Robert A. Butler

Factors That Influence the Localization of Sound in the Vertical Plane

The ability of listeners to locate sound in the vertical plane was investigated. The results showed that for auditory stimuli to be located accurately (1) the stimulus must be complex, (2) it must include frequencies above 7000 cps, and, (3) the pinna must be present.

Effect of Changes in Stimulus Frequency and Intensity on Habituation of the Human Vertex Potential

Interposed between periodic presentations of a 1000‐Hz tone were stimuli of different frequencies and intensities. The amplitude of the vertex (V) potential to the 1000‐Hz tone became larger as the frequency of the intervening stimuli departed progressively from 1000 Hz. It also became larger when the sensation level of the 1000‐Hz tone was raised progressively above that of the intervening stimuli. The notion that changes in stimulus frequency and increases in stimulus intensity activate new neural units was adopted to explain the data. It was asserted that habituation of the V potential is greatest when the same neural units are repeatedly activated and least when new neural units are brought into play.

Spectral cues utilized in the localization of sound in the median sagittal plane

A series of experiments was carried out to further elucidate the role of spectral cues in locating sounds in the median sagittal plane (MSP). Broadband noise bursts, generated at ±30°, ±15°, and O° re aural axis, were recorded via microphones placed in the external ear canals of 8 Ss. When these recorded sounds were played back dichotically through headphones, they were perceived as originating from the loudspeakers, not the headphones. In fact, Ss could identify that loudspeaker which originally generated the sound nearly as accurately as they could when listening under free‐field conditions. Analysis of the spectra of these recorded sounds revealed a notch in the frequency response curves which migrated toward the lower frequencies as the sound source was moved from above to below the aural axis. This feature of the spectrum may well be important for accuracy in locating sounds emanating from the frontal segment of the MSP. Four Ss were given additional tests to find out whether they could locate sounds...

Localization of Tonal Stimuli in the Vertical Plane

Listeners, when requested to judge the location of tone bursts on the vertical plane, tended to place the stimuli on a vertical scale in accordance with their respective pitch. Higher‐pitched sounds were perceived as originating above lower‐pitched sounds. This phenomenon was also observed in congenitally blind persons and in young children who presumably were unaware of the use of the words high and low in describing differently pitched sounds. In additional tests, visual cues were found to influence the range of the scale within which listeners perceived the sound sources and they could even bias the location judgments of high‐pitched tone bursts despite contrary binaural cues. But the main implication of the data is that tonal stimuli have intrinsic spatial characteristics, which result in the perception of frequencies with shorter periods as being higher in space than those with longer periods.

An Investigation of the Human Cortical Evoked Potential Under Conditions Of Monaural and Binaural Stimulation

The human cortical potential evoked by a 1000-Hz tone was recorded under conditions of binaural and monaural stimulation which covered a wide range of stimulus intensities. The results showed that (1) the curve for the amplitude as well as for the area confined by the potential increased with increases in stimulus intensity. At higher stimulus intensities, these curves leveled off. (2) Bilateral stimulation generated larger cortical potentials than did monaural stimulation. (3) Latency of the potential decreased with increases in stimulus intensity. When the monaural stimulus was delivered to the ear contralateral to the active scalp electrode, the various components of the potential showed consistently a shorter latency than when the monaural stimulus was presented ipsi-laterally.This research was supported by a grant from the National Institute of Neurological Diseases and Blindness, USA., and the Deutsche Forschungsgemeinschaft.

The bandwidth effect on monaural and binaural localization.

Listeners located, monaurally and binaurally, an 8.0 kHz centered noise burst whose bandwidth was set at 2.0, 4.0, 6.0 and 8.0 kHz. Loudspeakers, placed 7.5 degrees apart, covered an arc extending from 15 degrees to 165 degrees to the left of midline. Listeners reported the number of that loudspeaker from which the noise bursts appeared. A significant reduction in localization errors was associated with increments in bandwidth and for binaural localization, this was attained largely through fewer instances of confusing sounds from the front with those from the rear. While overall, binaural accuracy exceeded monaural accuracy when sounds came from the front or rear, no appreciable differences between the two listening conditions were noted when the sounds came from the middle section of the arc. Only when broadband noise was employed, as it was in a supplementary set of observations, did binaural localization uniformly surpass monaural localization in accuracy - a finding attributed to the introduction of low frequencies which resulted in the addition of interaural ongoing phase differences to the constellation of localization cues.

The influence of spatial separation of sound sources on the auditory evoked response.

Abstract The N1P2 component of the auditory evoked response in man was measured under conditions in which sounds alternated between loudspeakers located in the horizontal plane (Experiment 1) and in the vertical plane (Experiment 2). When compared with response amplitudes to sounds originating always at the same location, significant increments were found when sounds alternated between loudspeakers separated by 90 deg. in either the horizontal or vertical plane. The data were interpreted in terms of a place theory of cortical organization for auditory space perception.

Binaural and Monaural Localization of Sound in Two-Dimensional Space

Two experiments were conducted. In experiment 1, part 1, binaural and monaural localization of sounds originating in the left hemifield was investigated. 104 loudspeakers were arranged in a 13×8 matrix with 15° separating adjacent loudspeakers in each column and in each row. In the horizontal plane (HP), the loudspeakers extended from 0° to 180° in the vertical plane (VP), they extended from −45° to 60° with respect to the interaural axis. Findings of special interest were: (i) binaural listeners identified the VP coordinate of the sound source more accurately than did monaural listeners, and (ii) monaural listeners identified the VP coordinate of the sound source more accurately than its HP coordinate. In part 2, it was found that foreknowledge of the HP coordinate of the sound source aided monaural listeners in identifying its VP coordinate, but the converse did not hold. In experiment 2, part 1, localization performances were evaluated when the sound originated from consecutive 45° segments of the HP, with the VP segments extending from −22.5° to 22.5°. Part 2 consisted of measuring, on the same subjects, head-related transfer functions by means of a miniature microphone placed at the entrance of their external ear canal. From these data, the ‘covert’ peaks (defined and illustrated in text) of the sound spectrum were extracted. This spectral cue was advanced to explain why monaural listeners in this study as well as in other studies performed better when locating VP-positioned sounds than when locating HP-positioned sounds. It is not claimed that there is inherent advantage for localizing sound in the VP; rather, monaural localization proficiency, whether in the VP or HP, depends on the availability of covert peaks which, in turn, rests on the spatial arrangement of the sound sources.

Responses of Cochlear Potentials to Changes in Hydrostatic Pressure

Sustained and transient changes in hydrostatic pressure were transmitted directly to the perilymphatic space of guinea pigs. EP, as well as SP, CM, and AP, was measured before, during, and after pressure changes. Pressure applied to scala vestibuli led to an increase EP and SP, while pressure applied to scala tympani resulted in a decrease of these potentials. Comparable data were also obtained the anoxic cochlea. CM and AP failed to change when moderate amounts of pressure were applied. In guinea pigs whose organ of Corti was largely degenerated EP was normal, but it did not increase when pressure was applied to scala vestibuli, nor did it consistently decrease during pressure applications to scala tympani. These data were interpreted to mean that changes in EP normally occurring response to changes in hydrostatic pressure reflect variations in the output of dc generators located on the basilar membrane.

An analysis of the monaural displacement of sound in space

Twelve listeners, rendered monaural, were tested on their ability to localize two spectrally different stimuli presented 10° apart in the left hemifield (pretraining session). Both sounds were perceived to be displaced from the midline. Next, listeners were trained to localize one of the sounds (training sessions) and then tested again on both (posttraining sessions). Training temporarily reduced the extent to which listeners displaced the sound from the midline. No transfer of training, however, was observed when listeners localized the other noise band, which differed in its spectral content. It is proposed that no “natural” spectral cues are available for localizing, monaurally, sound originating on or near the midline, but that cues for these positions can be acquired through training. Head-related transfer functions recorded from the pinna and generated by differently centered narrow bands of noise were obtained, along with additional localization data, in an attempt to support the concept of a natural cue for monaurally localizing sound in the horizontal plane. By and large, the results were consonant with this concept.