Page 46 - Spring2022

WILLIAM A. YOST
Hirsh’s and Licklider’s initial papers in 1948, Bill has explored the various conditions under which the MLD does and does not occur (Yost, 1988). The effect has been shown by various researchers for tones, speech, and other signals, using both interaural phase (or time) differences and ILDs, and even in temporal masking paradigms, in which the signal and noise are not presented concurrently. In addition to publishing many influential articles on vari- ous aspects of the MLD, Bill along with his friend and colleague Tino Trahiotis in 1998 organized The MLD: A Collection of Seminal Papers to commemorate the 50th anniversary of the Licklider and Hirsh papers and to high- light and celebrate the vibrant psychoacoustics community, many of whom contributed to our understanding of this interesting phenomenon. The image in Figure 5 was taken from the cover of this collection.
Bill’s more recent work has focused on the maximum number of spatially separated sound sources in an auditory scene that listeners are able to successfully process (Yost et al., 2018, 2019b). These studies have found that for talkers simulating a cocktail party or noisy restaurant auditory scene, the maximum size of the auditory scene appears to be four. More specifically, listeners were relatively accurate in both identifying and discriminating the total number of talkers and reporting talker locations when there were up to four talkers. Listeners could also judge loudness differences based on individual source levels when there were four or fewer sources. With five or more sources, dis- crimination of the total number of talkers and localization accuracy approached chance, and listeners tended to use overall level to perform the loudness difference task rather than individual source levels, indicating an inability to
“hear out” individual sources or streams.
Most recently, Bill and his team have been interested in auditory motion and the effect of head turns on sound source localization, with a focus on cochlear implant (CI) users who are well-known for being poor local- izers (Brown, 2018). It was established some time ago that head movements are integrally related to local- ization (Wallach, 1940). The work by Pastore et al. (2020) in this area established that head turns signifi- cantly improved localization abilities for single-sided deafened individuals implanted with a CI with their CI both off (monaural condition), and on (so-called bimodal listening condition).
In fact, auditory motion is a sorely understudied topic. One very good reason for this is the many technical chal- lenges and other difficulties that interfere with the ability to exert sufficient scientific rigor so that the results are generalizable while also maintaining ecologically valid conditions. Ever fearless, Bill undertook the challenge, and the result is a listening room at Arizona State University, Tempe, that has been custom designed and purpose-built for auditory motion experiments (Figure 6). The room is sound deadened and contains a custom chair that allows precise measurement and control of rotational velocity and an array of loudspeakers with custom software that allows sound source motion to be accurately simulated.
Using this facility, Bill has collected a trove of interesting data, most of which have been used in published studies on the relationship between localization, source movement, and listener movement (e.g., Yost and Pastore, 2019). One goal of this work was to establish how individuals can use spatial cues during motion. Interaural difference cues are inherently head-centric and thus change with head turns as well as with any other movement of the source or the listener. How then does a listener disentangle a relatively complex scene wherein both the listener and the sound source are moving? Supported by compelling data, Bill has argued in several papers that sound source localization is not a purely a psychoacoustic phenomenon but rather is based on an integration of input from several systems, including auditory, visual, and very likely vestibular (e.g.,
Yost et al., 2019a, 2020).
Figure 6. Bill’s sound insulated room with rotating chair and surrounding speakers for studying auditory motion perception.
46 Acoustics Today • Spring 2022