Page 8 - Fall 2011

VOICES AND LISTENERS: TOWARD A MODEL OF VOICE PERCEPTION Jody Kreiman Department of Head/Neck Surgery University of California, Los Angeles, School of Medicine Los Angeles, California 90095 and Diana Sidtis Department of Communicative Sciences and Disorders New York University New York, New York 10012 and Nathan Kline Institute for Psychiatric Research Orangeburg, New York 10962 “The wide distribution of voice recognition abilities across species, combined with the clear survival value of such abilities and their strikingly full-blown ontogenetic appearance, suggests that familiar voice recognition is evolutionarily very old.” Introduction As humans, we are exquisitely tuned to voices and all that they are capable of conveying (Table 1). On hearing someone speak, we quickly infer details about gender, age, education, and geographical background (Sebastian and Ryan, 1985). We listen for signs of inter- est, well-being, competence, and cooper- ation, or coldness, ineptness, and resist- ance. Along with these, mood, emotional conditions, personality, and psychologi- cal status are simultaneously assessed by the listener, with varying accuracies. These speaker characteristics constitute a very large, complex array and pose huge challenges to analytic approaches. Not least important among the characteristics listeners extract from voices is the identity of the person who is speaking. The person may be someone familiar; or, much less commonly, we may try to identify a stranger, for example in a forensic situation. In this paper we will describe some of the important differences between these two classes of stim- uli—familiar and unfamiliar voices—and the cognitive and neuropsychological processes used in their perception. We then present a preliminary model of the manner in which lis- teners tackle each kind of information. taking into account underlying brain structures involved in these disparate processes. Finally, we explore the implications of our model for measurement of quality in the voice clinic and elsewhere. Which came first: Familiar or unfamiliar voices? Unfamiliar voices surround us in life, from the sound of the cashier greeting us at the market, to students talking in the hall outside a classroom, to the voices of other patrons convers- ing in a background of chattering and cheering at a sports event. When we pay attention to such voices, they can provide substantial amounts of information about the speaker, as noted above, and as a result it is easy to assume (as we ourselves have done in the past) that the unfamiliar voice is somehow the basis of the percep- tual processes used to extract information from all voices. After all, we reasoned, every voice was unfamiliar before it was familiar, so logically familiarity develops out of unfamiliarity, which implies that the unfamiliar is foundational. In the beginning was the familiar voice A substantial body of evidence sug- gests that the assumption that unfamil- iar voices are fundamental is fundamen- tally wrong. First, we note that the abili- ty to recognize a familiar voice (and especially the voice of a parent, off- spring, or mate) is very widespread among animals. Many, many species, including deer (Torriani et al., 2006), sheep (e.g., Sebe et al., 2010), wolves (Goldman et al., 1995), mares (Wolskia et al., 1980), many marine mammals (e.g., Insley, 2001; Pitcher et al., 2010), rodents (Fuchs et al., 2010), bats (Voigt-Heucke et al., 2010), amphibians (Bee and Gerhardt, 2002; Simmons, 2004), and birds ranging from penguins (e.g., Jouventin and Aubin, 2002) to parrots (Berg et al., 2011) also recognize the familiar voices of their kin. Recognition often begins very early in life, or even immediately; for exam- ple, the developing human fetus has been shown to recognize the voice of its mother (Kisilevsky et al., 2003). Scientists have only begun to appreciate the social complexity and sophistication of these behaviors. Recent studies reveal that seal mothers time their departure for food gathering to coin- cide with successful voice recognition by their pups, so that reuniting on their return will be successful (Charrier et al., 2001). In comparison, mother evening bats recognize the Voices and Listeners 7