885095980

his famous “residue” theory. With Licklider’s demonstration, the classical controversy, going back to Seebeck and Ohm, seemed to be solved definitely in favor of Seebeck. “Periodicity pitch” theory, as elaborated by Licklider in his “duplex” (1951) and “triplex" (1955) theories, seemed to have succeeded “place pitch” theory, promoted by famous scientists such as Helmholtz and von Bekesy, but for how long? The paper will review thc significance of Licklider’s contribution in the light of subsequent experimental evidence showing that frcquency resolution docs play an essential role in pitch perception.

2:30-2:45

Break

Contributed Papers

2:45

3PP6. Sharing place, period,'and concepts with Licklider. I. J. Hirsh and P. G. Singh (Washington Univ. and Central Inst. for the Deaf, 818 S. Euclid, St. Louis, MO 63110)

The naturę of pitch as a dual attribute comprising both “place” and “period” aspects was dramatically illustrated by Licklider (J. Acoust. Soc. Am. 26, 945 (1954)] in his famous masking demonstration. Lis-tener confusions in judgments of pitch in subsequent expcriments could be attributed to conflicting “place" and “period” information. Although thc conccpt of pitch duality has been updated by pitch theorists, one aspect that has been consistcntly ovcrlookcd is thc relation bctween “place” or "spectral” pitch and timbre. For example, the direction of a pcrceived pitch change is often confused with an ordinal aspect of timbre, like “brightness” and “sharpness.” The connection between “place” pitch and timbre will be discussed and demonstrated with examples from the literaturę and from our own experiments. (Work supported by USAFOSR and NIH.)

3:00

3PP7. Effects of signal-to-noise ratio on the frequency discrimination of complex tones with overlapping and nonoverlapping harmonics.

Brian C. J. Moore and Brian R. Glasberg (Dept. of Expcrimental Psych., Univ. of Cambridge, Downing St., Cambridge CB2 3EB, England)

Frequency differcnce iimens for multiple-component complex tones (DLCs) were measured using an adaptive two-interval, tw(valtemativc forced-choice task. The tones werc presented either in quiet or in pink noise. The tones to be discriminated either had all harmonics in com-mon or no c om mon harmonics. DLCs for the tones with no common harmonics were generally, but not always, larger than those for complex tones with common harmonics. For the former, performance did not worsen “monotonically” with increasing noise !evel, but tended either to stay constant or improve at first, only worsening when the tones were almost completely masked by the noise. For the lat ter, performance tended to worsen “monotonically” with increasing noise level, although, again, large changes only occurred when the tones were almost completely masked by the noise. Even at the highest noise level used, DLCs for complex tones with no common harmonics were usually larger than DLCs for tones with common harmonics. The results suggest that the worse performance for tones with no common harmonics does not re-sult from intemal noise in the channcls conveying information from the periphery to the mechanism determining residue pitch. Rather, spectral differences between tones with noncoincident harmonics appear to have a distracting eflfect that impairs pitch discrimination.

1888 J. Acoust. Soc. Am., Vol. 89. No. 4, Pt. 2. April 1991

3:15

3PP8. Dichotic pitch discrimination. William A. Yost (Parmly Hearing Inst., Loyola Univ., 6525 N. Sheridan Rd., Chicago, IL 60626)

Dichotic Huggins pitch was produced by generating broadband noise stimuli with narrow sections of the noise (bandwidths of 2, 4, 8, 16, 32, 64, and 128 Hz) interaurally phase shifted (by 90*, 135*, or 180*). The center frequencies of the narrow dichotic bands to which the interaural phase shifts were added were varied to change the value of the Huggins pitches. The just discriminable differences (obtained in a two-altemative, forced-choice adaptive procedurę) in center frequency were determined for six listeners and at four (250, 400, 500, and 750 Hz) basc center frequencies in an attempt to describe the pitch discrimination of dichotic Huggins pitches. For some listeners, at base center frequencies of 250 and 400 Hz and when the interaural phase shift used to produce the pitch was 180*, discrimination was almost as acute for these dichotic pitches (A///=0.05%) as it was for diotically produced pitches. Pitch discrimination becamc morę difficult as thc center fre-quency inereased to 500 and 750 Hz, the bandwidth of the interaurally altcred band decreased from 128 to 2 Hz, or thc interaural phase shift was reduced from 180° to 135* and 90*. The results will be discussed in terms of models of binaural processing and the use of interaural vari-ablcs to segregate one sound image from other sound images. [Work supported by thc NIDCD and thc AFOSR.]

3:30

3PP9. Interaural correlation and the discrimination of one from two delayed sources. Irwin Pollack (Mental Health Rcs. Inst., Univ. of Michigan, Ann Arbor, MI 48109-0720)

In 1948, J. C R. Licklider provided a powerful methodology for exploring the binaural system in terms of the cross correlations of sig-nals presented to the two ears. The Licklider procedurę is here extended to the discrimination between: (I) a binaural source with single ongoing interaural delay and (2) a combination of Iwo binaural sourccs, each with its own delay, with the mcan delay equal to thc single delay. Accuracy of discrimination suffers at longer mean interaural delays. Accuracy of performance varies with the difference between the two delays of thc combination. The direction of change is determined by the contrast in interaural correlation. The results are consistent with a binaural system where interaural correlation is evaluated at stations specific lo the component interaural delay(s); where the strength of interaction between the component correlations dccreases with the difference in the component interaural delays; and where the associated intemal noise leve! inereases with longer interaural delays (Jcffrcss). Licklider’* meth-ods and his corrclational theory of binaural localization continue to provide a rich source of hypotheses and procedurcs for the analysis of binaural listening.

121st Meeting: Acoustical Sodety of America 1888