrepresented. Audiograms based on thresholds determined from multiunit recordings demonstrate the specific sensitivities of the two bat species. In Hipposideros speoris the audiogram shows two sensitivity peaks, one in the nonecholocating frequency range (10-60 kHz) and one within the auditory fovea for echolocation (130-140 kHz). Megaderma lyra has extreme sensitivity between 15-20 kHz, with thresholds as low as −24 dB SPL, and a second sensitivity peak at 50 kHz (Fig. 8). In Megaderma lyra, as in common laboratory mammals, Q-values of single units do not exceed 30, whereas in Hipposideros speoris units with BFs within the auditory fovea reach Q-values of up to 130. In Megaderma lyra, many single units and multiunit clusters with BFs below 30 kHz show upper thresholds of 40-50 dB SPL and respond most vigorously to sound intensities below 30 dB SPL (Fig. 9). Many of these units respond preferentially or exclusively to noise. These features are interpreted as adaptations to detection of prey-generated noises. The two different tonotopic arrangements (compare Figs. 3 and 7) in the ICs of the two species are correlated with their different foraging behaviours. It is suggested that pure tone echolocation and auditory foveae are primarily adaptations to echo clutter rejection for species foraging on the wing close to vegetation. [ABSTRACT FROM AUTHOR]

Bats use the rich food resources of the night by specializing in audition. They emit short echolocation sounds and listen to the echoes returning from potential prey. The bat's auditory system analyzes spectral and temporal parameters of echoes for detecting, locating, and identifying a target. Different bat species have solved the problem of acoustic target detection and pattern recognition even in clustered situations by focusing on certain acoustical features of a target. The specialized motion detection by horseshoe bats, for instance, analyzes small echofrequency shifts modulated onto a long constant frequency echolocation signal. These frequency modulations are Doppler shifts within echoes returning from wing beating insects. For detecting modulations as small as 10 Hz or 0.01%, horseshoe bats have in the cochlea an extremely narrow filter (Q?500) matched to the carrier frequency (i.e., echolocation sound) of 83 kHz. The filter is realized by structural differentiations of the basilar membrane and the filter frequencies are represented on the basilar membrane in an expanded fashion. We have called this specialized patch of the basilar membrane an ''acoustical fovea.'' The ''foveal frequencies'' are largely overrepresented in the tonopical arrangement of the ascending auditory pathway. The bats have developed a feedback system which lowers the emitted frequency during flight in such a way that the Doppler shifted echofrequency is kept precisely at a fixed reference frequency of the fovea. This feedback system and other neuronal data disclose an intricate coupling of the auditory and vocalizing system. The evolution of echolocation in bats has driven the analyzing capacities of audition in both frequency and time domain close to theoretical limits. Investigations of such specialized systems give fascinating insights into capacities and possible general principles of auditory information processing. [ABSTRACT FROM AUTHOR]