Donald H. Mershon

Intensity and reverberation as factors in the auditory perception of egocentric distance

Both auditory intensity and reverberation have previously been shown to be sufficient to produce systematically varying judgments of perceived distance when several values of the variable are presented repeatedly to the same observer. Such studies do not, however, indicate clearly whether these cues are functioning in an absolute or in a relative manner. An absolute cue to auditory distance would require that two groups presented with different values of the variable in question should report different values of perceived distance. Two experiments are reported in which intensity variation and reverberation are examined. The results showed that auditory intensity differences over a range of 20 dB did not serve as an absolute cue to auditory distance, but could serve as a strong cue to changes in such distance. A comparison of data obtained in a normally reverberatory setting (Experiment 1) and an anechoic chamber (Experiment 2) indicated that the state of reverberation could serve as an absolute cue, with greater reverberation being associated with greater perceived distances. Some of the results were discussed in terms of the possibility that the specific distance tendency (a concept developed to handle some phenomena in visual space perception) might have applicability to the study of auditory perceived distance as well.

Absolute and relative cues for the auditory perception of egocentric distance.

Three experiments were performed to examine the reverberation cue to egocentric auditory distance and to determine the extent to which such a cue could provide ‘absolute’, as contrasted with ‘relative’, information about distance. In experiment 1 independent groups of blindfolded observers (200 altogether) were presented with broadband noise from a speaker at one of five different distances (0·55 to 8 m) in a normal hard-walled room. Half of each group of observers were presented with the sound at 0 deg azimuth, followed (after a delay) by the identical sound at 90 deg azimuth. The order of presentation was reversed for the remaining observers. Perceived distance varied significantly as a function of the physical distance to the speaker, even for the first presentations. The change in the binaural information between the 0 deg and 90 deg presentations did not significantly modify the results. For both orientations, near distances were overestimated and far distances were underestimated. Experiments 2 and 3 were designed to evaluate how much prior auditory exposure to the laboratory environment was necessary. A 200 Hz square-wave signal was presented from one of three distances (1, 2, or 6 m) to observers who had either minimal room information or an exposure which included talking within the room. Perceived distance varied significantly with physical distance regardless of exposure condition.

Effects of Room Reflectance and Background Noise on Perceived Auditory Distance

Perceptions of egocentric auditory distance were investigated within an environment for which the reverberation time could be systematically varied without changes in the size or shape of the room. Two levels of wide-band background noise, differing by 20 dB, were used as a masking stimulus. Target sounds were presented from distances between 0.75 and 6.0 m and verbal reports of distance were collected from 288 listeners in two separate experiments. Changes in physical distance produced variation in reported distance in each configuration. Reported distance was generally proportional to real distance, but considerably underestimated when room reflectance was low. When room reflectance was high (T60 ≈ 1.7 s for the range of frequencies used), initial reports of distance were often overestimates; upon repeated presentation, judgments in the high reflectance room became more nearly veridical. The effect of increasing the background noise level was to decrease the perceived distance. These findings are in accord with expectations based upon the importance of reverberation cue(s) to distance and upon previous analyses from this laboratory.

Spectral content as a cue to perceived auditory distance

Changes in the spectral content of wide-band auditory stimuli have been repeatedly implicated as a possible cue to the distance of a sound source. Few of the previous studies of this factor, however, have considered whether the cue provided by spectral content serves as an absolute or a relative cue. That is, can differences in spectral content indicate systematic differences in distance even on their first presentation to a listener, or must the listener be able to compare sounds with one another in order to perceive some change in their distances? An attempt to answer this question and simultaneously to evaluate the possibly confounding influence of changes in the sound level and/or the loudness of the stimuli are described in this paper. The results indicate that a decrease in high-frequency content (as might physically be produced by passage through a greater amount of air) can lead to increases in perceived auditory distance, but only when compared with similar sounds having a somewhat different high-frequency content, ie spectral information can serve as a relative cue for auditory distance, independent of changes in overall sound level.

Perceived Loudness and Visually-Determined Auditory Distance

Three experiments were conducted to determine whether variations in the perceived distance to a test sound could influence its loudness in the absence of physical changes in sound-level. The phenomenon of visual capture provided the means for manipulating apparent distance. A ‘dummy’ loudspeaker was used to vary the apparent source location of a short noise stimulus while the true source of this sound remained fixed (and hidden) with respect to the observer. Sound-levels from 40 to 75 dB(A) were presented to independent groups of observers in either anechoic or semi-reverberant acoustical environments. In general, reported loudness increased with perceived distance. This finding has implications for conceptualizing the phenomenon of loudness constancy.

Failure of familiar size to determine a metric for visually perceived distance

The use of knowledge of the familiar sizes of objects in determining the apparent distances of those objects is known as the familiar size cue to distance. If effective, this cue might be one of the factors responsible for supplying the metric (scalar) characteristics of perceptions of spatial extent within a visual display in which other information concerning scalar extents has been reduced to a minimum. Two groups of observers were presented with realistic objects of the same angular, but different assumed, sizes presented in such a cue-restricted display. Perceptions of size and distance within the display did not differ significantly as a function of the type of object initially presented. This result was consistent with the notion that scalar perceptions under these conditions probably are determined by a factor known as the specific distance tendency, rather than by the experiential factor of familiar size.

Time‐of‐day variations in oculomotor function: 1. Tonic accommodation and tonic vergence

Time‐of‐day variations were studied both in tonic accomodation and in tonic vergence. Thirty‐six male volunteers were measured once in the morning and at one other time on the same day. The mean baseline value of tonic accomodation was 2.02 D. Time‐of‐day variations were found, with a late evening measure being + 0.6 D greater than the morning value. No independent time‐of‐day fluctuations were found in tonic vergence. Post‐hoc analysis, however, showed that tonic vergence tended to shift in the same direction as an individuals tonic accommodation during the afternoon. During the evening, tonic vergences tended to shift oppositely to tonic accommodation.

Auditory Distance Judgments in Noise, with and without Hearing Protection

Previous studies of the effects of hearing protective devices (HPDs) on auditory localization have all been concerned with directional judgments. The present study examined the influence of such devices upon judgments of distance under conditions of noise. Blindfolded observers were required to verbally report the apparent distance to a target sound under one of three conditions: quiet (60 dBA), noise (90 dBA) or noise (90 dBA) while wearing inserted foam plugs. Judgments in quiet were found to be reasonably correct. The presence of high-level noise, however, decreased the judged distances significantly. The use of the HPD neither improved these latter judgments, nor did it produce any further error. The findings are fully consistent with present understanding of the bases for auditory distance perception.

Directional localization in high ambient noise with and without the use of hearing protectors

Personal hearing protective devices (HPDs) are becoming increasingly common in settings where levels of noise above 85 dB(A) represent a potential threat to hearing. Several previous studies have indicated that the use of such devices adversely affects the localization of sounds, but none of these experiments employed very precise measures of location discriminability. The major study in the present report examined just such fine discrimination, using a forced-choice task, and compared the effects of three different HPDS. Localization signals were 1.5 s bursts of noise (≥ 1 kHz). Both of the described experiments were conducted in acoustically complex, reverberant environments, to simulate real working conditions. Results indicated that greater attenuation (especially at higher frequencies) was the primary factor associated with poor performance on the fine discrimination task.

Organizational factors and the perception of motion in depth

When two stationary, stereoscopically separated targets are viewed in a completely dark surround, and no cues concerning their egocentric distances from the observer are salient, the farther target tends to be seen at the same distance it would have assumed if it were by itself. The nearer target is seen as being closer than it would have been if seen alone. The present studies extend this previous finding (now termed thefar-anchor effect) into the domain of targets that move in stereoscopic space. Observers viewed two small illuminated targets, which began at either the same or different stereoscopic distances. One of the targets was moved in depth and the observers identified the target that appeared to move. Conditions varied according to the initial depth location of the moving target. Significantly more correct responses were reported when the nearer target moved than when the farther one moved, consistent with the hypothesis that the perception of motion in depth is affected by the aforementioned perceptual anchoring effect of the farther target.