Page 54 - Summer 2015

HIstory of Psychoacoustics
Licklider (1956) also developed a “triplex” theory of hearing in which he proposed an autocorrelator for pitch processing. Pitch perception was studied extensively in the Netherlands and Germany by psychoacousticians such as Reinier Plomp, Burt de Boer, Frans Bilsen, and Ernst Terhardt. The autocor- relation approach of Licklider (and its later variations, Med- dis and Hewitt, 1991) can extract temporal regularity from a sound as a basis of pitch. However, there are equally success- ful models of pitch that are based on the spectral structure of a sound. Julius Goldstein, Ernst Terhardt, and Fred Wight- man each developed successful spectrally based models of pitch perception. As mentioned previously, the debate about spectralversustemporalaccountsofpitchperceptioncon- tinues today (Yost, 2009).
Neal Viemeister (see Table 1) developed the temporal mod- ulation transfer function (TMTF), similar to analogous functions in vision, to explain psychoacoustic phenomena associated with temporal modulation of sound (Viemeiser and Plack, 1993). This work paved the way for the current modulation filter bank models of temporal processing (Dau et al. 1996).
Psychoacoustics and Hearing Impairment: Audiology
Most of the work described in this history is based on nor- mal hearing. But, a great deal of psychoacoustic research is directly related to issues of hearing impairment. Many of the psychoacoustic procedures, data, phenomena, and theo- ries have been developed to diagnose and treat hearing loss. Data from listeners with hearing loss often shed light on the mechanisms of normal hearing.
The first audiometers to measure hearing loss were devel- oped at Western Electric in 1922 (e.g., Western Electric IA audiometer, which cost $1,500, only slightly less than a house at that time; Figure 10), and Fletcher coined the term “audiogram” at about the same time. Alexander Graham Bell spent a significant amount of his career developing hearing aids. Ray Carhart is often credited with being the “father of audiology” and starting the first audiology academic pro- gram at Northwestern University in the late 1940s. Before the term audiology was coined, the field was often known as "auricular training," but Hallowell Davis (“Hal” was an audi- tory physiologist and ASA Gold Medal winner and Presi- dent) thought auricular training sounded like someone who was taught to wiggle his/her ears.
Complex Sound Processing and Auditory Scene Analysis
A great deal of
psychoacoustic
research from
1920 to 1970 used
well-specified but
unnatural stim-
uli such as tones,
Gaussian noise, and brief transients in highly controlled ex- periments with very experienced listeners. By the late 1970s, psychoacousticians began to explore the limits of process- ing “simple” stimuli and to design stimuli and experimental conditions that better represented real-world sounds and listening situations. This was sometimes discussed as “com- plex sound processing” (Yost and Watson, 1987).
Although research on complex sound processing was pur- sued by many psychoacousticians, there was no overarch- ing theory or organizing principle to integrate the knowl- edge being accumulated and to make new predictions. This changed when a series of articles, chapters, and books ap- peared between 1988 and 1992 (Yost, 2014). The book by Al Bregman (1990), Auditory Scene Analysis, captured the essence of these other authors’ attempts at finding an orga- nizing principle for complex sound processing, and Auditory Scene Analysis captured the imagination of perceptual scientists in hearing as well as in perceptual and cognitive psychology.
Sounds from the various sources that make up an audi- tory scene interact physically and arrive at the ears as a single sound field representing the physical combination of the sounds from the various sources. The auditory pe- riphery uses biomechanical and neural processes to send a neural code to the brain representing the spectral/tem- poral features of that sound field. There are no peripheral mechanisms that process sounds as coming from individual sources. There is no representation in the neural code flow- ing to the brain that the scene may be one of a car driving by as the wind blows the leaves and a child giggles. Yet that is what we can perceive usually immediately and effortlessly. The sound is complex and the listener may be hearing some of the sounds for the first time, yet the auditory images are often vivid. These auditory images allow the listener to iden-
Figure 10. Western Electric IA audiometer, the first widely used audiometer for testing hearing impairment.
52 | Acoustics Today | Summer 2015