The perceptual grouping of a four-tone cycle was studied as a function of differences in fundamental frequencies and the frequencies of spectral peaks. F.ach tone had a single formant and at least 13 harmonics. In Kxperimcnt 1 the formant was created by filtering a Hat spectrum and in Experiment 2 by adding harmonics. Fundamental frequency was found to be capable of controlling grouping even when the spectra spanned exactly the same frequency range. Formant peak separation became more effective as the sharpness (amplitude of the peak relative to a spectral pedestal) increased. The effect of each type of acoustic difference depended on the task. Listeners could group the tones by either sort of difference but were also capable of resisting the disruptive effect of the other one. This was taken as evidence for the presence of a schema-based process of perceptual grouping and the relative weakness of primitive segregation. KKNL' MK Nous avons cludic le regroupcmenl perccptuel d'un cycle de qualre sons en fonction des differences dans les frequences londamentales et les frequences de pics spectraux. Chaquc son avail un scul forniant et au moins 13 hyrmoniques. Dans ('experience 1 Ic formant ctait eree par le filtrage d'un spectre plat et dans I'cxperience 2 par I'addition d'harmoniques. Nous avons trouve que la frequence fondamcntale etait capable de controlcr le regroupement meme quand les spectres traversaicnt exactement la meme elenduc dc frequences. l.a separation entre les pics dc formant devenait plus efficacc a mesure que I'acuitc (Tamplitude du pic relatif au picdcstal spectral) augmcntail. L'elfet dc chaque type dc difference acoustique dcpcndail de la tachc. Les sujets pouvaienl regrouper les sons par Tune ou Paul re des differences et elaient aussi capables dc rcsister a 1'effet pcrturbateur dc l'autre difference. Ccci a etc considere commc une evidence de la presence dun processus schema-dependant du regroupement percepluel et dc la faiblesse relative de la scgre gation primitive. When a repealing cycle of tones, some in a higher frequency region and some in a lower region, is repeated rapidly, the listener hears two streams of sound, one composed of the higher tones and the other of the lower tones (Bo/.zi & Vicario, 1960; Miller & Heise, 1950). This phenomenon has been referred to as auditory stream segregation (Bregman & Campbell, 1971). Stream segregation makes it hard to hear patterns that cross the two streams, be they melodic patterns (Bregman &
The fusion of two amplitude-modulated (AM) tones presented simultaneously was studied. Subjects were presented with an AM tone (A) followed by a copy of itself (B) which was accompanied by another AM tone (C). In different experiments, the subjects were asked either to rate how clearly they could hear Tone B in the BC mixture or whether Tone B was present or not. The stronger the fusion of Tones Band C, the harder it was to "hear out" Tone B. It was found that fusion was strongest when Tones Band C were modulated at the same rate; segregation curves were obtained for the degree of mismatch of modulation frequency.In the usual complex listening situation, the pressure waves arising from many sources mix together as they enter the ear. An important part of pattern recognition is the recovery of separate descriptions of some of the individual sources of sound that have contributed to the mixture. Therefore, the study of auditory segregation and its opposite, fusion, is the study of the methods used by the auditory system to decompose mixtures of sounds and to group their component frequencies or features together into appropriate descriptions (Bregman, in press).-There are two aspects of this grouping, which we can refer to as sequential grouping and simultaneous grouping. In sequential grouping, we must decide on the sequence of acoustic components that arise over time from the same source. In simultaneous grouping, the set of spectral components arriving at anyone instant must be decomposed into the contributions of the individual sources.It appears that there are heuristics for both sequential and simultaneous grouping that collaborate and compete in order to arrive at a correct decomposition of the incoming signal (Bregman, 1978(Bregman, , 1981. In simultaneous grouping, or fusion, many useful clues arise from the behavior of spectral components over time. It has, for example, been shown that frequency components with synchronous onsets and offsets tend to be heard as the spectral components of a single sound (Bregman & Pinker, 1978; Dannenbring & Bregman, 1978;Rasch, 1978). Generally speaking, the factors that promote the fusion of spectral components can be described loosely by the Gestalt term, "common fate" (see Koffka, 1935). If spectral components are doing the same thing at the same time (e.g., going on and off in synchrony) they probably "belong Copyright 1990 Psychonomic Society, Inc.together" as parts of the same signal (Bregman, 1978(Bregman, , 1981.One natural example occurs when a sound (e.g., human speech) is heard against a background of other sounds. The independent sound sources-target and background-will tend to have their own independent patterns of rise and fall in intensity. Evidence from experiments on "comodulation release from masking" has shown that the auditory system can use the correlations in the intensity variations of the different parts of the background spectrum to shield the target from masking (see Grose & Hall, 1989;Haggard, Harvey, & Carlyon, 1985;Hall, 1986;Hall, Haggard, ...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.