885096007

years, with normal or ncar-normal hearing). Both thc mean thresholds and standard deviation$ for older adults were elevated approximately fourfold compared to thc younger subjects. The age difference is con-sistent with previous reports of age-related loss in sensitivity to ITDs. However, in our study and in previous studies, results may have been confounded by a mcthodological factor. The older adults had very large excursions across stimulus levels in the adaptive (staircase) psycho-physical procedurę used. Thus thc age difference may reflect attentional as well as auditory factors. Further rcsearch using thc method of con-stant stimuli in a smali number of older adults is in progress. Prelimi-nary results indicate that there is still an age difference in interaural Processing, but that the difference is smaller than methods using the adaptive procedurę have indicated. The findings have implications not only for age-related changes in interaural proccssing. but also for the usc of psychophysical testing methods with older adults.

5SP34. Humań audition of ultrasound. Martin L. Lenhardt, Alexander M- Ciarkę, Ruth A. Skellctt, David A. Soltanpour, and Alfred L. Ochs (Med. College of Virginia, Virginia Commonwealth Univ., Box 168, Richmond, VA 23298-0168)

Tones above 20 000 Hz are not inaudible to humans as the term “ultrasound" implies, but rather are audible only via bonę conduction. Placing a high-frequency vibrator on the skin over the head, neck, or upper trunk results in a high-pitched auditory percept, yielding no sense of cutaneous stimulation. Audibility thresholds for tones of 25 and 62.5 kHz were determined in four normal-hearing subjects for various sites on the upper torso. The results of these ultrasonic “maps" indicate that the arca of the mastoid bonę is the most sensitive. Position and orien-tation of thc vibrator on the mastoid influence sensitivity. Smali changes of freąuency and/or vibrator orientation affect the “ear” in which the sound is heard. Thls finding is in contrast to a previous report from this laboratory of 60- to 80-dB attenuation between mastoid bones in a dry skuli [Dunlap et aL, Otolaryngol. Head Neck Surg. 99, 389-391, (1988)]. By placing water in thc cranial vaułt, skuli attenuation of 17 dB (25 kHz) and 36 dB (62.5 kHz) was obtained, values better match-ing the perceptual findings. It is hypothesized that standing waves are created intracranially, which in part determine the ear of perception. It is possible that cochlea inner hair celi stereocilia and/or vestibular type I cells in the striolar region of thc saccule are set into ultrasonic reso-nance [Strclioff et aL, Hear. Res. 18, 169-175 (1985)], inducing a dif-ferent form of stimulation than traditional air conduction hearing. [Work supported by Hearing Innovations, Inc., Tucson, AZ.] 5SP35. Gender classification based upon statistics of the spec trał distributions: Walking perception. Xiao-Feng Li. Richard E Pastore (Dept. of Psychol., SUNY, Binghamton, NY 13901), and Robert J. Logan (IBM, Poughkeepsie, NY 12602)

The current study continued the previous rcsearch using as stimuli the sounds of humans walking on a hard surface [Li et aL, J. Acoust. Soc. Am. Suppl. 1 87, S24 (1990)] to investigate the acoustic properties that delineated categories of ecologicałly valid auditory source events. Various statistics of the walking spectra were computed, ahd subjected to a factor analysis in order to identify the spectra! properties that differentiated between actual gender class of source events (małe and female walkers). Two classes of information were identified as impor-tant in distinguishing between the małe and female spectra: (1) central tendency and (2) slopes of spectral atlack and decay. A multiple re-gression analysis indicated the importance of these classes of information in determining the perception of the walker gender. A follow-up experiment then manipulated these statistical properties to verify their contributions in the perceptual classification of walker gender. [Work supported by NSF.]

LIBERTY B, 12:40 TO 3:15 P.M.

WEDNESDAY AFTERNOON, 1 MAY 1991

Session 5UW

Underwater Acoustics: Scattering and Reverberation

Stanley Chin-Bing, Chair

Naoal Ocean Research and Deuelopment Actwity, Codę 220, Numerical Modeling Dwision,

Stennis Space Center, Mississippi 39529-5004

Chairis Introduction—12:40

Contrihuted Papers

12:45

5UW1. Generalized Fresnel corrections for scattering in underwater sound channels with rough boundaries. Suzanne T. McDaniel (Appl. Res. Lab., Penn Stale Univ., University Park, PA 16804)

The Fresnel corrected Kirchhoff approximation for forward scattering from rough surfaces is extended to treat the generał case of bound-ary scattering in a refractive underwater sound channel where multiple interactions with the surface may occur. Expressions are obtained for the reflected intensity and the spatial covariance of the scattered acoustic fleld in the high-frequency limit. Numerical examples are presented

1940 J. Acoust. Soc. Am., Vol. 89, No. 4. Pt. 2, April 1991 to demonstrate the strong predicted dependence of both the reflectivity and spatial covariance on refraction. [Research supported by ONT with technical management provided by NCSC ]

IKK)

5UW2. Far-field considerations in surface and bottom scattering.

Darrell R. Jackson and Erie I. Thorsos (Appl. Physics Lab., College of Ocean and Fishery Sci., Univ. of Washington, Seattle, WA 98105)

It is sometimes argued that the concept of scattering cross section is 121st Meeting: Acoustical Sodety of America 1940