Search

Seemingly, there should be a close relationship between spatial release from masking and sound localization, but this is not always the case. For example, in binaural detection, randomizing the spatial parameters of the target or masker from trial to trial has little impact on threshold. In contrast, Simpson (Ph.D. dissertation, 2011) found that left/right localization judgments for a 60-ms target masked by a simultaneous 60-ms noise were considerably less accurate (equivalent to 10-dB reduction in SNR) when the location of the masker varied randomly from trial to trial than when the masker location was fixed. However, when a forward masker fringe was added, so that the noise was turned on 500 ms before the target, the impact of location variability was very small (about 1 dB, comparable to the detection literature). To determine if the presence of masker fringe could have limited the impact of spatial variability in previous detection experiments, the current study examines the effect of masker fringe and both target and masker spatial variability on detectability in conditions comparable to those of Simpson. The results will be compared to previous findings and models reported in the binaural detection and sound localization literature.

Localization accuracy for a target presented in a simultaneous masker, whose location varies randomly from trial to trial, improves when a preview of the masker location is provided (by playing a sound from that location) prior to the target + masker interval (i.e., a pre-trial cue) [B. Simpson, Ph.D. dissertation (Wright State University, 2011)]. One explanation is that knowing the masker location allows a listener to establish a “spatial attention filter” at the masker location. The present study compares the effect of such a pre-trial cue to the case in which the cue comes after the target + masker interval (post-trial cueing). That is, the cue is presented either 500 ms prior to the onset of a 60-ms, 100-Hz click-train target embedded in a 60-ms broadband masker, or 500 ms subsequent to the offset of the target + masker stimulus. The data indicate that both cue types lead to similar improvements in performance over the no-cue condition, with the greatest improvement from cueing (~6 dB) seen for localization in the left/right dimension. While these data are roughly consistent with previous results, they cannot be explained by a simple spatial filter hypothesis.

The Virtual Environment Research, Interactive Technology, And Simulation (VERITAS) facility is designed to support basic and applied research in complex multisensory environments including: 3D audio displays for improving safety in general aviation flight operations, advanced audio displays for improving performance in combat search and rescue missions, basic and applied research on multisensory (audio/visual/haptic) interactions, and displays for instrument landing systems. The central component of the facility is a CAVE®, a room-sized virtual environment presentation system. Several of our multi-user distributed experiments also utilized a virtual environment display system in the Appenzeller Visualization Laboratory (AVL). The facilities include a set of commercial-off-the-shelf (COTS) software solutions that work together with custom applications developed by our team to create immersive virtual environments. To facilitate software development and coherence, we created the Integrated Virtual Platform (IVP) as a flexible, expandable, and portable software platform for human performance experimentation. The experiments, experiences, and results in the VERITAS facility presented here, provide an example of a VE platform being actively used for human performance experimentation. The success of the research we have completed suggests that immersive virtual environments are useful tools in investigating complex human performance tasks.

Many of today’s military operations take place in large urban environments, which present unique challenges due to limited line of sight and increased concealment for enemy forces. This experiment evaluated the potential of a 3D audio display to aid threat detection and localization in a complex visual environment. Subjects rode as part of a convoy through a simulated city where they encountered snipers surrounded by distracting personnel. We compared 3D audio cues to verbal descriptions of the sniper’s location and to simple audio warnings of the presence of a sniper. Consistent with past research, subjects located the sniper more quickly in the 3D audio condition compared to both the semantic description and simple warning conditions. The 3D audio display was faster than the other displays across all azimuths but, in contrast to previous findings, the advantage did not increase consistently with increasing azimuth.

Navigating in any environment can be a very tedious task, and it becomes considerably more difficult when the environment is unfamiliar or when it contains threats that could possibly jeopardize the success of the mission. Often in difficult navigation environments there are external sensors present in the area that could provide critical information to the ground operator. The challenge is to find a way to transmit this information to the ground operator in an intuitive, timely, and unambiguous manner. In this study, we explore a technique called “audio annotation” where the sensor information is transmitted to a remote observer who processes it and relays it verbally to an operator on the ground. Spatial information can be conveyed intuitively by using a spatial audio display to project the apparent location of the remote observer's voice to an arbitrary location relative to the ground operator. The current study compared the “audio annotation” technique to standard monaural communications in a task that required a remote observer with a high-level view of the environment to assist a ground operator to avoid threats while locating a downed pilot in an urban environment. The overall performance in the audio annotation condition was found to be superior to the standard monaural condition.

Visual and audio navigations aids were compared in a virtual environment that depicted an urban combat search and rescue mission (CSAR). The participants' task was to rapidly move through a virtual maze depicted in a CAVE° to find a downed pilot, while dealing with automated hostile and friendly characters. The visual and audio displays were designed to present comparable information, which in separate conditions could be a simple realtime indication of the bearing to the pilot or to intermediate waypoints along the way. Auditory displays led to faster response times than visual displays (p = .011) and the waypoint display led to faster response times than the simple bearing display (p = .002). The results are considered in the context of the target CSAR application.

The effects of variable feedback delays on movement time were examined in a three-dimensional (3D) virtual environment. The participants' task was to use a 3D controller to position a cursor in targets as they appeared in a cubic workspace. Both the mean and standard deviation of the delay between the movement of the controller and the displayed position of the cursor were manipulated. In addition, the size of the targets and the distance between targets were varied. The results suggested that movement times are much more strongly affected by mean delay than by delay variability and that the effect of both variables is greatest during the closed-loop component of the movement. The results are discussed in relation to buffering strategies for reducing delay variability, Fitts' law, and other descriptions of aimed movements.

A task designed to estimate the impact of communication channel performance on individual and collaborative human performance in distributed, shared, virtual environments was evaluated. The 'Fitts-like' task required participants to press buttons as fast as they could in a three-dimensional virtual workspace. In addition, in the collaborative task the 2 participants were required to press the buttons simultaneously. The size of the buttons, the distance between buttons, and the delay between movement and visual feedback were manipulated. The results were well described by Fitts' law. Movement Time increased with increases in Index of Difficulty and delay. The Index of Performance decreased with delay and was lower in the collaborative task. Overall, orderly changes in performance were observed with manipulation of task parameters, suggesting that future studies to examine more complicated manipulations of network performance and multisensory displays are appropriate.

Most current display systems in general aviation (GA) environments employ, at best, relatively simple audio displays that do not fully exploit a pilot's auditory processing capabilities. Spatial audio displays, however, take advantage of the spatial processing capabilities of the auditory system and have the ability to provide, in an intuitive manner, comprehensive information about the status of an aircraft to the pilot. This paper describes a study conducted in order to assess the utility of spatial audio as (1) a navigation aid, and (2) an attitude indicator in an actual flight environment. Performance was measured in tasks requiring pilots to fly in the direction of a spatial audio “navigation beacon” and use an auditory artificial horizon display to detect changes in attitude and maintain straight and level flight when no visual cues were available. The results indicate that spatial audio displays can effectively be used by pilots for both navigation and attitude monitoring, and thus may be a valuable tool in supporting pilot situation awareness and improving overall safety in GA environments.

Modern Traffic Advisory Systems (TAS) can increase flight safety by providing pilots with real-time information about the locations of nearby aircraft. However, most current collision avoidance systems rely on non-intuitive visual and audio displays that may not allow pilots to take full advantage of this information. In this experiment, we compared the response times required for subjects participating in a fully-immersive simulated flight task to visually acquire and identify nearby targets under four different simulated TAS display conditions: 1) no display; 2) a visual display combined with a non-spatialized warning sound; 3) a visual display combined with a clock-coordinate speech signal; and 4) a visual display combined with a spatialized auditory warning sound. The results show that response times varied in an orderly fashion as a function of display condition, with the slowest times occurring in the no display condition and the fastest times occurring in the 3D audio display condition, where they were roughly 25% faster than those without the 3D audio cues.

This study investigated binaural detection of tonal targets (500 Hz) using sets of individual masker waveforms with two different bandwidths. Previous studies of binaural detection with wideband noise maskers show that responses to individual noise waveforms are correlated between diotic (N0S0) and dichotic (N0S(pi)) conditions [Gilkey et al., J. Acoust. Soc. Am. 78, 1207-1219 (1985)]; however, results for narrowband maskers are not correlated across interaural configurations [Isabelle and Colburn, J. Acoust. Soc. Am. 89, 352-359 (1991)]. This study was designed to allow direct comparison, in detail, of responses across bandwidths and interaural configurations. Subjects were tested on a binaural detection task using both narrowband (100-Hz bandwidth) and wideband (100 Hz to 3 kHz) noise maskers that had identical spectral components in the 100-Hz frequency band surrounding the tone frequency. The results of this study were consistent with the previous studies: N0S0 and N0S(pi) responses were more strongly correlated for wideband maskers than for narrowband maskers. Differences in the results for these two bandwidths suggest that binaural detection is not determined solely by the masker spectrum within the critical band centered on the target frequency, but rather that remote frequencies must be included in the analysis and modeling of binaural detection with wideband maskers. Results across the set of individual noises obtained with the fixed-level testing were comparable to those obtained with a tracking procedure which was similar to the procedure used in a companion study of rabbit subjects [Zheng et al., J. Acoust. Soc. Am. 111, 346-356 (2002)].

Contents: Part I:Sound Localization. F.L. Wightman, D.J. Kistler, Factors Affecting the Relative Salience of Sound Localization Cues. E.A.G. Shaw, Acoustical Features of the Human External Ear. R.O. Duda, Elevation Dependence of the Interaural Transfer Function. J.C. Middlebrooks, Spectral Shape Cues for Sound Localization. R.A. Butler, Spatial Referents of Stimulus Frequencies: Their Role in Sound Localization. Part II:Lateralization and Binaural Masking. L.R. Bernstein, Detection and Discrimination of Interaural Disparities: Modern Earphone-based Studies. T.N. Buell, C. Trahiotis, Recent Experiments Concerning the Relative Potency and Interaction of Interaural Disparities. R.H. Dye, Jr., The Relative Contributions of Targets and Distractors in Judgments of Laterality Based on Interaural Differences of Level. A. Kohlrausch, R. Fassel, Binaural Masking Level Differences in Nonsimultaneous Masking. Part III:Echoes, Precedence, and Depth. W.M. Hartmann, Listening in a Room and the Precedence Effect. E.R. Hafter, Binaural Adaptation and the Effectiveness of a Stimulus Beyond Its Onset. R.K. Clifton, R.L. Freyman, The Precedence Effect: Beyond Echo Suppression. D.H. Mershon, Phenomenal Geometry and the Measurement of Perceived Auditory Distance. Part IV:Motion Perception. D.R. Perrott, T.Z. Strybel, Some Observations Regarding Motion-Without-Direction. D.W. Grantham, Auditory Motion Perception: Snapshots Re-visited. K. Saberi, E.R. Hafter, Experiments on Auditory Motion Discrimination. Part V:Sound Source Segregation and Free-Field Masking. W.A. Yost, The Cocktail Party Problem: Forty Years Later. M.D. Good, R.H. Gilkey, J.M. Ball, The Relation Between Detection in Noise and Localization in Noise in the Free Field. T.J. Doll, T.E. Hanna, Directional Cuing Effects in Auditory Recognition. Part VI:Physiology of Spatial Hearing. S. Kuwada, R. Batra, D.C. Fitzpatrick, Neural Processing of Binaural Temporal Cues. T.C.T. Yin, P.X. Joris, P.H. Smith, J.C.K. Chan, Neuronal Processing for Coding Interaural Time Disparities. J.F. Brugge, R.A. Reale, J.E. Hind, Auditory Cortex and Spatial Hearing. E.D. Young, J.J. Rice, G.A. Spirou, I. Nelken, R.A. Conley, Head-related Transfer Functions in Cat: Neural Representation and the Effects of Pinna Movement. Part VII:Models of Spatial Hearing. R.M. Stern, C. Trahiotis, Models of Binaural Perception. H.S. Colburn, S.K. Isabelle, D.J. Tollin, Modeling Binaural Detection Performance for Individual Masker Waveforms. J.A. Janko, T.R. Anderson, R.H. Gilkey, Using Neural Networks to Evaluate the Viability of Monaural and Interaural Cues for Sound Localization. Part VIII:Development of Spatial Hearing. R.Y. Litovsky, D.H. Ashmead, Development of Binaural and Spatial Hearing in Infants and Children. Part IX:Applications. J. Blauert, An Introduction to Binaural Technology. B. Shinn-Cunningham, H. Lehnert, G. Kramer, E. Wenzel, N. Durlach, Auditory Displays. M.D. Burkhard, Binaural Measurements and Applications. R.L. McKinley, M.A. Ericson, Flight Demonstration of a 3-D Auditory Display. M.A. Ericson, R.L. McKinley, The Intelligibility of Multiple Talkers Separated Spatially in Noise. J. Koehnke, J. Besing, Binaural Performance in Listeners with Impaired Hearing: Aided and Unaided Results. B. Kollmeier, Signal Processing for Hearing Aids Employing Binaural Cues.

The sound localization ability of human observers has been frequently examined in quiet environments, but there have been relatively few studies that have considered the effect of noise on sound localization. In this study, three subjects judged the perceived direction of broadband click-train signal in the quiet and in the presence of a broadband noise at nine signal-to-noise ratios, which varied over a 23 dB range. The signal could originate from any of 239 spatial locations that completely surrounded the subjects in azimuth 360 degrees) and ranged from -45 degrees to (+)90 degrees in elevation; the masker (when present) was always located directly in front of the subjects at 0 degrees azimuth and 0 degree elevation. The subjects indicated the perceived direction of the signal by pointing at a 20-cm-diam spherical model of auditory space. As the signal-to-noise ratio was lowered, the accuracy of localization judgments decreased nearly monotonically. However, the accuracy of judgments relative to the median plane (i.e., the left/right dimension) was less strongly influenced by the presence of noise than was the accuracy of judgments relative to the horizontal plane (i.e., the up/down dimension). The accuracy of judgments relative to the frontal plane (i.e., the front/back dimension) was most strongly influenced by noise.

A technique is described for rapidly collecting responses in auditory-localization experiments. Subjects indicate the perceived direction of the sound by pointing at a 20-cm-diam spherical model. In Experiment 1, the subjects judged the direction of a broadband signal, which could originate from any of 239 directions ranging through 360° of azimuth and between −45° and +90° of elevation. Using this technique, the subjects responded 2–8 times more rapidly than previous subjects who used either a verbal-reporting technique or a head-pointing technique. Localization judgments were as accurate as they had been with verbal reports, but were not as accurate as judgments collected using the head-pointing technique. In Experiment 2, the signal was turned off and the experimenter read the spherical coordinates of the signal location to the subjects. The subjects pointed to these coordinates more accurately than they had judged the direction of the sounds in Experiment 1, suggesting that the response technique had not been the limiting factor in that experiment. Circumstances relevant to the choice of response techniques for auditory-localization experiments are discussed.

Signal-in-noise detection improves when the masker duration extends beyond that of the signal (masker fringe) relative to the case in which the signal and masker are pulsed on/off simultaneously. This improvement has been attributed to the fact that the fringe provides a baseline set of stimulus parameters that serves as a context against which a signal may be detected. Conversely, when the fringe parameters are inconsistent with those of the masker, signal detectability may be reduced. In this study, the impact of masker fringe on sound localization is examined. The results demonstrate the importance of stimulus parameters prior to and subsequent to the portion of the stimulus containing the signal for sound localization.

Objectives: We evaluate visual search performance in both static (nonmoving) and dynamic (moving) search environments with and without spatial (3-D) auditory cues to target location. Additionally, the effects of target trajectory, target location, and practice are assessed. Background: Previous research on aurally aided visual search has shown a significant reduction in response times when 3-D auditory cues are displayed, relative to unaided search. However, the vast majority of this research has examined only searches for static targets in static visual environments. The present experiment was conducted to examine the effect of dynamic stimuli upon aurally aided visual search performance. Method: The 8 participants conducted repeated searches for a single visual target hidden among 15 distracting stimuli. The four main conditions of the experiment consisted of the four possible combinations of 3-D auditory cues (present or absent) and search environment (static or dynamic). Results: The auditory cues were comparably effective at reducing search times in dynamic environments (—25%) as in static environments (—22%). Audio cues helped all participants. The cues were most beneficial when the target appeared at large eccentricities and on the horizontal plane. After a brief initial exposure to 3-D audio, no training or practice effects with 3-D audio were found. Conclusion: We conclude that 3-D audio is as beneficial in environments comprising moving stimuli as in those comprising static stimuli. Application: Operators in dynamic environments, such as aircraft cockpits, ground vehicles, and command-and-control centers, could benefit greatly from 3-D auditory technology when searching their environments for visual targets or other time-critical information.

The development of optimal three-dimensional auditory displays requires a more complete understanding of the interactions among spatially separated sounds. Free-field masking was investigated as a function of the spatial separation between signal and masker sounds within the horizontal, frontal, and median planes. The detectability of filtered pulse trains in the presence of noise maskers was measured using a cued, two-alternative, forced-choice, adaptive staircase procedure. Signal and masker combinations in low (below 2.3 kHz), middle (1.0–8.5 kHz), and high (above 3.5 kHz) frequency regions were examined. As the sound sources were separated within the horizontal plane, signal detectability increased dramatically. Similar improvement in detectability was observed within the frontal plane. As suggested by traditional binaural models, interaural time cues and interaural intensity cues are likely to play a major role in mediating masking release in both the horizontal and frontal planes. Because no interaural cues exist for stimuli presented within the median plane, traditional models would not predict a release from masking when the stimuli are separated within this plane. However, with high frequency signals, masking release similar to that observed in the horizontal and frontal planes could be observed in the median plane. The current literature suggests that sound localization in the median plane may depend on direction-specific spectral cues that are introduced by the pinna at high frequencies. The masking release observed here may also depend on these “pinna cues.”

Subject responses were measured for individual narrow-band reproducible stimuli in a low-frequency tone-in-noise detection task. Both N0S0 and N0Spi conditions were examined. The goal of the experiment was to determine the relative importance of envelope and fine-structure cues. Therefore, chimeric stimuli were generated by recombining envelopes and fine structures from different reproducible stimuli. Detection judgments for noise-alone or tone-plus-noise stimuli that had common envelopes but different fine structures or common fine structures but different envelopes were compared. The results showed similar patterns of responses to stimuli that shared envelopes, indicating the importance of envelope cues; however, fine-structure cues were also shown to be important. The relative weight assigned to envelope and fine-structure cues varied across subjects and across interaural conditions. The results also indicated that envelope and fine-structure information are not processed independently. Implications for monaural and binaural models of masking are discussed.

Previous research has repeatedly shown that people can find a visual target significantly faster if spatial (3D) auditory displays direct attention to the corresponding spatial location. However, previous research has only examined searches for static (non-moving) targets in static visual environments. Since motion has been shown to affect visual acuity, auditory acuity, and visual search performance, it is important to characterize aurally-aided search performance in environments that contain dynamic (moving) stimuli. In the present study, visual search performance in both static and dynamic environments is investigated with and without 3D auditory cues. Eight participants searched for a single visual target hidden among 15 distracting stimuli. In the baseline audio condition, no auditory cues were provided. In the 3D audio condition, a virtual 3D sound cue originated from the same spatial location as the target. In the static search condition, the target and distractors did not move. In the dynamic search condition, all stimuli moved on various trajectories at 10 deg/s. The results showed a clear benefit of 3D audio that was present in both static and dynamic environments, suggesting that spatial auditory displays continue to be an attractive option for a variety of aircraft, motor vehicle, and command & control applications.

Processing mechanisms used for detection of tones in noise can be revealed by using reproducible noise maskers and analyzing the pattern of results across masker waveforms. This study reports detection of a 500-Hz tone in broadband reproducible noise by rabbits using a set of masker waveforms for which human results are available. An appetitive-reinforcement, operant-conditioning procedure with bias control was used. Both fixed-level and roving-level noises were used to explore the utility of energy-related cues for detection. An energy-based detection model was able to partially explain the fixed-level results across reproducible noise waveforms for both rabbit and human. A multiple-channel energy model was able to explain fixed-level results, as well as the robust performance observed with roving-level noises. Further analysis using the energy model indicated a difference between species: human detection was influenced most by the noise spectrum surrounding the tone frequency, whereas rabbit detection was influenced most by the noise spectrum at frequencies above that of the tone. In addition, a temporal envelope-based model predicted detection by humans as well as the single-channel energy model did, but the envelope-based model failed to predict detection by rabbits. This result indicates that the contributions of energy and temporal cues to auditory processing differ across species. Overall, these findings suggest that caution must be used when evaluating neural encoding mechanisms in one species on the basis of behavioral results in another.

A single-interval, yes-no, tone-in-noise detection experiment was conducted to measure the proportion of "tone present" responses to each of 25 reproducible noise-alone and tone-plus-noise waveforms under narrowband (100 Hz), wideband (2900 Hz), monotic, and diotic stimulus conditions. Proportions of "tone present" responses (estimates of the probabilities of hits and false alarms) were correlated across masker bandwidths and across monotic and diotic conditions. Two categories of models were considered; one based on stimulus energy or neural counts, and another based on temporal structure of the stimulus envelope or neural patterns. Both categories gave significant correlation between decision variables and data. A model based on a weighted combination of energy in multiple critical bands performed best, predicting up to 90% of the variance in the reproducible-noise data. However, since energy-based models are unable to successfully explain detection under a roving-level paradigm without substantial modification, it is argued that other variations of detection models must be considered for future study. Temporal models are resistant to changes in threshold under roving-level conditions, but explained at most only 67% of the variance in the reproducible-noise data.

Artificial neural networks were trained to identify the location of sound sources using head related transfer functions (HRTFs). The simulated signals were filtered clicks presented from virtual speakers placed at 15 degree steps in azimuth and 18 degree steps in elevation. After signals were passed through the HRTFs, quarter-octave spectra were computed. The inputs to the networks were either monaural spectra or binaural difference spectra. In some cases a broadband cross-correlation term was also provided. Backpropagation was used to train the networks. Separate networks were trained for each combination of spectral information. In some cases the networks achieved performance comparable to that of human observers in both accuracy and number of front-back reversals. With a hierarchy of neural networks accuracy better than human performance was obtained. >

Previous research from our laboratory has shown that uncertainty about the spatial location of a masking sound (randomly selected from 1 of 239 locations) can dramatically reduce localization accuracy for a simultaneous target relative to the case in which the masker location is known exactly. One possibility is that knowing the masker location enables the listener to establish a spatial attention filter at that location to suppress the masker and better localize the target. In this experiment, the level of masker spatial uncertainty was systematically varied across blocks by varying the number of possible masker locations (1, 2, 4, 8, or 239) and informing subjects about these possible locations prior to the start of each block. Localization errors were found to increase systematically in the left/right, front/back, and up/down dimensions as the number of potential masker locations increased; this effect was most prominent in the left/right dimension, where localization errors increased by nearly 30 degrees across conditions. Moreover, for a masker in a given location, errors generally increased across these levels of masker spatial uncertainty, consistent with the notion that there is a cost to distributing attention across multiple locations.

We have submitted, revised, and/or published 17 papers, chapters, and books. In addition we have made 17 presentations a various meetings. Some of the results reported in these papers ar based on the research efforts begun under AFOSR NL-91-0289 (including work on masked detection and sound localization in noise. We continue our theoretical work on spatial hearing. We have collected new data concerning the localization of speech stimuli, the effects of the listening environment on the perception of virtual audio and auditory-aided visual search. We have developed hardware and software to support planned experiments in a number of topic areas and we have published an edited book on binaural and spatial hearing in real and virtual

The CAVE is a multi-person, room-sized, high-resolution, 3D video and auditory environment, which can be used to present very immersive virtual environment experiences. This paper describes the CAVE technology and the capability of the CAVE system as originally developed at the Electronics Visualization Laboratory of the University of Illinois- Chicago and as more recently implemented by Wright State University (WSU) in the Armstrong Laboratory at Wright- Patterson Air Force Base (WPAFB). One planned use of the WSU/WPAFB CAVE is research addressing the appropriate design of display and control interfaces for controlling uninhabited aerial vehicles. The WSU/WPAFB CAVE has a number of features that make it well-suited to this work: (1) 360 degrees surround, plus floor, high resolution visual displays, (2) virtual spatialized audio, (3) the ability to integrate real and virtual objects, and (4) rapid and flexible reconfiguration. However, even though the CAVE is likely to have broad utility for military applications, it does have certain limitations that may make it less well- suited to applications that require 'natural' haptic feedback, vestibular stimulation, or an ability to interact with close detailed objects.© (1997) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Many real-world auditory scenes are dynamic and complex, with multiple sounds that may change location over time. In this experiment, we examined the ability of listeners to localize a spatially consistent target sound in a dynamic, spatially varying auditory scene. The target and masker stimuli were composed of sequences of 60-ms bursts of uncorrelated noise (2 to 16 bursts in duration) and differed only in their degree of spatial consistency. Specifically, each target burst within a sequence came from the same spatial location (which varied from trial to trial), whereas each masker burst within a sequence came from a different, randomly chosen spatial location. The listener’s task was to localize the spatially-consistent sequence. Localization errors decreased by approximately 11° with each doubling of the sequence duration, and approached quiet performance with 16-burst sequences. Adding a second masker increased localization errors by approximately 14° overall. These results suggest that spatial information can be combined across multiple observations over time to identify and localize a spatially consistent target in a dynamic auditory scene. These data will be discussed in terms of the information obtained from each burst and the manner in which the information is combined across bursts.

Previous research suggests that the localization of a target sound in a masking noise is more accurate when the masker is preceded by a forward masker fringe. It is unclear what information listeners are exploiting to produce this improvement in performance. Does the fringe simply reduce spatial uncertainty (by cuing the location of the masker) or does it provide a “context” against which the onset of the signal is more easily discerned? In order to address these questions, we examined conditions in which the masker is (a) pulsed on and off with a 60 ms target (no fringe), (b) immediately preceded by a 500‐ms masker fringe; or (c) preceded by a 500‐ms fringe + a 500‐ms gap. Masker spatial uncertainty was also manipulated by presenting the noise from a fixed and known location throughout a block, or from a location that varied across trials. The results suggest that listeners benefit from the presence of a forward masker fringe even when the masker location is known. However, localization performance is best when this fringe occurs immediately prior to the masker (i.e., no gap), suggesting that onset‐related cues also contribute to the improvement in localization. [Work supported by the AFOSR]

one goal of this program of research is to examine masking among spatially separated sound sources. The results indicate that masking release on the order of 8-20 dB can be observed in free-field masking situations when the signal and the masker are spatially separated by 900. This magnitude of masking release is comparable to that observed in traditional binaural masking level difference experiments, where the stimuli are presented through headphones. However, while masking release observed in headphone studies is typically assumed to be based on interaural differences in phase or on interaural differences in intensity, we observed substantial masking release in the median plane where interaural differences are minimal. our own headphone masking research is also questioning traditional models of binaural masking. Work is underway to develop a neural network based model of sound localization. The results of these studies will have implications for the development of virtual environments and auditory displays.

Several studies have shown that detection of a dichotic target signal in a concurrent diotic noise masker improves when the masker onset occurs prior to that of the target and/or the masker duration extends beyond the target offset (so‐called masker fringe; McFadden, 1966; Trahiotis et al. 1972). It has been argued that this improved performance results from the fact that the masker fringe provides a baseline set of interaural parameters against which a change in parameters, associated with the signal onset or offset, can be detected. As the duration of the fringe increases, more time is available to estimate these baseline parameters, thus improving signal detectability. However, little is known about the relation between masker fringe and spatial hearing. In this study, the localization of a 250‐ms click train signal presented in a broadband directional noise masker was measured as a function of the duration of the masker fringe and the signal‐to‐noise ratio. The results indicate that the presence of ma...

Several psychophysical models for masked detection were evaluated using reproducible noises. The data were hit and false-alarm rates from three psychophysical studies of detection of 500-Hz tones in reproducible noise under diotic (N0S0) and dichotic (N0Spi) conditions with four stimulus bandwidths (50, 100, 115, and 2900 Hz). Diotic data were best predicted by an energy-based multiple-detector model that linearly combined stimulus energies at the outputs of several critical-band filters. The tone-plus-noise trials in the dichotic data were best predicted by models that linearly combined either the average values or the standard deviations of interaural time and level differences; however, these models offered no predictions for noise-alone responses. The decision variables of more complicated temporal models, including the models of Dau et al. [(1996a). J. Acoust. Soc. Am. 99, 3615-3622] and Breebaart et al. [(2001a). J. Acoust. Soc. Am. 110, 1074-1088], were weakly correlated with subjects' responses. Comparisons of the dependencies of each model on envelope and fine-structure cues to those in the data suggested that dependence upon both envelope and fine structure, as well as an interaction between them, is required to predict the detection results.

Real‐world listening experiences suggest that listeners possess a much greater ability to recover spatial information from multiple simultaneous sounds than would be predicted from the results of most multisource listening experiments. Recent data from our laboratory are consistent with these real‐world experiences, and indicate that listeners can reliably report the location of a sound that has been deleted from a multisource scene. However, performance degrades as the number of competing sounds increases, and it is unclear if this 'set size effect' arises from changes in signal‐to‐noise ratio (SNR) related to the number of competing sources, or from attentional constraints that limit the number of simultaneous sources that can be monitored. In this study, SNR was held roughly constant by fixing the number of competing sources at 8, and the 'relevant set size' was varied by displaying on a visual monitor, prior to an observation interval, a list of 1,2,3,4,6, or 8 environmental sounds, one of which was the target sound (the sound that would be deleted). The ability to localize this target sound decreased as the relevant set size increased, suggesting that the ability to monitor complex auditory scenes is mediated by attentional constraints.

Yost [J. Acoust. Soc. Am. 78,901-907 (1985)] found that the detectability of a 30-ms dichotic signal (S pi) in a 30-ms diotic noise (No) was not affected by the presence of a 500-ms dichotic forward fringe (N pi). Kollmeier and Gilkey [J. Acoust. Soc. Am. 87, 1709-1719, (1990)] performed a somewhat different experiment and varied the onset time of a 25-ms S pi signal in a 750-ms noise that switched, after 375-ms, from N pi to No. In contrast to Yost, they found that the N pi segment of the noise reduced the detectability of the signal even when the signal was temporally delayed well into the No segment of the noise and suggested that the N pi segment of noise acted as a forward masker. To resolve this apparent conflict, the present study investigated the detectability of a brief S pi signal in the presence of an No masker of the same duration as the signal. The masker was preceded by quiet or an N pi forward fringe and followed by quiet, an No, or N pi backward fringe. The present study differs from most previous studies of the effects of the masker fringe in that the onset time of the signal was systematically varied to examine how masking changes during the time course of the complex fringe-masker-fringe stimulus.(ABSTRACT TRUNCATED AT 250 WORDS)

The goal of this project is to specify the transformations used by the auditory system in order to determine the presence of the signal in an auditory masking task, with particular emphasis on the role of processes that compare information in the frequency domain and in the time domain. Classical models that restrict analysis to a single frequency band and a single temporal window are evaluated. In addition, the role of pattern or 'profile' analysis in auditory processing is being assessed by fitting more complex models to the data. The results show that similar cues determine performance in monaural and binaural masking tasks. Information remote from the signal in frequency can be used to overcome the effects of uncertainty about the stimulus level. Masking noise which does not overlap with the signal in time can either improve or degrade the detectability of the signal, depending upon the interaural phase relations of the masker and the signal. The analyses yield a quantitative description of processes that compare information within the frequency band and temporal window that contain the signal to information in other spectral/temporal regions. Other significant results include a more complete description of internal noise processes and evidence that the external masker is not cancelled by the binaural processor.

The threshold of a short interaurally phase-inverted probe tone (20 ms, 500 Hz, S pi) was obtained in the presence of a 750-ms noise masker that was switched after 375 ms from interaurally phase-inverted (N pi) to interaurally in-phase (No). As the delay between probe-tone offset and noise phase transition is increased, the threshold decays from the N pi S pi threshold (masking level difference = 0 dB) to the No S pi threshold (masking level difference = 15 dB). The decay in this "binaural" situation is substantially slower than in a comparable "monaural" situation, where the interaural phase of the masker is held constant (N pi), but the level of the masker is reduced by 15 dB. The prolonged decay provides evidence for additional binaural sluggishness associated with "binaural forward masking." In a second experiment, "binaural backward masking" is studied by time reversing the maskers described above. Again, the situation where the phase is switched from No to N pi exhibits a slower transition than the situation with constant interaural phase (N pi) and a 15-dB increase in the level of the masker. The data for the binaural situations are compatible with the results of a related experiment, previously reported by Grantham and Wightman [J. Acoust. Soc. Am. 65, 1509-1517 (1979)] and are well fit by a model that incorporates a double-sided exponential temporal integration window.

Historical reports from suddenly deafened adults describing a sense of detachment from a world devoid of auditory input suggest that auditory cues may be crucial for achieving a sense of presence in virtual environments [R. H. Gilkey and J. M. Weisenberger, Presence 4, 357–363 (1995)]. However, factors other than acoustic fidelity can affect the perceived quality of a virtual auditory display. For example, Simpson et al. [J. Acoust. Soc. Am. 100, 2633(A) (1996)] showed that the sense of presence in virtual auditory environments was driven by characteristics of the real listening environment in which the virtual audio was heard. Specifically, when the virtual auditory environment matches the real listening environment, the listening experience is particularly compelling. However, it was not clear whether auditory or visual experience with the real room mediated this effect. This paper describes attempts to separate the impact of auditory and visual information on presence by systematically manipulating these cues in synthesized environments. Implications for design of virtual environments and spatialized auditory displays are considered. [Work supported by AFOSR and Ohio Board of Regents.]

Although most sound localization research has examined the ability of listeners to determine the location of a single sound presented in a quiet (typically anechoic) environment, most real‐world listening situations are more complex, with multiple simultaneous sounds. Here, an initial experiment, designed to examine localization in multisource environments, is described. Listeners judged the location of a target signal (speech or environmental sound, presented normally or time‐reversed) masked by up to four simultaneous sounds. In each block of trials, the observation interval was either preceded by, or followed by, a cueing interval, during which the stimulus to be localized was identified. It was expected that these two approaches would lead to different answers, as the associated tasks presumably address different listening strategies (i.e., analytic listening versus monitoring). The results indicate that, in all conditions, localization errors increase as the number of simultaneous sources increases. Moreover, performance degrades more rapidly in the post‐cue condition than in the pre‐cue condition. Surprisingly, this difference occurs for as few as two simultaneous sources, suggesting that there is a substantial cost when listeners are asked to remember what sounds were present and where those sounds were located in a complex auditory environment. [Work supported by AFOSR.]

Gilkey, Mason, and Kidd [J. Acoust. Soc. Am. 109, 2468 (2001)] found that although informational masking has typically been thought to depend more strongly on the variability in the ensemble of masking stimuli presented across trials, trial‐by‐trial performance was strongly related to the specific masker waveform presented on each trial. Moreover, hit and false‐alarm rates to individual masker stimuli under multiple‐burst‐same (MBS) conditions were well predicted by the patterns of energy in those stimuli. Hit and false‐alarm rates under the multiple‐burst‐different (MBD) condition were more difficult to predict. This paper reports on further efforts to model those data. The results indicate that performance under both MBS and MBD conditions can be predicted using a linear combination of envelope statistics computed from seven narrow frequency bands surrounding the 1000‐Hz signal frequency. MBS responses are related to the average height of the envelope in these bands and MBD responses are related to the standard deviations of the envelopes. [Work supported by NIH/NIDCD.]

To elucidate the roles of envelope and carrier cues in detection of a 500‐Hz tone in 50‐Hz bandwidth noise under N0Sπ conditions, hit‐ and false‐alarm rates were collected for 4 sets of reproducible stimuli. The envelopes and carriers from two sets of stimuli (each with 25 signal‐plus‐noise and 25 noise‐alone waveforms) were extracted and recombined to form 2 new sets (each having the carriers from one of the original sets and the envelopes from the other). Noises were preselected to produce minimal spectral splatter in the recombination processes. Preliminary results for signal‐plus‐noise trials suggest that carrier cues (e.g., temporal fine structure) are dominant over envelope cues (i.e., response patterns were more similar between sets with the same carriers but different envelopes, than between sets with the same envelopes but different carriers); however, carrier cues cannot completely explain the data. Results for noise‐alone trials suggest that both envelope cues and carrier cues contribute to listener responses. Further analyses will explore potential interactions between monaural and binaural stimulus properties, as well as interactions between interaural envelope cues and carrier cues. [Supported by NIH DC007798(SAD), DC001641(LHC), and DC00100(HSC).]

Both energy‐ and envelope‐based models have been successfully used to predict narrowband tone‐in‐noise detection thresholds. To distinguish between these models, hit and false‐alarm rates for 25 reproducible maskers were measured in the N0S0 and N0Sπ interaural configurations and 4 cue conditions: (1) normal energy and envelope cues, (2) restricted envelope cues, (3) restricted energy cues, or (4) restricted energy and envelope cues. Preliminary results show that hit and false‐alarm rates were strongly correlated between the conditions with and without restricted energy cues, indicating that the reduction of energy cues did not substantially affect the detection process. However, hit and false‐alarm rates were more weakly correlated between the conditions with and without restricted envelope cues, suggesting that the reduction of envelope cues did affect the detection process. Restricting envelope cues also had a larger (but still modest) effect on d′ under the N0Sπ configuration, but restricting energy c...

Informational masking has been distinguished from energetic masking in that it seems to depend less on the specific characteristics of the stimulus presented on a particular trial (e.g., energy in the critical band) and more on the variability in the ensemble of stimuli presented across trials. We employ techniques that have been used in reproducible noise masking experiments [D. M. Green, Psychol. Rev. 71 (1964)] to examine the degree to which subjects responses in informational masking tasks are driven by the characteristics of the individual maskers. A single‐interval task was used to collect hit and false–alarm rates to individual maskers under multiple‐burst‐same (MBS) and multiple‐burst‐different (MBD) conditions [G. Kidd et al., J. Acoust. Soc. Am. 95 (1994)]. During each block the 25 MBS or 25 MBD reproducible maskers were presented in random order across trials. Both hit and false–alarm rates showed significant dependence on the masker waveform under both MBS and MBD conditions (p

This work examines sound source lateralization of ongoing broadband noise targets in the presence of a single ongoing broadband noise jammer, which was either a simple simulated reflection of the target or was uncorrelated with the target. Identification thresholds of ±300‐μs ITD targets and lateralization thresholds were determined. Identification thresholds of the reflection jammers were 3.2 dB lower (P=0.027) than the identification thresholds of the jammers that were uncorrelated with the targets. Lateralization thresholds of the reflection jammers were 2.85 dB lower (P=0.040) than the lateralization thresholds of the uncorrelated jammers. There was a slight trend, in agreement with the current understanding of the precedence effect, for both the identification and lateralization thresholds of the simple reflection jammers to be dependent on jammer ITD. The 0‐μs ITD reflection jammer produced a threshold that was 1.9 dB lower (P=0.104) than the threshold of the 643‐μs ITD reflection jammer. Additionally comparisons between the lateralization thresholds of normal‐hearing listeners and two cross‐correlation models were made. The models obtained lateralization thresholds as low as −13 dB, up to 10 dB better than normal‐hearing performance. [Work supported by NIH Grant No. R01 DC00100.]

Most often the cocktail party effect has been evaluated for relatively simple stimuli (e.g., isolated words or simple sentences masked by noise) in anechoic environments. Here, a stimulus situation is used which is similar to that employed by Plomp [Acustica 34, 200–211 (1976)] (i.e., continuous discourse for both target and masker) in a reverberant room. However, instead of asking subjects to make a subjective judgment of intelligibility, the ‘‘tracking procedure’’ of DeFillipo and Scott is used [J. Acoust. Soc. Am. 63, 1186–1192 (1978)] to simulate a conversation. This procedure measures the number of words per minute (WPM) successfully transmitted between the target talker and the subject. Preliminary results show increases in WPM of about 20% when the masker is separated from the target by 90 deg in the horizontal plane relative to the case when both the target and masker are at the same location in front of the subject. (The change in S/N necessary to maintain constant WPM between these conditions indicates that the magnitude of the effect is similar to that observed by Plomp.) Separating the target and masker(s) in elevation has little effect on WPM. [Work supported by AFOSR and the Ohio Board of Regents.]

Janko et al. [Binaural and Spatial Hearing in Real and Virtual Environments, edited by Gilkey and Anderson (Erlbaum, Hillsdale, NJ, 1996), pp. 557–570] used neural networks to describe auditory localization and found that data in quiet were insufficient to discriminate among models. To further constrain the models, localization in the presence of a spatially fixed masker is considered here. The wideband targets and maskers were filtered by head‐related transfer functions, then by a gammatone filter bank. A model for binaural interaction similar to that of Lindemann [J. Acoust. Soc. Am. 80, 1608–1622 (1986)] was used to process the filter‐bank outputs. The inhibited cross‐correlation output of the binaural processor was sampled and used as input to a neural network (e.g., 24 correlation lags at each of 13 frequency channels for 312 input nodes). The network was trained using back propagation to estimate the target location in quiet, then tested on targets in noise at several values of signal‐to‐noise ratio...

Anecdotal reports from casual listening in auditory virtual environments suggest that the ‘‘compellingness’’ of virtual sounds depends not only on the fidelity of the synthesized environment but also on the characteristics of the real environment in which the sounds are presented (i.e., even though the sounds are presented through headphones, nonauditory characteristics of, or previous auditory experience in, the listening environment may influence the perception of the virtual environment). In particular, when the synthesized environment ‘‘matches’’ the listening environment the auditory experience can be strikingly realistic. To quantify this phenomenon, binaural recordings of ‘‘everyday’’ sounds were made through the KEMAR manikin in each of three real rooms (16 to 194 m3). A 3×3 between‐subjects design was used; each untrained subject listened to one of the three recordings over headphones while seated with eyes open in one of the three rooms (the recording might or might not match the room). The effects of recording room size and listening room size on objective measures of 3‐D localization accuracy (azimuth, elevation, and distance) and on subjective measures of the listening experience (sense of presence, sound quality, and vividness) are evaluated. [Work supported by AFOSR F49620‐95‐1‐90106, NRC, and Wright State University.]

Neural networks were trained with back propagation to localize ‘‘virtual’’ sounds that could originate from any of 24 azimuths (−165° to +180°) and 6 elevations (−36° to +54°). The sounds (clicks) were filtered with head related transfer functions and transformed into 22‐point quarter‐octave spectra. The networks were composed of 22 to 44 input nodes, followed by 50 hidden nodes and 30 output nodes (24 azimuth nodes and 6 elevation nodes). With the ‘‘binaural’’ configuration, the interaural delay and the interaural difference spectrum were provided as inputs to the network. With the ‘‘monaural’’ configuration, separate networks were trained with the left‐ear and the right‐ear spectra; a third, arbitrator, network learned to localize based on the output of these two monaural networks (i.e., the activation levels of their azimuth and elevation nodes). The monaural configuration did not allow for binaural interaction in the traditional sense. Both configurations achieved performance comparable to humans, sug...