Binaural Intensity Differences Research Articles

Psychoacoustics in the United Kingdom in the period 1900 to 1950

Much research in psychoacoustics during this period was conducted in the “Psychological Laboratory” in Cambridge (H. Bannister, A. F. Rawdon-Smith) and in Manchester (T. S. Littler, A. W. Ewing). Although Lord Rayleigh had published his influential duplex theory of sound localization in 1907, lingering doubts remained about whether interaural phase could truly be discriminated—rather, it had been proposed that somehow interaural phase differences were converted into interaural intensity differences. Bannister conducted a series of experiments to assess this idea and concluded that “binaural phase differences are appreciated in some manner which is distinct from the appreciation of binaural intensity differences.” Bannister also reported experiments on time-intensity trading in sound localization and reported that sometimes two images were perceived. Rawdon Smith and independently Littler and Ewing studied the phenomenon of auditory fatigue (temporary threshold shift, TTS). They showed that the maximum TTS produced by an intense fatiguing tone occurred for test tones with frequencies well above that of the fatiguing tone, a phenomenon that was repeatedly confirmed in later studies and that has only recently been explained. A common feature of the papers describing this work was a detailed description of the (often ingenious) apparatus that was developed to conduct the experiments.

Posttransient shifts in auditory lateralization

Presenting an intense (e.g., 80-dB [SPL]) "transient" (e.g., 50-msec) inducer to the ear reduces the loudness of subsequent signals at or near the frequency of the inducer. In this study, we ask whether similar inducers also affect lateralization. In two experiments, we asked how inducing tones presented to one ear (the exposed ear) affect judgments of the lateral position of subsequent target tones having various interaural intensity differences. In Experiment 1, inducers had the same frequency as the targets, and, as predicted, reduced the tendency to lateralize the targets to the exposed ear. In Experiment 2, the frequency of the inducers and the target differed (different critical bands), thereby eliminating the effect on lateralization. These results are consistent with the hypothesis that inducers temporarily reduce the magnitude of the representation of intensity signals in the frequency region around them and that this reduction occurs, at least partly, peripherally to the site at which binaural intensity differences are encoded. The results imply further that the reduction in loudness previously reported under similar stimulus conditions reflects a more general reduction of intensity-based information in hearing.

Evolution of mammalian hearing

The comparative study of mammals reveals systematic differences in their hearing abilities. First, the ability to hear above 10 kHz (which apparently evolved for using pinna cues to localize sound) varies inversely with the size of an animal’s head such that smaller animals usually have better high-frequency hearing than larger ones. Second, low-frequency hearing ability is bimodally distributed: Although most mammals can hear below 125 Hz, about one third do not hear much below 500 Hz and probably do not use temporal coding for pitch perception. Third, sound localization acuity varies directly with the width of the field of best vision, indicating that a major source of evolutionary pressure on hearing is to direct the eyes to the source of a sound. Finally, while some mammals do not use the binaural time-difference locus cue and a few others do not use the binaural intensity difference cue, it is likely that all mammals that possess pinnae use pinna locus cues to localize sound. Taken together, these findings suggest that some of the hearing abilities of extinct mammals may be inferred from fossil evidence bearing on their head size, eyes, and the presence of pinnae.

Auditory lateralization in bushcrickets: a new dichotic paradigm

Pair formation in the bushcricket Gampsocleis gratiosa is achieved through acoustic signalling by the male and phonotactic approaches of the female towards the calling song. On a walking belt in the free sound field, females tracked the position of the speaker broadcasting the male calling song with a remarkable precision, deviating by no more than 10 cm in either direction from the ideal course. Starting with stimulus angles of 6-10 degrees the females significantly turned to the correct side, and with stimulus angles greater than 25 degrees no incorrect turns were made. Using neurophysiological data on the directionality of the ear we calculated that with such stimulus angles the available binaural intensity difference is in the order of 1-2 dB. We developed a dichotic ear stimulation device for freely moving females with a cross-talk barrier of about 50 dB, which allowed to precisely apply small binaural intensity differences. In such a dichotic stimulation paradigm, females on average turned to the tronger stimulated side starting with a 1 dB difference between both ears. The significance of such a reliable lateralization behaviour with small interaural intensity differences for phonotactic behaviour under natural conditions is discussed.

Auditory receptive area in the cerebellar hemisphere is surrounded by somatosensory areas

We mapped the neuronal discharges in response to sound in the granule cell layer of the cerebellar hemisphere of the rat. An auditory receptive area was located in the lateral part of Crus IIb. The size of the auditory area was approx. 1 mm 2. It was surrounded by somatosensory receptive areas representing the regions in and around the mouth, particularly the lips, the incisors, and areas inside the mouth. Frequency selectivity of neurons in the auditory area was broad that it resembled the audiogram of the ear of the animal. The auditory responses were not particularly sensitive to binaural intensity differences. On the basis of the response properties of these neurons to sound and the receptive field properties of the adjacent somatosensory areas, we suggest that the function of the auditory area in the cerebellar hemisphere may be in the control of movements involved in vocalization.

Biophysical Aspects of Directional Hearing in the Tammar Wallaby, Macropus Eugenii

ABSTRACT The biophysical properties of the external ear of the Tammar wallaby, Macropus eugenii (Desmarest), have been investigated using probe microphones implanted in the ear canal. An acoustic axis of the pinna exists above 2kHz which is located close to the horizonal plane for natural ear positions, whereas azimuthal location of the acoustic axis is determined by pinna orientation on the head. The maximum on-axis acoustic pressure gain of the external ear reaches 25–30 dB for frequencies near 5 kHz. This results from pressure transformation by the horn-like pinna combined with resonance of the auditory meatus. The directionality of the pinna is similar to the sound diffraction properties of a circular aperture with an average radius based on the circumference of the pinna face. These properties determine the acceptance angle of the main lobe containing the acoustic axis and the spatial location of nulls. Large binaural intensity differences, exceeding 30dB, can be produced by the interaction of peaks and nulls between monaural directivity patterns, depending on the relative position of each pinna.

Localization of tones by horses: Use of binaural cues and the role of the superior olivary complex.

The ability of horses to use binaural time and intensity difference cues to localize sound was assessed in free-field localization tests by using pure tones. The animals were required to discriminate the locus of a single tone pip ranging in frequency from 250 Hz to 25 kHz emitted by loudspeakers located 30 degrees to the left and right of the animals' midline (60 degrees total separation). Three animals were tested with a two-choice procedure; 2 additional animals were tested with a conditioned avoidance procedure. All 5 animals were able to localize 250 Hz, 500 Hz, and 1 kHz but were completely unable to localize 2 kHz and above. Because the frequency of ambiguity for the binaural phase cue delta phi for horses in this test was calculated to be 1.5 kHz, these results indicate that horses can use binaural time differences in the form of delta phi but are unable to use binaural intensity differences. This finding was supported by an unconditioned orientation test involving 4 additional horses, which showed that horses correctly orient to a 500-Hz tone pip but not to an 8-kHz tone pip. Analysis of the superior olivary complex, the brain stem nucleus at which binaural interactions first take place, reveals that the lateral superior olive (LSO) is relatively small in the horse and lacks the laminar arrangement of bipolar cells characteristic of the LSO of most mammals that can use binaural delta I.

Sound localization and high-frequency hearing in horses

The ability of mammals to hear high-frequency sounds has often been attributed to the need to use binaural intensity differences (ΔI to localize sound [e.g., R. S. Heffner and H. E. Heffner, J. Comp. Physiol. Psychol. 96, 926–944 (1982)]. However, this explanation does not seem to account for the ability of horses to hear high-frequency sounds. Specifically, while horses are able to localize low-frequency tones, they are unable to localize tones above 1 kHz even though they can hear up to 35 kHz. These results suggest that while they can use interaural phase or time differences (Δt) to localize sound, they lack the ability to use the binaural ΔI cue. However, localization tests using filtered noise and sound sources located off to the side indicate that horses use pinna cues to localize sound and rely heavily on the high-frequency content of a sound in order to prevent front-back reversals. It appears, then, that horses have lost the ability to use binaural ΔI and rely primarily on binaural Δt and pinna cues in order to localize sound. Furthermore, since it seems unlikely that horses engage in ultrasonic communication, it apepars that they have evolved high-frequency hearing primarily in response to the need to use pinna cues to localize sound. [Work supported by NSF and NIH.]

Monaural and binaural intensity discrimination in normal and cochlear-impaired listeners.

Monaural and binaural intensity difference limens for 75-dB SPL pure tones were determined for 6 normal subjects and for 6 subjects with cochlear hearing loss. The magnitude of binaural masking level difference (BMLD) was also determined. Both normal and hearing-impaired subjects showed a 0.45-dB binaural advantage for intensity discrimination. In contrast, the BMLD was appreciably reduced with hearing impairment. Results are discussed in terms of specific effects of hearing loss on binaural hearing.

Binaural intensity discrimination data compared with left‐right‐averaged and with superior‐ear data

Binaural intensity difference thresholds were compared with monaural thresholds based upon left‐ and right‐ear means, and with the thresholds of the ear showing superior intensity difference sensitivity for the test frequency. Consistent with other recent finding [W. Jesteadt and C. C. Wier, J. Acoust. Soc. Am. 61, 1599–1603 (1977)] the ΔI/I ratios for binaural listening were lower than the averaged monaural ratios for frequencies 250, 1000, and 3000 Hz. The superior‐ear data, however, matched the binaural data well, and appeared to separate from left‐right‐averaged ΔI/I ratios. The suggestion is offered that the contribution of the less sensitive ear, in amplitude discrimination, may be negligible, at least for some varieties of stimuli.

Auditory lateralization in monkeys: an examination of two cues serving directional hearing.

In assigning binaural ongoing time differences (phase) as the cue for localization of low frequencies, and binaural intensity differences as the cue for localization of high frequencies, the duplex theory has successfully accounted for human directional hearing of tones. Sensitivity of monkeys to these cues was examined in two experiments. The dependencies on frequency of interaural intensity difference thresholds (lateralization experiment I) and time difference thresholds (lateralization experiment II) were determined behaviorally on three monkeys (M. nemestrina). The range of frequencies was from 125 Hz to 8 kHz in experiment I and from 250 Hz to 2 kHz in experiment II. The results indicate that the duplex theory is applicable to monkeys. However, monkeys are less sensitive than man to both binaural cues. The shortest time disparity monkeys discriminate is 42 microseconds at 1.5 kHz and the smallest intensity difference is 3.5 dB at 500 Hz. Good agreement between the present findings and localization measurements [C. H. Brown et al., J. Acoust. Soc. Am. 63, 1484-1492 (1978)] suggests: (a) that monkeys utilize time disparity cues through higher frequencies than man; and (b) that inaccurate localization by monkeys at high frequencies reflects decreasing sensitivity to interaural intensity cues.

The response properties of auditory neurones in the midbrain of the domestic fowl (Callus gallus) to monaural and binaural stimuli

Extracellular recordings were made from auditory units in the midbrain (NMLD) of the domestic fowl. Tonal stimuli delivered monaurally to either ear revealed that contralateral excitation predominated and was coupled with ipsilateral inhibition (EI, 46%), excitation (EE, 21%) or no effect (EO, 17%). Conventional V-shaped tuning curves were readily obtained from contralateral excitatory thresholds, however complex effects were noted such as, inhibitory sidebands and double sensitivity peaks. The distribution of best frequency (BF) thresholds ranged from 60 Hz-5.0 kHz, with the most sensitive region between 1.0–2.0 kHz. Analysis of cell types with BF revealed wide distributions for EI and EE cells, however EO cells had a distinct bias for BF's above 1.0 kHz. Binaural stimuli were used to study unit sensitivity to experimentally produced differences in interaural intensity and time (phase). EI cells generally showed good sensitivity for interaural intensity differences up to 20 dB, across a wide range of BF's. EE cells which demonstrated interaural time sensitivity did so over several milliseconds (maximum physical delay 100 μs). These effects appeared to be related to the period of the stimulus (below 800 Hz) and reflected a tendency for unit responses to be phase-locked. Considering the limits of head size and high frequency sensitivity in the domestic fowl, the present neural data suggests that binaural intensity differences may be the only viable cue during sound localization.

Sensory parameters of the ultrasonic binaural spatial sensor with particular reference to use by blind babies

The sensory parameters of the ultrasonic binaural spatial sensor for use by blind persons are discussed. These comprise (1) the distance cue coded in kHz/m which can be varied from 1 kHz/m for long (5m) range use to 10 kHz/m for short (0.5 m) range use; (2) the direction cue coded in binaural intensity difference in dB/degree and may be varied in practice from 0.2 to 1 dB/degree giving either a compressed expanded angular resolution; (3) the field of view which may be wide (120°) or narrow (10°). These parameters are discussed to show what choices are available in deciding on the parameters of an aid to be used by a blind baby. It is shown that the effect of motion causing Doppler shifts can be minimized in a baby's aid and greater resolution obtained by the appropriate choice of sensory parameters. It would seem that specialists, similar to opticians or audiologists, are needed in this field to make the choices and fit the sensory aid. Results of work with babies and young children will be discussed.

Responses of single cells in cat inferior colliculus to binaural click stimuli: Combinations of intensity levels, time differences and intensity differences

Click stimuli were used to study the responses of single units in the inferior colliculus to interaural time and intensity differences, using values for these differences that approximate those estimated to be produced by positioning of an actual source of sound. Effects of sound intensity level on interaural time and intensity responses were studied, as well as effects of binaural intensity difference on interaural time responses. Spike counts were plotted as a function of variation in interaural time or intensity. Four populations of neurons were described on the basis of these spike count functions; units sensitive to (1) interaural intensity differences, (2) interaural time differences, (3) neither interaural time nor intensity, or (4) both interaural time and intensity. With manipulation of intensity a unit may alter its properties so as to appear to shift from one of these populations to another. Four general types of interaural time sensitive functions and three general types of interaural intensity functions were described. It was proposed that complex types of unit interaural time codings suggest a convergence of lower order units, with simple codings, onto higher order complex units. Time-intensity trading was demonstrated by shifts of interaural time sensitivity functions along the time axis with binaural intensity manipulation. Other effects of both binaural intensity difference and intensity level were demonstrated. These results suggest dynamic and differential unit properties concerned with the representation of auditory space in the central nervous system.