Orientation Sounds Research Articles

Acuity of Echolocation in Collocalia hirundinacea (Aves: Apodidae), With Comments on the Distributions of Echolocating Swiftlets and Molossid Bats

The acuity of echolocation in Collocalia hirundinacea was studied by observing the birds' performances in an obstacle course constructed 10 m inside the mine they used as a roost. Birds flying into the mine missed 10 mm diameter rods significantly more often than rods 4 or 1.5 mm in diameter (73 % vs. 67% or 55%, respectively), but avoided even the 1.5 mm diameter rods. Birds flying through the obstacle course on their way out of the mine apparently, on some occasions, used vision to detect the rods, for they generally hit fewer rods than birds flying in. The orientation clicks of C. hirundinacea and their performances in the obstacle course were very similar to previous reports for the closely related Collocalia vanikorensis. The benefits of echolocation for Collocalia and the coincident distribution of Collocalia and a depauperate molossid bat fauna are discussed. Occupation of a cave or mine by the swiftlets does not preclude use of the same site by bats. THE ABILITY TO ECHOLOCATE is known from two groups of birds, the caprimulgid Steatornis caripensis (Griffin 1954) and at least four species in the apodid genus Collocalia (Griffin and Suthers 1970). In all of these birds the highest energy components of the orientation vocalizations are of low frequency ( < 10 KHz) relative to the orientation cries of microchiropteran bats, and consequently the acuity of their echolocation would not be expected to be as good as that of the Microchiroptera. The genus Collocalia includes from 12 to 20 species, depending upon which classification one follows (Mayr 1937; Peters 1964; Thomson 1964), and of these, C. brevirostris (Novick 1959), C. maxima (Medway 1959), C. fuciphaga (C. salangana; Medway 1967), and C. vanikorensis (Griffin and Suthers 1970) echolocate, while C. esculenta does not. To date there appear to be no other published data on orientation sounds, or their lack, in the remaining species. This study was undertaken to determine if Collocalia hirundinacea, a species very similar and perhaps closely related to C. vanikorensis, produced orientation sounds, and if it was capable of echolocation.

Coordinated activity of the middle-ear and laryngeal muscles in echolocating bats

The middle-ear muscles contract synchronously with the emission of orientation sounds and attenuate vocal self-stimulation. Apparently the vocalization center activates the middle-ear and laryngeal muscles in a coordinated manner for echolocation. With Myotis lucifugus, further experiments were performed on the coordinated activities of the crichothyroid (CTM) and stapedius muscles (SM) for vocalization and attenuation of vocal self-stimulation. We have also found that both the SM and CTM responded to acoustic stimuli. The CTM discharged action potentials to acoustic stimuli with a latency of 11 msec on the average. The shortest latency of the reflex was 8.0 msec, which was 4.6 msec longer than that of the acoustic SM reflex. Unlike SM fibers, tuning curves of CTM fibers were a simple triangle. The best frequencies ranged between 25 and 40 kHz and the Q−10-dB values ranged between 2 and 5. The lowest threshold measured was 50 dB SPL. [Supported by NSF Grant GB-40018.]

Peripheral specialization for analysis of Doppler-shifted echoes in the auditory system of the “CF-FM” bat, Pteronotus parnellii. I. Cochlear microphonic

The mustache bat (Pteronotus parnellii) emits orientation sounds which consist of long constant-frequency (CF) components followed by short frequency-modulated (FM) components. The second harmonic (61–62 kHz) of the CF component is an information-bearing element and is probably used for Doppler-shift compensation. The auditory system of this animal is sharply tuned to sounds at 61–62 kHz in terms of cochlear microphonic (CM). This sharp tuning is due to the mechanical specialization of the inner ear. Because of this specialization, CM evoked by 61–62-kHz tones slowly increases and decreases in amplitude at the onset and cessation of these stimuli. The CM after the cessation of the stimuli is called the after-response or CM-aft. The area of the inner ear tuned at 60–61 kHz shows a prominent nonlinearity and is located close to the sharply tuned area, so that CM responses to 60–61-kHz tone bursts show prominent transients at the onset and cessation of the stimuli. At the cessation, the CM first increases, then decreases in amplitude. Such a CM-aft is called the off-response or CM-off. Both the CM-on and CM-aft, are tuned at 61–62 kHz, while the CM-off is tuned at 60–61 kHz. [Supported by NSF Grant GB-40018.]

Peripheral control of acoustic signals in the auditory system of echolocating bats.

Many species of echolocating bats emit intense orientation sounds. If such intense sounds directly stimulated their ears, detection of faint echoes would be impaired. Therefore, possible mechanisms for the attenuation of self-stimulation were studied with Myotis lucifugus. The acoustic middle-ear-muscle reflex could perfectly and transiently regulate the amplitude of an incoming signal only at its beginning. However, its shortest latency in terms of electromyograms and of the attenuation of the cochlear microphonic was 3-4 and 4-8 msec, respectively, so that these muscles failed to attenuate orientation signals by the reflex. The muscles, however, received a message from the vocalization system when the bat vocalized, and contracted synchronously with vocalization. The duration of the contraction-relaxation was so short that the self-stimulation was attenuated, but the echoes were not. The tetanus-fusion frequency of tha stapedium muscle ranged between 260 and 320/sec. Unlike the efferent fibres in the lateral-line and vestibular systems, the olivo-cochlear bundle showed no sign of attenuation of self-stimulation.

Dynamic properties of the compensation system for doppler shifts in the bat,Rhinolophus ferrumequinum

1. Echolocation sounds of the batRhinolophus ferrumequinum were played back to the bat as artificial echoes shifted in frequency by a few kHz (0–4 kHz) without distortion of the amplitude and frequency pattern of the sound. The simulated Doppler shifts in the echoes were modulated either sinusoidally or in step functions. 2. The bats did not compensate for frequency shifts within one sound, but in the sounds subsequent to the introduction of the shift. 3. Responses to symmetrical sinusoidal modulations of the echo from −2 kHz to +2kHz demonstrate that in the dynamic situation the bat compensates for positive frequency shifts and ignores negative frequency shifts (Fig. 2). Asymmetrical sinusoidal modulations between 0 Hz and different positive frequency shifts (1, 2, 3, and 4 kHz) yielded amplitude and phase response curves for modulation frequencies between 0.01 and 2 Hz (Fig. 5, 6). The response to step functions gave the time course for positive and negative changes of the frequency shift (Fig. 7). 4. The Doppler compensation system shows nonlinearities: The response of the system depends on the sign of the frequency shift change. Positive changes of the frequency shift in sinusoidal modulations are optimally followed by the system, when it is considered as a sampling system, in which the pulse repetition rate is the sampling rate. The response to negative frequency shift changes is much slower (Fig. 6). With an increase of the modulation frequency the response is DC-shifted. The amplitude and phase response curves depend on the amplitude of the frequency shifts presented in the modulation (Table 1, 2). 5. The maximum compensation between two succeeding bat orientation sounds was 1400 Hz, giving a frequency change of about 10 Hz/msec. This value is about the same as the minimum frequency modulation of the constant frequency portion of the sound detected by the bat. 6. The dynamic properties of the Doppler shift compensation system are discussed in the context to neurophysiological findings on the hearing system of the bat.

Neurophysiological studies on echolocation systems in awake bats producing CF-FM orientation sounds.

ABSTRACT The bats Pteronotus parnelli, P. suapurensis and Noctilio leporinus emit orientation sounds first containing a constant-frequency (CF) and then a frequency-modulated (FM) component. P. parnellii produced a long CF with a second harmonic at 62 kHz to which its auditory system was sharply tuned. In the other two species, the CF was shorter and there was no sharp tuning. Electrical stimulation of the midbrain reticular formation and/or the central grey matter elicited vocalizations which were indistinguishable from those used for echolocation. The electrically-elicited vocalization was enhanced by acoustic stimuli. In P. parnellii, this vocal response was sharply tuned at 62–63 kHz and also to downward sweeping FM sounds. In P. suapurensis and N. leporinus, the vocal responses were prominent only to downward sweeping FM sounds. This indicates that the FM is important to echolocation in all these bats and that the CF component is more essential to echolocation in P. parnellii than to that in P. suapurensis and N. leporinus. The responses of primary auditory neurons to the onset and cessation of pure tone stimuli were due to mechanical events, not due to a rebound from neural inhibition. Masking experiments with P. parnellii indicate that the neural response at the cessation of a CF-FM sound similar to its orientation sound mainly consisted of the response to the FM component and not the off-response to the CF component. During vocalization, self-stimulation was reduced by contraction of middle-ear muscles. This was not due to the acoustic reflex which started to occur with a 6 msec latency.

Echolocation of insects by horseshoe bats

MOST bats of the Suborder Microchiroptera use echolocation during the pursuit of insect prey1. This conclusion was originally based on observation that bats emitted high frequency orientation sounds during pursuit of insects and that their repetition rate increased sharply during close pursuit manoeuvres. Experiments in controlled laboratory conditions later demonstrated that little brown bats (Myotis lucifugus) could capture fruit flies (Drosophila sp.) by echolocation when vision and passive hearing of insect flight sounds were not possible2.

Site of neural attenuation of responses to self-vocalized sounds in echolocating bats.

Bats of the genus Myotis emit intense orientation sounds for echolocation. If such sounds directly stimulated their ears, the detection of echoes from short distances would be impaired. In addition to the muscular mechanism in the middle ear, the bat has a neural mechanism in the brain for attenuation of responses to self-vocalized orientation and nonorientation sounds. This neural attenuating mechanism operates in the nucleus of the lateral lemniscus, reducing its activity by about 15 decibels, and it is synchronized with vocalization.

Directional sensitivity of echolocation system in bats producing frequency-modulated signals.

ABSTRACT Radiation patterns of the 55, 75 and 95 kHz components in frequency-modulated sounds emitted by the grey bat (Myotis grisescens) were studied. FM sounds similar to species-specific orientation sounds were elicited by electrical stimuli applied to the midbrain while the head of the animal was immobilized by a nail cemented to its skull. The main beam was emitted 5–10° downward from the eye-nostril line. The radiation angle at one half of maximum amplitude was 38° lateral, 18° up and 50° down at 55 kHz, 34° lateral, 8° up and 32° down at 75 kHz, and 30° lateral, 5° up and 25° down at 95 kHz. At 95 kHz, two prominent side lobes were present. The directional sensitivity of the auditory system (DSA) measured in terms of the potential evoked in the lateral lemniscus was studied with the grey bat (M. grisescens) and the little brown bat (M. lucifugus). The maximally sensitive direction moved toward the median plane with the increase in frequency from 35–95 kHz. The slope of the DSA curve increased from 0·3–0·6 dB/degree with frequency. The directional sensitivity of the echolocation system (DSE) was calculate using both the DSA curve and the radiation pattern of the emitted sound. The maximally sensitive direction of the echolocation system was 15° lateral to the median plane at 55 kHz and 2·5° lateral at 95 kHz. The slope of the DSE curve increased from 0·6 to 1·0 dB/degree with frequency. Thus, the higher the frequency of sound, the sharper was the directional sensitivity of the echolocation system. The interaural pressure difference (IPD), which appeared to be the essential cue for echolocation in Myotis, changed linearly with the azimuth angle from 0–30° lateral regardless of the frequency of sound, at respective rates of 0·4, 0·7, 0·3 and 0·4 dB/degree for 35, 55, 75 and 95 kHz sounds. Beyond 30°, the change in IPD was quite different depending on frequency. For 75 and 95 kHz sounds, the IPD stayed nearly the same between 30° and 90°. Thus, the 75–95 kHz components in FM orientation sounds were not superior to the 35 and 55 kHz components in terms of the IPD cue for echolocation. Assuming the just-detectable IPD and ITD to be 0·5 dB and 5 μsec respectively, as in man, the just-detectable azimuth difference of Myotis around the median plane would be 0·7–1·7° with the IPD cue and 11° with the ITD cue.

Neural Attenuation of Lateral Lemniscal Responses to Self-Vocalized Sounds in Echolocating Bats

Bats of the genus Myotis emit strong orientation sounds for echolocation. If such sounds directly stimulate their ears, the detection of echoes from short distances would be impaired. To reduce self-stimulation, the middle-ear muscles contract prior to and during vocalization, and quickly relax after its termination. In addition to this muscular mechanism, bats have neural mechanisms for the attenuation of responses of lateral lemniscal neurons to self-vocalized sounds. Three recording electrodes were placed in the auditory nerve, superior olivary complex (presumably the S-segment), and the nucleus of the lateral lemniscus, in order to record their summated activity evoked by self-vocalized sounds and the same sounds played back with a tape recorder. When the response of the auditory nerve was similar to both the sounds, the responses of the superior olivary complex were also the same, but the responses of lateral lemniscal neurons to emitted sounds were much smaller than those to playback sounds. Apparently the activity of the lateral lemniscal neurons was different, depending on whether the animal vocalizes or the same signal was delivered from an external source. Such neural attenuation was found when the bats emitted orientation sounds or squeaks [Supported by an NSF grant.]

Orientation sounds evoked from echolocating bats by electrical stimulation of the brain.

The brains of several representative species of echolocating bats were found to contain regions concerned with the emission of species-specific orientation sounds. When the dorsal part of the reticular formation in the midbrain or the lateral part of the central gray matter is stimulated with a short train of electric pulses, the bat moves its mouth and pinnae and emits sounds identical to orientation sounds, with a latency of 25 to 60 msec. The acoustic parameters of the electrically elicited sounds and the temporal patterns of emissions vary with the location of the stimulating electrodes. Bats of the genus Myotis (little brown bat) emitted short frequency-modulated (FM) sounds both in nature and when electrically stimulated. Pteronotus parnellii (mustache bat) responded with sounds containing long constant-frequency (CF) and short FM components that were typical of the species. Eptesicus fuscus (big brown bat), Pteronotus suapurensis (naked-backed bat), and Noctilio leporinus (fish-catching bat) also made characteristic orientation sounds when electrically stimulated. It appears that the midbrain contains neural circuits for the emission of species-specific orientation sounds and that a part of the vocalization system is located in the regions juxtaposed to the inferior colliculus, an important neural center for echolocation.

Vocal Responses of “CF-FM Bats” to Electric and Acoustic Stimuli

The experiments were performed for the exploration of coupling between the auditory and vocalization systems of awake bats, Pteronotus parnellii and P. suaprensis. The electric stimulation of the midbrain elicited the emission of sounds identical to species-specific orientation sounds consisting of CF and FM components. When acoustic stimuli were delivered with the electric stimuli, the electrically evoked sounds increased in amplitude and number. In P. parnellii these vocal responses were selectively sensitive to 62–63 kHz sounds, comparable to the CF component in its orientation sound, and also to downward-sweeping FM sounds comparable to the FM components in the orientation sound. In P. suaprensis, the threshold of the vocal response was high for pure tones and upward-sweeping FM sounds, but low for downward-sweeping FM sounds. The vocalization system was thus coupled with the auditory system by particular types of auditory neurons. These data indicate that the responses to acoustic stimuli were related not only to these power spectra, but also to properties of single auditory neurons and information-bearing elements in an emitted signal.

Coding and processing in the auditory systems of FM-signal-producing bats.

For echolocation, many species of bats emit frequency-modulated (FM) sounds. When the central gray matter or reticular formation in the midbrain is electrically stimulated, these bats produce FM sounds similar to FM-orientation signals. The auditory system is excited by these emitted sounds, but the responses of auditory neurons in the midbrain are attenuated by a neural mechanism operating synchronously with vocalization. This neural attenuating mechanism is present between the auditory nerve and the inferior colliculus and may be a part of mechanisms for effective echo-detection. Among the various orientation sounds emitted by bats, FM sounds are the most suited for echo-ranging. In the lateral lemniscus, the majority of neurons show a very short recovery period, as well as frequent facilitation of their responses to a second tone pulse, so that these neurons are specialized for echo-detection and carry information necessary for echolocation. In the inferior colliculus, single neurons exhibit a broad spectrum of recovery cycles, so that these are probably able to scale echoes from different distances for the distance measurement. Neural-network models of a clock for echo-ranging are discussed. In the inferior colliculus and the auditory cortex, there exist neurons specialized for processing FM signals. For the excitation of these FM-specialized neurons, the direction, range, and speed of frequency sweep are important factors. The FM-specialized neurons always have a large inhibitory area and respond to FM sounds sweeping across this area. Although seemingly paradoxical, such properties are easily explained by neural-network models. Orientation sounds of several species of bats consist of constant frequency (CF) and FM components. Neurons specialized for the analysis of CF sounds have inhibitory areas on both sides of a very narrow excitatory area. The responses of the FM-specialized neurons to certain FM sounds in the presence of either FM or CF sounds are discussed in relation to the inhibitory area.

Directionality of an Echo-Locating System of an “FM Bat.”

In order to understand the directional sensitivity of an echolocating system, directionalities of both emitted sounds and ears should be measured. The skull of Myotis grisescens was rigidly held with a nail cemented to it. FM sounds nearly identical to the species-specific orientation sounds were elicited by electric stimuli applied to the midbrain. The fields of 55-, 75-, and 95-kHz components in the FM sounds were measured with microphones moved around the bat's head. The amplitudes of these components were half at 30°–40° lateral, 25°–50° down, and 5°–20° up from the eye-nostril line. As expected, the higher the frequency, the narrower was the sound field. Directionality of the auditory system in terms of lateral lemniscal evoked potential was also measured at 55, 75, and 95 kHz. From these measurements, the directional sensitivity of the echolocating system was obtained, which was sharper for higher frequencies, but the most sensitive direction moved toward the midline with an increase in frequency, so that the most important cue for echolocalization, the interaural intensity difference, became smaller for higher frequencies.

Neural processing mechanisms in echolocating bats, correlated with differences in emitted sounds.

Bats that emit pulses containing constant wavelength (CW) components preceding an FM sweep show neural adaptations that differ conspicuously from those of bats emitting purely FM pulses. In contrast to “FM bats,” which have sharply tuned units distributed broadly throughout the range of their emitted pulses, “CW bats” have auditory systems tuned predominantly to a much smaller range corresponding to the restricted range of emitted frequencies. There is an “off”-response sharply tuned to the emitted CW frequencies. “On” response sensitivity is greater at just lower frequencies (CW component of less than 8-msec duration) or on either side of the “off”-frequency, with a null at that point (CW component of 10–20 msec or longer). The “off”-response, which is present at the level of the auditory nerve, apparently represents rebound excitation following non-neural suppression peripheral to the primary afferent fibers. It reflects a mechanism for increasing frequency discrimination and has the effect as well of enhancing response to the FM portion of the emitted pulses. High sensitivity to frequencies above that of the emitted CW, in “long CW” bats, is apparently tied to a use of the Doppler-shifted echo CW to detect and determine the angle and/or relative velocity of echo sources. CW bats show much less dramatic temporal resolution ability than FM bats, although it seems probable that the same mechanism of distance determination—measurement of elapsed time between emitted and returning FM sweeps—is used by both. CW bats, on the other hand, appear to have more units that are sharply sensitive to signal angle, showing strong binaural interaction. The consistent and conspicuous differences in neurophysiology in different species indicates not only that central analytic mechanisms are highly adaptable, but that these adaptations are important and closely tied to the types of orientation sounds used.

Rebound excitation (Off-responses) following non-neural suppression in the cochleas of echolocating bats

Auditory nerve evoked potential responses (N1) occur at the termination of tone pips as well as the onset in several species of bats that emit orientation sounds having a constant frequency component of several msec duration. 2. These “off”-responses are present only in a narrow band of frequencies around the emitted constant frequency. The “on”-response is most sensitive at slightly lower or higher frequencies and shows a relative null at the best frequency for the “off” (Fig. 2). 3. The “off”-response is similar in waveform to the “on” response, and follows the termination of a tone pip with a latency (0.9 ±0.1 msec) indistinguishable from that with which the “on”-response follows the onset of the signal (Fig. 1). 4. “Off”-responses are seen at signal durations as short as 0.5msec and are typically maximal with a signal duration of only 2 msec. Summation of “on” and “off” responses at durations shorter than 1.5 msec suggests that separate populations of units are contributing to the two responses. 5. Many single units at higher neural levels respond only to the “off ” of a signal of the frequency that produces an N1 “off ” evoked potential, and only to the onset of a slightly lower frequency that produces a maximal “on” evoked potential (Fig. 3). 6. A frequency that produces “off”-responses can either suppress or enhance the “on”-response to a lower frequency tone pip, depending on the relative intensity of the two sounds (Fig. 4). The suppression probably represents a form of two-tone inhibition comparable to that seen in other mammals. Similar enhancement has not been reported in other mammals, however. The “off”-response appears to represent rebound excitation of cells tuned to slightly lower frequencies than the suppressant tone. 7. The “off”-response is most sensitive to masking by frequencies that elicit maximal “on”-responses, well below those that produce a maximal “off”-response (Fig. 5). 8. It is concluded that the suppression is unlikely to be mediated by chemical neural inhibition. The properties of the “off” response appear compatible with a mechanism whereby vibration of a certain part of the basilar partition causes an electrical or mechanical bias that reduces spontaneous activity in a population of primary fibers innervating an adjacent part of the partition, possibly by reducing tonic hair cell transmitter release, with termination of this suppression resulting in synchonous resumption of spontaneous firing at normal or greater than normal rate. 9. The apparent function of this interaction in the constant frequency emitting bats is a peripheral sharpening of frequency resolution, and a “sensitization” of primary afferent pathways to the lower frequencies of the downward sweep at the end of the emitted pulses.

Respiration, Wing-Beat and Ultrasonic Pulse Emission in an Echo-Locating Bat

ABSTRACT During flight, echo-locating bats must adjust their pulmonary ventilation to satisfy a variety of requirements. The basic role of respiration, in gas exchange and in the maintenance of physiological homeostasis under exercise stress, must be achieved in a way that is compatible not only with the rhythmic contraction of powerful flightmuscles but also with the continual production of high-intensity vocalizations necessary for acoustic orientation. Moreover, the demand for accurate navigational information is often greatest during the stresses of rapid erratic flight. At such times bats continually emit brief, high-intensity orientation sounds at repetition rates varying from a few to many per second. Nothing is known regarding the relationship between respiration, sound production and wing movements during flight, or the extent to which changes in pulse-repetition rate affects respiration.

Coding and Processing in the Nervous Systems of FM-Signal-Producing Bats

For echolocation, certain insectivorous bats emit frequency-modulated (FM) sounds in which frequency always sweeps from high to low. Neurons in the cochlear nerve and nucleus respond to various types of sounds, such as pure tones, FM sounds, and noise bursts, and are therefore not specialized for the analysis of the bat's orientation sounds. In the inferior colliculus and auditory cortex, however, there exist neurons which selectively respond to FM sounds. For the excitation of these FM-specialized neurons, the direction, range, and speed of frequency sweep are important factors. The FM-specialized neurons always have a large inhibitory area and respond to FM sounds sweeping across this area. Although seemingly paradoxical, such properties are easily explained by neural network models. Bats emit short FM sounds when certain parts of the brain are stimulated with electric pulses. The temporal pattern of emission differs according to the location of the stimulating electrodes. Many neurons in the lateral lemniscus and inferior colliculus are excited by the bat's own outgoing sounds, in spite of the contraction of the middle ear muscles, synchronized with sound emission. The lateral lemniscal neurons appear to be specialized for echo detection because of the facilitation of their responses to a second tone pulse (artificial echo), while the inferior collicular neurons appear to be specialized for echo ranging because they exhibit various types of recovery cycles, including temporary recovery. [Supported by grant GB-13904 and PIB-2103 from the N.S.F. and grant HSAA 53837 from the P.H.S.]

Adaptations of the auditory nervous system for echolocation

Adaptations of the auditory nervous system for echolocation

The importance of atmospheric attenuation for the echolocation of bats (Chiroptera)

Atmospheric attenuation of high frequency sound in air limits the effective range of bats' echolocation. Attenuation is proportional to frequency squared, but also increases with absolute humidity, passes through a peak, and declines again. The humidity causing maximum attenuation is very low at 1 kHz, but shifts progressively to higher humidities at ultrasonic frequencies, and by 100 to 150 kHz the peak is located at humidity levels found in tropical rain forests. The adaptive advantages of various frequencies for an echolocating bat are thus humidity dependent. This may explain the use of 8 to 4 kHz orientation sounds by an unidentified insectivorous bat in the Amazon Valley.