Behavioral vs. Cognitive Views of Speech Perception and Production
Schlinger, Henry D., Jr., The Journal of Speech-Language Pathology and Applied Behavior Analysis
The study of speech perception and production has been dominated largely by cognitively oriented researchers and theorists. However, cognitive theories do not translate well into actually teaching speech and language. Because of its emphasis on behavior and its foundation of experimentally derived principles of learning, operant learning theory, on the other hand, has for more than 40 years been used to successfully teach speech and language, especially to people with a variety of speech and language disorders (e.g., Camarata, 1993; DeLeon, Arnold, Rodriguez-Catter, & Uy, 2003; Hegde, 1998; Hegde, 2007; Hegde & Maul, 2006; Johnston, & Johnston, 1972; Lancaster et al., 2004; Pena-Brooks & Hegde, 2007; Wagaman, Miltenberger, & Arndorfer, 1993). Although there is a fair amount of behavioral research on teaching verbal behavior to people with speech and language disorders, with few exceptions (e.g., Guess, Sailor, Rutherford, & Baer, 1968; Whitehurst, 1972; Whitehurst, Ironsmith, & Goldfein, 1974) behavioral scientists have not contributed much basic research on speech perception or production. What behavior analysis can offer language researchers and speech-language pathologists (SLPs) is a coherent and parsimonious interpretation of speech consistent with experimentally established scientific principles of learning that has immediate practical applications. The purpose of this paper is to illustrate a general behavioral approach to speech perception and production and to contrast it with the traditional cognitive account. The task of interpreting the traditional speech perception and production literature from a behavior analytic perspective is not as onerous as one might imagine, given that much of the research either incorporates relatively straightforward operant conditioning methods or methods that are amenable to an operant analysis.
Contrasting Views of Perception
Cognitive Views of Perception
Traditional treatments describe sensation in terms of the effects of stimuli on sensory receptors. Perception, on the other hand, has generally been referred to as how the brain interprets sensory experience, or more formally as:
...the process by which animals gain knowledge about their environment and about themselves in relation to the environment. It is the beginning of knowing and so is an essential part of cognition. More specifically, to perceive is to obtain information about the world through stimulation. (Gibson & Spelke, 1983, p. 2)
The main problem with such descriptions is that gaining knowledge, knowing, and obtaining information are inferred solely from observable behavioral evidence and, thus, do not add much to our understanding. Moreover, placing perception inside the brain, as some descriptions do, only moves the level of analysis further away from the actual behaviors involved when one speaks of perception. And talking about the brain as if it is doing something--perceiving--is an example of what I refer to as the brain-as-person metaphor because organisms, not brains, perceive, that is, behave.
Descriptions of the development of perception in infants are equally vague. For example, according to Gallahue (1989), "newborns attach little meaning to sensory stimuli" but very soon infants begin to attach meaning and attend to specific stimuli and to identify objects (p. 184). But what does that mean? These verbs do not refer to specific behaviors that can be studied but rather labels for a variety of behaviors. To understand what it means to attend, to identify or to attach meaning, we would need to observe what newborns actually do and the circumstances under which they do it when researchers use such terms. And we should not be surprised if the behaviors and circumstances vary considerably from instance to instance. Furthermore, such characterizations raise a more fundamental question: Why do infants begin to attach meaning or attend to specific stimuli? …