The View from the Auditory System

For much of the past 50 years, the main theoretical debate in the scientific study of speech perception has focused on whether the processing of speech sounds relies on neural mechanisms that are specific to speech and language or whether general perceptual/cognitive processes can account for all of the relevant phenomena. Starting with the first presentations of the Motor Theory of Speech Perception by Alvin Liberman and colleagues (Liberman, Cooper, Harris, MacNeilage, & Studdert-Kennedy, 1964; Liberman, Cooper, Shankweiler, & Studdert-Kennedy, 1967; Studdert-Kennedy, Liberman, Harris, & Cooper, 1970) and the critical reply from Harlan Lane (1965), many scientists defended “all-or-none” positions on the necessity of specialized speech processes, and much research was dedicated to demonstrations of phenomena that were purported to require general or speechspecific mechanisms (see Diehl, Lotto, & Holt, 2004 for a review of the theoretical commitments behind these positions). Whereas the “speech-is-special” debate continues to be relevant (Fowler, 2008; Lotto, Hickok, & Holt, 2009; Massaro & Chen, 2008; Trout, 2001), the focus of the field has moved toward more subtle distinctions concerning the relative roles of perceptual, cognitive, motor, and linguistic systems in speech perception and how each of these systems interacts in the processing of speech sounds. The result has been an opportunity to develop more plausible and complete models of speech perception/production (Guenther & Vladusich, 2012; Hickok, Houde, & Rong, 2011). In line with this shift in focus, in this chapter we concentrate not on whether the general auditory system is sufficient for speech perception but rather on the ways that human speech communication appears to be constrained and structured on the basis of the operating characteristics of the auditory system. The basic premise is simple, with a long tradition in the scientific study of speech perception: the form of speech (at the level of phonetics and higher) takes advantage of what the auditory system does well, resulting in a robust and efficient communication system. We review here three aspects of auditory perception—discriminability, context interactions, and effects of experience— and discuss how the structure of speech appears to respect these general characteristics of the auditory system. It should be noted that we include in our conception of the “auditory system” processes and constructs that are often considered to be “cognition,” such as memory, learning, categorization, and attention (Holt & Lotto, 2010). This is in contrast to previous characterizations of “Auditorist” positions in speech perception that appeared to constrain explanations of speech phenomena to peculiarities of auditory encoding at the periphery. Most researchers who have advocated for general auditory accounts of speech perception actually propose explanations within a larger general auditory cognitive science framework (Holt & Lotto, 2008; Kluender & Kiefte, 2006). Recent findings in auditory neuroscience provide support for moving beyond simple dichotomies of perception versus cognition or topdown versus bottom-up or peripheral versus central. There have been demonstrations that manipulation of attention may affect the earliest stages of auditory encoding in the cochlea (Froehlich, Collet, Chanal, & Morgon, 1990; Garinis, Glattke, & Cone, 2011; Giard, Collet, Bouchet, & Pernier, 1994; Maison, Micheyl, & Collet, 2001) and experience with music and language