885095954

9:15

8PP2. Monaural spatial resolution in the horizontal piane under dynamie and static listening conditions. D. Wesley Grantham and David W. Chandler (Bill Wilkerson Ctr. and Vanderbilt Univ. Div. of Hearing and Speech Sci._t 1114 I9th Ave. South, Nashville, TN 37212)

Auditory spatial resolution was measured in the horizontal piane for three normal-heanng adult subjects in a darkened anechoic chamber. In all conditions. subjects’ right ears were occluded by an EAR foam insert pług plus an external sound-attenuating rnufF, providing a total of about 40 dB of attenuation. In the dynamie conditions, the minimum audible movement angle (MAMA) was measured—that is, the angular extent a moving target had to traverse to bc just discriminable from a stationary target. In the static conditions, the minimum audible angle (MAA) was measured—that is, the minimum angular separation between two se-quentially presented stationary targets that was just discriminable from a single stationary target presented twice in succession. In generał, MA-MAs and MA As decreased as stimulus bandwidth inereased from 0 Hz (pure tonę at 3000 Hz) to wideband. MAAs were extremely variable across and within subjects (varying from 10° to 40* of arc). MAM As for slow-ve!ocity targets (10%) were usually loiver than MAAs measured for the same signals, a result that contrasts with the results from bin-aural resolution tasks [D. W. Chandler and D. W. Grantham, J. Acoust. Soc. Am. Suppl. I 87, S64 (1990)). These results suggest that special-ized mechanisms sensitive to dynamie stimuli may play a role in monaural spatial resolution. [Work supported by NIH.)

9:30

8PP3. Monaural localization, revisited. Frederic Wightman, Doris Kistler, and Mariannę Arruda (Dept. of Psychol. and Waisman Ctr., Univ. of Wisconsin, Madison, W1 53705)

There are numerous reports in the psychoacoustical literaturę that human listeners can localize sound sources reasonably well with one ear. Since interaural differcnce cues are presumably eliminated in monaural conditions, the so-callcd “monaural spectral cues" introduced by pinna filtering are assumed to provide the information necessary for accurate localization in monaural conditions. In these experiments, listeners localize wideband noise bursts presented either in free-field (with one ear occluded) or via headphones (with the signal to one phone either attenuated or disconnected). In the headphone conditions, pinna filtering effects are added digitally, such that the waveforms at a listcn-er’s eardrum are nearly the same as those produced by free-field sources. The noise spectrum is scrambled from trial to trial to prevent learning. With the noise bursts presented at about 30 dB SL in free-field, the results are consistent with other recent reports and suggest that some abilily to localize, especially in the vertical direction, is retained in monaural conditions. However, when the identical stimuli are presented via headphones, there is no indication that sources can be localized monaurally. In other conditions, listeners localize conslant spectrum stimuli, free-field stimuli at 70 dB SL, and binaural headphone stimuli with one ear attenuated. Results from these conditions suggest that monaural localization in free-field is most likely mediated by smali head movcments, a priori knowledge of the stimulus spectrum, and acoustical leakage through the ear-occluding devices uscd to monauralize the listeners. [Work supported by NIH and NASA.)

9:45

8PP4. Influence of onset cues in lateralization. Richard L. Freyman (Dept. of Commun. Disord., Univ. of Massachusetts, Amherst, MA 01003) and Patrick M. Żurek (MIT. Cambridge, MA 02139)

When pairs nf clicks delivered binaurally via earphones are repeated to form a click train, interaural cues at stimulus onset can exert a strong influence on the perceived intracranial position of the entire stimulus [Sabcn and Perrott, J. Acoust. Soc. Am. Suppl. 1 86, SI 1 (1989)). This phenomenon has been observed most clearly under conditions in which the interaural limę delays in the ongoing click trains—because they

1995 J. Acoust. Soc. Am., Vol. 89. No. 4, Pt. 2. April 1991 alternate between two valucs—are ambiguous. For example, when a single click pair with a 500-^-time lead to the right is followed by a 250-ms train consisting of alternating diotic and left-leading click pairs, the entire stimulus is lateralized to the right. This paper will describe conditions in which onset cucs dominate the perceived lateral position as well as those in which steady-state cues determine the lateralization. It will also describe stimulus manipulations that appear to produce a relcasc from the influence of the onset cucs. For cxample, a bnef silent gap inserted in the middle of a click train releascd the listener from the influence of the original onset, while a hrief burst of binaural white noise presented during the middle of the click tram did not. [Work supported by NSF and NIH.)

10:00

8PP5. DifTerential lateralization interfercnce for interaural time and interaural !evel differences. Laurie M. Heller and Virginia M. Richards (Dept. of Psychol., 3815 Walnut St., Univ. of Pennsylvania, Philadelphia, PA 19104)

Lateralization thresholds were obtained in a 2IFC task using 20()-ms noise bands presented with either an interaural time or level ditference. The elevation in lateralization threshold (interference) caused by a si-multaneously presented noise band was measured. When the ITD was applied to a 50-Hz-widc noise band centered at 500 Hz, there was minimal interference from a diotic 400-Hz-wide noise band centered at 4000 Hz. When the ITD was applied to the noise band centered at 4000 Hz, there was substantial interference from a diotic band of noise centered at 500 Hz [D. McFadden and E. G. Pasancn, J. Acoust. Soc. Am. 59, 634-639 (1976)). In contrast, the interfering effect of a diotic noise band was greater when the ILD was applied to the 500-Hz band of noise than when it was applied to the 4000-Hz band of noise. A similar patiem of results was obtained when the ITD or ILD of the interfering band of noise was randomized from interval to interval, although ran-domization produced greater interference in all conditions. The com-plementary paltem of results for time and !evel differences favors an explanation based on averaging between simultaneous interaural param-eters rather than a masking-based cxplanation.

10:15

8PP6. Modeling of binaural discrimination of multiple sound sources: A contribution to the development of a “cocktail party processor.” H.

Slatky (Lehrstuhl fuer allgemeine Elektrotech. und Akust., Ruhr-Univ. Bochum, D-4630 Bochum I, Germany)

The human hearing system when being stimulated simultaneously by narrow-banded signals from two loudspeakers at ditfcrent locations is able to determine the directions of both sound sources. Howcver, con-ventional models of binaural interaction which are based on interaural cross correlation in critical bands fail in discriminating the source directions. The reason is that the cross-correlation functions at the output of the model display severe fluctualions in time. A revised binaural cross-correlation model that contains an additional “recomputation mechanism” will be presented. This mechanism estimates directions and energies of sound sources from cross-correlai ion functions fluctuating in time. Using this mechanism the results of relcvant auditory experiments can bc reproduced. This model has been applied to construct a cocktail party processor, which is able to enhance the signal-to-noise ratio con-siderably.

10:30

8PP7. Changing echo thresholds. Richard Freyman (Commun. Disord., Univ. of Massachusetts, Amherst, MA 01003), Racłicl Clifton, and Ruth Litovsky (Univ. of Massachusetts, Amherst, MA 01003)

Echo thresholds for a 4-ms white noise burst were measured in an anechoic chambcr with the leading loudspeaker located at 45* left and three lagging loudspeakers at 35*, 45*. and 55* right. The listener judged

121 st Meeting: Acoustical Society of America 1995