Verbal behavior by B.F. Skinner: contributions to analyzing early language learning.
Two pieces of literature appeared quietly and without fanfare in 1957. Each book unalterably affected how we have come to view language, human behavior, and language learning. In 1957, Noam Chomsky published Syntactic Structures (1957), his germinal work that established the foundations of psycholinguistics. This work, in combination with a second publication, Aspects of the Theory of Syntax (Chomsky, 1965), broadly influenced research in linguistics and the theoretical relationships between the mind and language. Together, they represented a firm anchor point on one end-the nativistic end-of the philosophical continuum established in 1957.
With regard to explaining language development, one of the more publicized features of this theory was the Language Acquisition Device, or LAD. Although there was no intention to correlate this "device" to any underlying neurological structure, Chomsky proposed the LAD as the presumed innate mechanism in the human brain (or mind) to explain the apparent ease and rapidity with which children acquire language. Although the concept of the LAD has been modified through the years, in the 1960s it became the widely accepted explanation for children's acquisition of language, largely removing caregivers from any active role in their children's language abilities.
The opposite anchor point-the behavioral end-on this philosophical continuum was established that same year, when B.F. Skinner published Verbal Behavior (1957). (See Hegde, in this issue, for a comprehensive review of Skinner's Verbal Behavior.) In contrast to Chomsky's approach, Skinner proposed an analysis of verbal behavior based on a natural science account of behavior that had evolved since the earlier publication of his The Behavior of Organisms (1938). In Verbal Behavior, Skinner applied a functional analysis approach to analyze language behaviors in terms of their natural occurrence in response to observable environmental circumstances and the measurable effects they have on human interactions. In this view, language was characterized as the result of, as opposed to the reason for, complex human behavior. The complexities of language do not exist prior to or independent of human behavior; instead the complexities of language behavior reflect our capacity to respond verbally to the complex and subtle intricacies inherent in human experiences and interactions.
In Verbal Behavior, Skinner (1957) did not emphasize explaining the nature of early language development; perhaps the explanation seemed obvious to him from the fact that he had invoked an operant model in the overall analysis. Verbal Behavior primarily focused on an explication of the causal variables for the verbal interactions of accomplished speakers and listeners whose learning histories were in place and preceded the verbal behaviors in question-in essence, adult speakers. It is perhaps unfortunate that Skinner emphasized this level of analysis-adult verbal interactions-throughout Verbal Behavior as it ostensibly posed difficulty for some who have tried to explain the development of language under his model. Chomsky critiqued Skinner's functional analysis in a book review that many found puzzling due to several lengthy criticisms put forth by Chomsky that did not relate to principles or concepts contained in Verbal Behavior. Specifically, Chomsky criticized Skinner for proposing imitation and conscientious parental tutoring as the major explanations for language development. Chomsky also took Skinner to task for his supposed reliance on memorized Markovian chains as an explanation for grammar. However, none of these elements were included in Skinner's model. Over the years, attempts to describe how Skinner's model might be applied to children's language development, including recent sources (Owens, 2005; Berko Gleason, 2005; Hulit, 2006), have continued to rely on perpetuating these misconceptions contained in Chomsky's (1959) "bewildering" (MacCorquodale, 1970, p. 83) critique of Verbal Behavior. However, others (McLaughlin, 2006; Winokur, 1976) have attempted to provide integrated explanations of early language
learning that are more consistent with the original principles set forth in Verbal Behavior (1957).
This paper will revisit some basic terms and concepts central to understanding the functional analysis set forth by Skinner (1957). In the context of early language learning, these elements will be explored as to how they relate to our understanding of language learning in young children, with particular emphasis on the foundations laid down in the first months and years of life. Finally, Skinner's basic classifications of verbal behaviors-verbal operants-will be described and correlated with traditional concepts and research from language development literature.
The Foundational Contingency: Infant-Caregiver Interactions
According to Skinner (1957), a functional analysis of verbal behavior must build on the fundamental task of describing the behaviors of interest and then extend itself to explaining the causes of those behaviors. Such an analysis attempts to identify the variables that cause certain verbal behaviors to occur in certain circumstances. In the same way that one might say combustion is increasingly likely as a function of the combined presence of fuel, air, and heat, one might say that certain verbal behaviors are dependent upon, or occur as a function of, certain causal variables. The interrelationship of multiple and complex variables that influence verbal behavior comprises the contingencies that are established through a learning history-even though sometimes quite momentary-in which the setting, a behavior, and the consequences that have attended to the behavior influence the likelihood that the behavior will occur again in similar circumstances.
In the context of language learning, analyzing the independent variables-the causes-is no simple proposition. To say that human beings are complex in their functioning is probably less a platitude than an understatement. Furthermore, developing humans-from infancy through childhood-are especially complex due to the multitude of concurrent changes that occur so rapidly during a relatively short period of time. There is the underlying physiological and neurological maturation that provides for an array of increasingly refined sensory and motor capabilities. Supporting and nurturing all these physiological changes requires the social interactions associated with caregivers' routines and activities. Eventually, the child's social interactions expand to an entire supporting cast of individuals with whom the child will interact in various settings and roles. In addition, the increasingly refined motor skills and the expanding and sophisticated social behaviors set the stage for new and more complex verbal behaviors to evolve. This is a process that is not only multifaceted; the various facets intertwine with and drive each other forward in the overall process.
Although Skinner's analysis is focused on the behavioral layer in all this development, it acknowledges that there is a myriad of accompanying genetic, neurophysiological and social influences (Skinner, 1974). They simply do not occur at a level that can be included in this particular analysis. They no doubt exist and their potential roles would be addressed where necessary, but they are for the purposes of a functional analysis the inaccessible undercurrents that are left to other fields to explore and analyze. The functional analysis for the present purpose will confine itself to defining the roles of the antecedent setting events, consequences, and their influence on the onset and development of communication behaviors in infants and language learning in children.
Antecedents for Infant Behaviors: The Stimulus Setting for Language Learning
The antecedents of behavior are those stimuli or stimulus events that set the stage for behavior. Every moment that we experience contains a multitude of external stimuli that can be detected and internal stimuli that can be sensed at some level. As a result, any single stimulus or combination of stimuli has the potential to serve as the setting event for a response. If that response proves to be useful in some way that has reinforcing consequences, the stimulus that was present when reinforcement occurred becomes a discriminative stimulus for similar behaviors in the future. In our natural interactions in the "real world," it is likely that more often than not, the combinations of stimuli that comprise a stimulus setting are more arbitrary, random, and coincidental than they are planned, staged, or contrived. Things just come together at times and present us with the occasion to respond in some way.
External Stimuli. The human infant is immersed in stimuli from the beginning--even prior to birth. Although the fetal surroundings are darkened and relatively silent, it has been determined that the cochlea and auditory nerve have developed at around 26 weeks of gestation and the fetus is subsequently capable of detecting sound (Locke, 1993). The auditory system does have stimuli available to it, primarily in the form of the internal sounds of the mother's heartbeat and blood flow. However, the mother's voice is also detectable--albeit muffled-at conversational levels above this ambient noise. In this way, the fetus has several weeks of hearing at least the prosodic character of the mother's voice, a sound that will be immediately associated with the most significant events in the first moments and days of the infant's life following birth (Locke, 1993).
At birth, the events that essentially integrate the social and functional nature of communication are present almost immediately. There is a significant convergence of human interaction in which caregivers begin to consistently (some might say constantly) provide for their infants' very survival. Human infants are uniquely dependent on their caregivers for survival from the start and for several years following. Caregivers in turn provide for these needs through various interactions and routines that take on a strongly social character.
It is critical to note that in this convergence (e.g., feeding sessions) the newborn's first experiences are both life-sustaining and social by nature. As noted earlier, the infant's auditory system is fully functioning at birth and the infant is probably already familiar with the mother's voice patterns. Although his or her visual acuity is limited, stimuli within a foot are in fact visible to the infant. By the nature of most feeding sessions, the infant nestled in a feeding position at the breast, or in some similar placement if bottle fed, places the caregiver's face and voice in perfect proximity to be intimately associated with feeding (Locke, 1993). This proximity and intimacy will persist as caregivers will need to assist with feeding for some time. As a result, this interaction, so vital for simple survival, consistently associates the caregiver's face, voice, and physical contact with survival needs. As the earliest and most familiar stimulus setting for developing infants, this convergence of attention and social stimulation coincides with their most critical period of rapid brain growth and neurological maturation. This is a crucial principal in understanding the learning process that ensues. As Locke surmised, "One can hardly think of a developmental circumstance that would more favorably affect acquisition of complex behavior" (Locke, 1993, p. 264).
Although most of the infant's behaviors-eye gaze, hand movements, facial movements, and vocalizing-are primarily reflexive at first, they do coincide with the earliest most important routines that by their nature include the caregiver's face and voice as conspicuous elements in the overall setting event. Most caregivers, at least in Western societies, are inclined to respond to their infants with eye gaze behaviors, hand play, and speech (often in some form of "motherese") while the infant is being cuddled, bathed, dressed, and fed. Again, these earliest social experiences, repeatedly associated with the caregiving routines, become the foundational and consistent setting events for infants interacting with the world around them. Simultaneously, caregivers are at the center of their infants' experiences as the most important source for survival and stimulation. As maturation progresses, infant motor behaviors-hand movements, facial expression, and so forth-become less reflexive and more voluntary and serve to initiate interactions in the context of routines and new forms of interaction with the caregivers.
Beyond those earliest intimate experiences, with time the child's developing motor system eventually permits exploring a wider territory, which includes a vast array of smells, textures, temperatures, shapes, colors, sizes, locations, actions, objects, persons, and relationships among them. This is hardly a complete list of the possible physical and functional features that could present themselves to any of us given the complexity of our world. As children grow, they become more adept at handling and exploring objects, more aware of the diverse events around them, and more responsive to the ways these all interact and relate to the significant people in their lives. All these very complex stimuli-both subtle and significant-become salient setting events for the development of an increasingly complex array of verbal behaviors. In other words, it would follow that the increasing complexity of the child's verbal behavior correlates with his or her responses to increasingly complex stimuli and relationships in the environment.
Internal Stimuli. Not all stimuli that contribute to setting events for verbal behavior are external. From birth, the infant experiences various states that presumably represent internal stimuli. Although the internal physiological stimuli cannot be directly observed, the process that prompts them to occur can be. The time period in which the infant is deprived of some basic need that results in the internal stimulus can be observed and measured. The length of time a child goes between feedings, the amount of time a child is uncovered and exposed to a chilly draft, or goes without social interaction would each be indicative of the degree of the corresponding internal states of hunger, cold, or loneliness (social deprivation) that may serve as motivation for the child's subsequent behavior. Deprivation states can be real, accessible, and measurable variables when analyzing internal stimuli as antecedent events for certain behaviors in children.
In early months, the overt behaviors may be purely physiological, reflexive, and strictly vocal in nature-for example, the hunger cry. Interestingly, caregivers respond differentially to negative vocalizations (whining, fussing, crying, and so forth) as opposed to positive vocalizations (cooing, babbling, laughing, and so forth). Negative vocalizations prompt more attempts at changing the infant's physical state through tactile or positional changes. Caregivers tend to respond to positive vocalizations with vocal-verbal behaviors (Keller & Scholmerich, 1987). With the infants' development of more precise verbal (as opposed to vocal) behaviors, a repertoire of overt responses that are more specific to different deprivation states and their underlying internal stimuli becomes possible and it would follow that caregivers would become more precise in responding verbally to address their children's internal states (pain, hunger, cold, etc.).
Consequences: The Nature of Reinforcement in Language Learning.
The other major element in the foundational contingency for language learning described by Skinner (1957) was the nature of consequences. On one level, that reinforcement probably plays a role at some level in verbal behavior would appear to be quite straightforward. Most may have a basic understanding of what constitutes reinforcement. However, a purely academic understanding that simply equates reinforcement to food and praise, or a knowledge of how clinicians explicitly and systematically reinforce target skills may hamper our ability to recognize the subtlety in caregiver-infant interactions. Appreciating the intricate strands of a reinforcement history that link the potency of the earliest experiences with the most basic reinforcers to the myriad of fleeting moments of social reinforcement that follow during the day-to-day caregiver-infant interactions is essential to understanding how language evolves from the earliest germinal moments in an infant's life.
Primary reinforcers. Reinforcers that have survival value--food, warmth, social contact-are considered primary reinforcers and are a normal part of care provided to most infants. Certainly the simplest and most obvious primary reinforcer is food. As with the youngest in any species, food plays a vital role as a primary reinforcer with survival value for human infants. However, it is perhaps even more crucial because human infants are entirely dependent on caregivers to provide this life sustaining primary reinforcer a number of times each day for several years. This earliest contingency between caregiver responses and infant needs is vital to the infant's survival and development. However fundamental it may be, providing for the infant's survival is merely the first level of interconnection between infants and their caregivers.
Secondary reinforcers. Secondary reinforcers are those stimuli or events that have the potential to reinforce behaviors due to their prior association with primary reinforcers. If a secondary reinforcer is established by being conditioned through association with a primary reinforcer, the social interactions noted previously that cause caregivers' faces, speech, and physical contact to coincide with the delivery of food, by definition, should establish potent secondary reinforcers for an infant from the first day of life. Through this frequent association, the caregiver's presence, attention, facial expressions, eye contact, touch, speech, and overall interaction are established from the beginning as powerful secondary reinforcers. Caregivers' contingent attention and responsiveness to the needs and behaviors of their infants are predictive of the infant's physical, social, intellectual, and emotional development as well as their eventual language development (Dunham, Dunham, Hurshman, & Alexander, 1989; TamisLamonda, Bornstein, & Baumwell, 2001).
Any behavior on the infant's part that evokes one or more of those social response elements from a caregiver-a touch, speech sounds, eye gaze, or a smile-in response will be effectively reinforced. This observation-the potency of the caregiver's natural behavior to serve as a secondary reinforcer as a result of this convergence-is perhaps most crucial to understanding the dynamics of early learning so essential to interpreting and extrapolating Skinner's Verbal Behavior to early language learning. As most would agree, it has been documented that infants are in turn responsive to caregivers' presence, interaction, and attention in many domains of learning (Kuhl, 2004).
Natural consequences. Most caregivers are naturally responsive to infants' noncry vocalizations. In turn, infants appear to learn that their caregivers do respond and come to have expectations that they will respond contingently. This reciprocity and infants' expectation connecting infants' vocalizations and caregivers' contingent responses suggests the reinforcing nature of these interactions (Goldstein, Schwade, & Bornstein, 2009). It has been demonstrated that infants' vocalizations can be influenced by social reinforcement in controlled experimental studies (Rheingold, Gewirtz, & Ross, 1959; Routh, 1969; Todd & Palmer, 1968; Wahler, 1969; Weisberg, 1963). Furthermore, it has been shown that the frequency of infant vocalizations varies depending upon the nature of engagement with their caregivers (Keller & Scholmerich, 1987). At least one experimental study has demonstrated that infants' can be conditioned to produce different responses based on differential social reinforcement (Routh, 1969). It should be noted that caregivers do not need to be conscious of providing reinforcement for these communicative behaviors; it occurs merely as the result of the natural interactions with their children. Infant behaviors that are attended to and effectively responded to by the caregiver are naturally reinforced by the immediacy and relevance of the caregiver's response. This is the essence of natural consequences, a critical concept in Skinner's analysis that has been missed or misconstrued in many attempts to review, critique, or interpret Verbal Behavior (1957).
Selective reinforcement. Another closely related concept, selective reinforcement, has been misinterpreted by some as well. A number of sources (Berko Gleason, 2005; Owens, 2005; Hulit, 2006) attempting to describe Skinner's model in the context of language learning, have portrayed selective reinforcement as a conscious, even conscientious, process painstakingly applied by caregivers. Again, this interpretation seems to be related to Chomsky's review of Verbal Behavior in which he contended that there is no evidence that caregivers use "slow and careful" reinforcement applied with "meticulous care" to teach language to their child (Chomsky, 1959, pp. 39,42,43). However, this is simply another one of Chomsky's "straw men" in that Skinner never indicated such requirements. More than 30 years after he published his Verbal Behavior, Skinner (1988, p. 486) wrote that:
Chomsky and others often imply that I think that verbal behavior must be taught, that explicit contingencies must be arranged. Of course, I do not, as Verbal Behavior makes it clear. Children learn to speak in wholly noninstructional verbal communities. But the contingencies of reinforcement are still there, even though they may be harder to identify. (p. 486)
Perhaps some confusion stems from the distinction between reinforcement as part of an experiment and as it occurs in the natural environment. In Routh's (1969) study, different classes of infant vocal responses--consonant--like sounds, vowel-like sounds, and all vocalizations--were chosen by the study design for the purpose of determining if infants' vocal behaviors were responsive to differential reinforcement. The specific vocal behavior that was reinforced in each infant in fact increased selectively. This was a useful and purposeful distinction; the isolated increase in the specified vocal behavior being reinforced demonstrated the effects of selective reinforcement. However, in this demonstration the vocal behaviors were chosen by the investigator as part of the experimental design. Selective reinforcement under normal circumstances is the natural result that follows when one response survives and is strengthened because it is more effective than another. In the natural environment's differential reinforcement-intentional or unintentional--is contingent on the value of a response and it "selects" the response by strengthening its future probability.
Selective reinforcement is analogous to the natural selection that perpetuates useful variations in traits within a species. As Skinner pointed out in About Behaviorism, in the same way that "accidental traits, arising from mutations, are selected by their contribution to survival, so accidental variations in behavior are selected by their reinforcing consequences" (1974, p. 114). When reinforcement is more likely to follow an effective response, it selectively strengthens that response over less effective responses. The caregiver does not need to be conscious of "selecting" one response over another; children's responses are selectively reinforced by virtue of communicating more effectively and evoking favorable outcomes. Caregivers do not wake up each morning planning which behaviors they will reinforce and what schedule of reinforcement they will use that day; they wake up hoping to help their offspring thrive, learn, and be happy. Both natural consequences and selective reinforcement are central aspects of the subtle consequences that determine children's communication behaviors, including those that will become integrated into the lasting verbal behaviors that comprise language.
Summary. The basic elements of Skinner's functional analysis of verbal behavior (1957) include the stimuli--external and internal-that set the stage for verbal behavior and the consequences that affect its reoccurrence. For the young infant, the environment is generally composed of those stimuli that either emanate from the caregiver-speech, touch, eye contact, facial expression, as so on-or occur as a result of actions by the caregiver-change in position, proximity to other stimuli, presentation of food, and so on. The infant's frequent, consistent, and contingent interactions with the caregiver while having primary survival needs addressed automatically keys in the caregiver's attention as a powerful secondary reinforcer that occurs naturally and is capable of selectively strengthening various infant responses.
A Functional Analysis of Verbal Behavior: Correlates in Early Language Learning
Skinner (1957) developed his model based on verbal behavior as a conspicuous and ubiquitous example of complex social behavior. He defined verbal behavior as a social behavior that is "reinforced through the mediation of other persons" (p. 2)--an especially fitting description of early language learning. His interlocking verbal paradigm captured the reciprocal nature of social interaction that is intrinsic to human verbal behavior. Applying this framework, he then illustrated the types of verbal behaviors-verbal operants--that occur in episodes or exchanges based on their circumstances, each speaker's behavior, and the consequences that follow. Different contingent relationships serve to functionally define different primary verbal operants. Each category of verbal operant represents a set of verbal behaviors that tend to occur under similar setting events and are reinforced in characteristic ways. Beyond the primary verbal operants, autoclitic behaviors represent the essence of "grammar" and more in Skinner's analysis (see Hegde in this issue). Along with some basic concepts central to the analysis, several verbal operants and their relevance to early language learning will be described. Correlated traditional concepts and terminology will be integrated into the discussion of each.
Concepts Central to a Functional Analysis
Unit of Analysis
Description: In analyzing verbal behavior, especially in the context of early language learning, it is important to note that Skinner's unit of analysis-the verbal operant-was functional, not structural. It was more important for him to consider the settings and the consequences of verbal behavior than its formal structural characteristics (i.e., noun, noun phrase, declarative). With infants, of course, the initial level of behavior is more vocal than verbal. Vocal behaviors that evoke caregiver attention and interaction evolve from cooing, babbling, and jargon that increasingly contain adult-like speech elements, including syllable patterns and intonational contours, and evoke verbal responses from caregivers (i.e., motherese). As a result, infants' vocal behaviors gradually transition toward verbal behaviors (i.e., the first word).
With children, the nature of the verbal operant will evolve over time.
In addition, the size of the functional verbal operant will expand over time. As the child becomes capable of coordinating longer motor sequences in his or her speech, his or her utterances can orchestrate more relations and address social agendas more deftly as they expand, for example, from mere grunts and gestures to "Cookie" to "More cookie" to "Two big cookies" all the way to "I can tell from how they smell that you make the best chocolate chip cookies!"
Correlates in Early Language Learning: Traditional literature includes few, if any, close correlates to Skinner's functional unit of analysis. Dore's "Primitive Speech Acts" (Dore, 1974) and "Conversational Speech Acts" (Dore, 1986) perhaps come closest due to their pragmatic basis. Given the preceding example of how a "functional unit" might evolve from primitive grunts and gestures to quite sophisticated and diplomatic indirect requests, some might be led to equate Skinner's unit of analysis with the pragmatic concept of "purposes," such as described in Speech Acts theory (Searle, 1969). It is important to remain aware that Skinner used the term "function" to refer to the causal explanations of verbal behaviors, as when noting that a certain response seems to be causally related to certain circumstances and a past history of reinforcement. Invoking the mentalistic concept of "purpose" as is done in pragmatic analyses makes that model different from Skinner's.
Interlocking Verbal Episodes
Description: A vital aspect of Skinner's model was casting the functional analysis in terms of interlocking verbal episodes intended to capture the "back and forth" reciprocal nature of verbal behavior between a speaker and listener whose roles frequently interchange throughout the interaction. In most cases, there are at least two participants in a verbal exchange. Each alternately serves as the speaker and the listener. As the exchange progresses, the verbal behavior of the first serves as a discriminative stimulus ([S.sup.D]) and possibly a reinforcer ([S.sup.r]) for the second. In turn, the second individual may respond verbally to the verbal SD produced by the first person. The verbal behavior of the second person may serve to reinforce the first person and serve as a verbal SD that evokes a subsequent response from the first person and so on until the series of verbal exchanges (i.e., the conversation) terminates.
In early interactions, episodes are frequently prompted by the caregiver with eye contact, physical contact, facial expressions, or speech ("motherese"). Recent studies have continued to demonstrate the significant role that the caregiver's presence and interaction can play in influencing infant attention, vocalization, reciprocity, and even enhancing their discrimination of the speech around them (Gartstein, Crawford, & Robertson, 2008; Goldstein, Schwade, & Bornstein, 2009; Keller & Scholmerich, 1987; Kuhl, Tsao, & Liu, 2003). Mothers adjust their behavior to their children's and children as young as 9 and 13 months make active adjustments to their caregiver behaviors based on their developmental needs to evoke more meaningful responses from the mother (Tamis-Lamonda Bornstein, & Baumwell, 2001). These observations suggest that even prior to the first birthday interactions that fit the verbal interlocking paradigm have already become established to enhance language growth "in the context of responsive social exchanges between caregivers and children" (Tamis-Lamonda, Bornstein, & Baumwell, 2001, p. 763).
Correlates in Early Language Learning: For a number of years, the literature has described the occurrence of phenomena that illustrate the concept of the verbal interlocking paradigm. Beginning with proto-conversations in caregiver-infant feeding interactions (Bateson, 1971; Locke, 1993) and the earliest forms of conversational interaction (Dore, 1986), caregiver-child discourse has been recognized as a rich source of teaching and learning about language. These phenomena represent important interactions that could be captured as interlocking verbal episodes and analyzed for the social and verbal transactions that occur through them. A cumulative history across ongoing exchanges between caregivers and children would probably allow an analysis of the diverse set of subtle teaching tools used by caregivers that may not even be apparent to the caregivers.
Hart and Risley (1999) collected monthly recordings for 30 months of the interactions between 42 children and their families. In their analysis they discovered surprisingly rich opportunities in families for most children to be exposed to language models. They found that an average of 700-800 utterances per hour was produced by people within hearing distance of a child. The average of utterances directed by the parent to the child was 300-400 per hour. The parents' language to others contained richer vocabulary and longer clauses and used declarative statements 60% of the time. In contrast, the parents used more questions and directives with their children, averaging 83 questions per hour. Hart and Risley analyzed much of their data in terms of "episodes," defined as occasions in which families and children directed social behavior toward each other terminated with 5-second boundaries of no social behavior. An average of 96 interactional episodes was recorded each hour and, interestingly, more than half of the episodes were initiated by the children. However, social responsiveness remained constant; in episodes containing utterances by both parents and children, the ratio of utterances was 50:50 throughout the 2 year period of learning to talk. These findings document extensive reciprocal and symmetrical interactions that correlate to Skinner's interlocking verbal episodes and occur frequently throughout a child's experience with adults, forming an important foundation of early language learning.
The Influence of Stimulus Control
Description: When a response has been reinforced in the presence of a stimulus, the stimulus potentially gains the ability to influence or control the likelihood of that response occurring in the future. The very first productions of "mama" or "dada" may or may not be verbal operants ("intentional" or "meaningful"), but they are nonetheless fortuitous. These two syllable pairs occur frequently in the early sound repertoire of children babbling in most every human language, possibly due to their underlying connection to oral-motor actions involved in feeding. The younger the infant when these syllable pairs are first heard, the less likely they are the true "first words." Nonetheless, at any age, mothers and fathers respond differentially to those two productions. When those syllables occur in their absence, it is not likely that others respond quite as enthusiastically, especially in the early months. Over time, however, after being reinforced by the relevant parent, those syllables are produced more frequently and differentially depending on who is immediately present. Once the child produces "mama" and "dada" reliably in their presence, mother and father have become the controlling stimuli for those responses.
Stimulus discrimination and stimulus generalization. Stimulus discrimination occurs when a restricted set of stimuli exclusively controls or influences a response, apparently due to a relatively narrow set of controlling or "defining" features. Conversely, stimulus generalization occurs when different stimuli that are similar in some defining manner become controlling stimuli for a response.
Initially, for a young child "mama" carries the sense that there is only this one person in his or her world who deserves this moniker. That response initially exhibits strong stimulus discrimination. In most cases, there is one adult female present to respond to and reinforce the early random forms of "mama." However, at some point, perhaps following more exposure to additional adult females, stimulus generalization occurs and soon every adult female is called "mama" or "mommy." Eventually, through a counterbalancing process of generalization and discrimination, the child learns that only certain adult females qualify as "mommies" and, at least implicitly, the child comes to reserve the "capitalized" version, "Mommy," for his or her own. He or she learns that there is a broader category of "mommies" and a specific instance of "Mommy."
Correlates in Early Language Learning: Children's vocabulary learning in language development has always been of interest to researchers. The various responses children produce as they attempt to label and categorize people, objects, and events around them gives us that special sense of having a "window into their minds." In traditional literature it has long been acknowledged that children only gradually learn to use words according to the adult conventions or boundaries. In the process, children have been described as producing "overextensions"-extending the use of a word beyond its conventional meaning (Bloom, 1973). A familiar example used to illustrate overextensions is children calling every creature with four legs, fur, and a tail, a "doggy." With experience and corrective feedback, children's overextensions are remediated by their verbal community. However, their occurrence provides a straightforward illustration of the behavioral process of stimulus generalization, albeit over-generalization with respect to the conventional meanings.
Theories on children's cognitive processes as they develop their concepts and word meanings have focused on hypothetical mental processes that may underlie these phenomena. The prevailing notion is that the child is hypothesis testing in attempts to develop the meaning of a word (Brown, 1958). Some theories have emphasized the child's apparent attention to perceptual features (Clark, 1973) and others have stressed that children base their conceptual meanings for words on the central or core function of an object (Nelson K. , 1974). In a functional analysis, it may turn out that these aspects are operative as different levels of controlling stimuli across individual children, but behaviorally a child will only demonstrate conceptual behavior if his or her response generalizes within a group of stimuli that are similar in some defining way and discriminates across groups of stimuli that differ in defining ways.
Contingency Shaped Behavior
Description: Contingency shaped behavior consists of behaviors that evolve through a gradual process of differential reinforcement. Very few, if any, voluntary behaviors, especially complex behaviors, are present at birth. As the infant gradually gains voluntary control over various movements and responses, the consequences that follow-some positive and some negative-cause behaviors to gradually emerge, evolve, and refine.
To the extent that caregivers are responsive to infant behaviors, it appears that the contingent nature of their responses is important to the infant. Goldstein, Schwade, and Bornstein (2009) found that 5-month-old infants had learned that an adult had been responding contingently to their vocalizations. After a period of responding contingently to infants' vocalizations, when the adults assumed a still face, infants exhibited extinction bursts in which their smiling decreased and vocalizations increased. Additionally, 9-month-old infants, when presented with vocal production models by their mothers contingent on their own babbling, modified their babbling and included the mother's phonological patterns in their babbling (Goldstein & Schwade, 20008).
Skinner (1957) emphasized the importance of response contingent behavior between speaker that maintains responding and shapes responses during the overall process of learning language. In addition, to the extent that it provides differential reinforcement or corrective feedback it would also shape the outcome of the process by strengthening response forms that evolve toward the conventions of the language. Tamis-LeMonda, Borstein, and Bauwell (2001) found that maternal responsiveness to infants' play and vocalizations at 9 months and 13 months were predictive of the infants' achievement of several language milestones.
Correlates in Early Language Learning: This concept has special relevance to Skinner's analysis of verbal behavior. Following Chomsky's model, Brown (1973), among many others, had asserted that the child was acquiring the rules of grammar by extracting the regularities or patterns in the language he or she was exposed to. The regularities or patterns in the child's language presumably revealed the internalized rules of grammar that governed the child's language production. The distinction between "rule -governed behavior" and "contingency-shaped behavior" became a central issue that crystallized a key difference between Chomsky's theory and Skinner's analysis. The former placed the controlling mechanism in the child's mind (the LAD) and the system of rules being extracted from exposure to the language. The latter placed the controlling mechanism in the history of reinforcement that incrementally shapes behaviors toward conventionally accepted and consistent patterns of verbal behavior. Such consistent patterns may be described as evidence of "rules." However, the behavioral evidence for such "rules" contradicts what most would expect. In characterizing rule-governed behavior, one would expect a more rapid and complete change in behavior that quite promptly complies with the newly learned rule. However, in the case of early language learning, most behaviors evolve gradually (Brown, 1973).
In About Behaviorism,, Skinner wrote, "Certainly for thousands of years people spoke grammatically without knowing that there were rules of grammar. Grammatical behavior was shaped, then as now, by the reinforcing practices of verbal communities in which some behaviors were more effective than others..." (1974, pp. 127-128). In fact, in a seldom quoted section in his landmark book about children's mastery of grammatical morphemes, Brown (1973) very clearly described this shaping process in describing the subjects' "unaccountable regressions and unexplained abrupt advances," stating that
the learning involved must be conceived as gradual change in a set of probabilities rather than as the sudden acquisition of quite general rules. If our conception is correct, it means that the learning of the ... 14 grammatical morphemes is more like habit formation and operant conditioning than anyone has supposed. Skinner's definition of operant strength in terms of response probability is surprisingly apt. (p. 388)
Still, in his overall conclusions, Brown (1973) determined that there was no evidence in his data of "selective pressures" being exerted through adult models or through direct reinforcement by adults. Moerk (1992) obtained Brown's entire data set and performed a different analysis on it. Brown had used frequency counts of parent use of the morphemes in question and found that neither parental models for, nor parental expansions of, a child's utterance correlated with the children's mastery for that morpheme. In contrast, Moerk used a microanalysis of the learning process present in caregiver-child interactions across adjacent time periods. This approach revealed the extent to which the parent's modeling, corrective feedback (reinforcement), and massing (repeated, intense modeling of a structure) were at work in shaping the child's subsequent productions. Brown's simplified operational definition (1973) of models, reinforcement, and feedback had essentially eliminated recognizing their roles in his own data (Moerk, 1992).
Verbal Operants in Developing Language
Skinner designated the basic elements of verbal behavior as primary verbal operants. As the building blocks in his functional analysis, each primary verbal operant is based on the contingency represented between the setting event that "sets the stage" for a characteristic type of verbal behavior due to the reinforcement that has followed that type of verbal behavior. These primary verbal operants become the elements or fragments that, once reinforced, are available to be rearranged through the speaker's secondary verbal behavior into more complex responses. Consistent arrangements of these primary verbal operants evolve in ways that, by convention, reflect the different relationships among them and other variables influencing the speaker. The following is a brief summary of each primary verbal operant. There is certainly more to understanding each one than can be included here. (Again, the reader is referred to Hegde, in this issue, for a more complete review and discussion of Skinner's Verbal Behavior.)
Description: A mand is a verbal operant in which the verbal behavior specifies its reinforcement. It occurs in response to a deprivation state or an aversive state. In general terms, the speaker's verbal behavior tells the listener how to reinforce the speaker by addressing a deprivation state (hunger, thirst, social contact, etc.) or by terminating an aversive state (cold, wet, pain, lack of information, etc.).
Correlates in Early Language Learning: For infants, the earliest cry behaviors, although not conscious or intentional, probably represent the first demands placed on the environment around them (Formby, 1967). Although the infant's cry cannot "specify" the reinforcement, there are not too many needs to choose from-food, dry diaper, sleep, or company-and caregivers typically address the specific need either by recognizing the particular cry through past experience or through a process of elimination. Obviously, for the infant, this responsive and contingent behavior is crucial to implanting the sense that the world responds in some way when he or she behaves and help establish the earliest mand. With motor development, mands expand to include reaching and pointing and as new verbal behaviors are learned, they come to include additional specifics with more socially appropriate and effective forms-"I'd like the donut with white icing and sprinkles, please." The literature in early development of pragmatic abilities is replete with examples of early communicatio n skills that essentially illustrate mands. The most straightforward example is Primitive Speech Acts (PSAs) (Dore, 1974). Of the 10 types of PSAs included in the classification of pragmatic behaviors, four of them are essentially mands-Calling, Protesting, Requesting Answer, and Requesting Action.
Description. The obvious reference for the term echoic is the word "echo." An echoic is verbal behavior that reproduces the acoustic properties of another's verbal behavior. Imitation is the traditional sense of echoics, but traditionally carries with it a sense of intention. We imitate when asked or for some purpose, where there are aspects of others' verbal behavior that are sometimes reproduced without any awareness on the part of the speaker. After some time in a different geographic region, we take on the accent or the expressions of that region without necessarily being conscious of the change in our verbal behavior.
Correlates in Early Language Learning: Infants may exhibit the earliest echoics in the nature of their babbling. Although there is a measure of controversy surrounding this phenomenon, it cannot be denied that infants' babbling gradually sounds increasingly like the native language to which they are exposed. One theory (Mowrer, 1952) that actually predated Verbal Behavior, described the process as one in which speech sounds produced by the infant that more closely resemble the caregivers' sounds that have been long associated with feeding sessions were automatically self-reinforcing. This process causes those sounds that are more natural to the verbal community to gradually become more prominent in the infant's babbling while those that are not, and as a result are not self-reinforced, decrease in frequency. The hypothetical result of Mowrer's autism theory-the gradual evolution of babbling toward the sounds of the parents' language-has been documented even at the level of echoing both the phonemes and intonational contours of another speaker (deBoysson-Bardies, Sagart, & Durand, 1984).
Beyond the first year, echoics would appear to be most useful for children in learning new vocabulary when caregivers and teachers prompt them to repeat a new word. Children appear to imitate vocabulary terms under a variety of circumstances. Leonard, Schwartz, Folger, Newhoff, and Wilcox (1979) found that children's imitations of lexical items were more likely when the term and its referent were novel and when the referent was informative in the present context. Keenan (1974) suggested that at times children imitate their partner's utterance when it appears that they are unable to add significant information to the exchange-probably not unlike the ill-prepared student who simply repeats the instructor's unexpected question.
In addition, some have suggested that in the process of selective imitation, producing partial echoics might provide children with the opportunity to include words and phrases from the language models around them as their repertoire of verbal behavior expands. The correspondence between the models being echoed and the attempted communication might be reinforced differentially. Echoing or including a relevant word, word ending, or phrase might result in an utterance that is reinforced through a better outcome. Folger and Chapman (1978) found that children at the one- and two-word stage were more likely to imitate an adult's utterance if it was an expansion of their own preceding utterance. Interestingly, because the adult's expansion was typically partially based on the child's utterance, it would constitute a partial echoic, and the children were more inclined to imitate these than an adult's exact repetition of their utterance.
Farrar (1992), like several others (Moerk, 1992; Nelson K. E., 1977) found that 2-year-old children were two to three times more likely to imitate the correct grammatical feature in response to corrective recasts which replace incorrect or missing grammatical features, than any other maternal responses. Again, in all these examples, the children's selective imitation through reproducing a partial echoic appeared to be an important learning tool used to shape their production of grammatically progressive forms.
With some fear of belaboring the point, it is important to ask why these instances of echoic behaviors can be considered as learning tools as opposed to simply moments of meaningless, nonprogressive imitation. As indicated throughout this paper, the connection between the caregiver's attention and continued interactions were established as powerful secondary reinforcers from the beginning. To the extent that a child approximates the caregiver's (and through generalization, all adults') recast or expansion of his or her ungrammatical response, the child's echoic behaviors are automatically self-reinforced (see Hegde, in this issue, for more on echoics and language learning).
Description: Verbal behaviors that allow listeners to be in "contact" with the speaker's environment are called tacts in Skinner's model Tacts are essentially verbal behaviors that "name" or "describe" the elements of the environment. Because tacts, to be reinforced, must exhibit conventional correspondence to how others in the verbal community would talk about the same circumstances, they allow sharing with listeners the objects, persons, events, and relationships that originally evoked the speaker's verbal behavior. We share our experiences with others through tacting.
Tacting takes its first form in the proverbial first words, when children begin to consistently respond to their caregiver's presence with "mama" or "dada" or name their favorite toy. Of course, these tacts evoke much attention and celebration, but otherwise caregivers continue to attend and respond to most tacts to encourage children to maintain and expand descriptions of their experiences well into the future, especially those that are out of the caregiver's view.
For the child, the causation for tacting is different than for manding. The two verbal operants might be functionally confused, for example when the toddler's tact is misinterpreted as a mand. Manding is caused by a state of motivation and will be reinforced by receipt of the specified reinforcer-say, when a child truly asks for more ice cream and then eats it. If the words "ice cream" were not due to an actual desire for ice cream, and instead were simply tacts noting its presence to evoke a comment from the caregiver, any ice cream that is provided may be used for finger painting or end up on the floor.
Correlates in Early Language Learning: Children's first words are a source of joy for caregivers and a source of fascination for researchers. Beyond the traditional first word occurring on the first birthday, researchers have been intrigued by the grammatical composition of first words and the first 50 words. At least from the superficial perspective, there has been a predominance of nouns reported in both cases (Nelson K. , 1973). This "noun bias"-naming objects, people, animals, and so forth-has been found in maternal speech to one-year-olds (Goldfield, 1993). Some (e.g., Gentner, 1982) have taken this as evidence of an innate predisposition toward learning nouns, in which children learning any human language should be predisposed to learning nouns. Testing this premise, however, others (e.g., Tardif, Liang, Zhang, Fletcher, & Kaciroti, 2008) have found that there are not only cross-linguistic differences but evidence of strong influence from parental input.
Description. Skinner noted that "stimulus control is by no means precise" (1957, p. 91). When a response is reinforced in the presence of an object or event, there is actually a variety of features that accompany the primary stimulus. When another stimulus appears that shares at least one of those features, the original response may occur as an extended tact-it is extended to the new stimulus based on stimulus generalization. Skinner described a number of extended tacts that illuminate a variety of most interesting verbal behaviors; these will not be addressed here, but are nonetheless fascinating. (See Hegde, in this issue, for a discussion of generalized tacts.)
Correlates in Early Language Learning: Perhaps the most obvious manifestation of extended tacts in our observations of children learning language is what has been widely called "overextensions." The classic examples include the young child who having become familiar with the family pet-the one that has four legs, fur, a tail, and barks-proceeds to call everything with four legs and fur a "doggy." These and the conceptual behavior they illustrate were discussed previously in relation to the influence of stimulus control.
Other examples of extended tacts include the development of figurative language. Metaphors, similes, idioms and proverbs are examples of figurative language that are essentially based on stimulus generalization. In most such expressions, a relationship shared by two sets of stimuli is expressed through an analogy as in the idiomatic expression, "Don't let the cat out of the bag," which likens letting a feline escape to giving out a secret. Due to their abstract nature, these expressions are less likely to be understood by younger children. An individual expression may be learned by rote and used in very limited context until the stimulus generalizations that underlie such figurative expressions become well established. The concept of metaphoric transparency has been used to describe the degree to which the literal meaning and figurative meaning are similar (Gibbs, 1987). Because, in contrast to the previous example, the literal meaning of the expression, "Don't beat around the bush" is less similar to its figurative meaning, this latter idiom would be said to exhibit less metaphoric transparency. Perhaps it would be expeditious to not "beat around the bush" and simply recognize that these two sets of stimuli involved using and understanding this idiom have less in common, which behaviorally makes them less likely to evoke or support the stimulus generalization necessary to understand them.
Autoclitics: Secondary Verbal Behaviors
Description: Perhaps the most subtle and interesting element in Skinner's (1957) functional analysis of verbal behavior is autoclitic behavior. Skinner characterized autoclitic behavior as a secondary level of verbal behavior in which the speaker's behavior comments on the relationships among the primary verbal behaviors by inserting certain words or word endings or through rearranging their order. Skinner classified these secondary behaviors as autoclitic words ("is", "was," "but," "and," etc.), autoclitic tags ("-ed," "-ing," "-s," "-ion," etc.), and autoclitic orders, (such as those used in English for statements, questions, imperatives, etc.).
Autoclitic behavior is most essentially Skinner's way of addressing the traditional concepts of syntax and grammar. Traditionally, syntax and grammar are comprised of the mental rules that dictate the order of words and word endings. In contrast, Skinner conceived of these words, word endings and orders as being reinforced and shaped to conform to the verbal community. Due to the established practice in a verbal community, a certain word order is more effective in sharpening the listener responses than an order that is not part of that practice; the proverbial headline "Man Bites Dog" is eye-catching not because of the three words involved, but because the order in which they have been arranged corresponds to a relationship that occurs infrequently and evokes effective responses as a consequence.
Correlates in Early Language Learning: An important correlate in the traditional literature was again provided by Brown (1973) in his landmark book, A First Language. Brown noted that as the children in his study began producing multi-word utterances, they began to express certain semantic roles and relationships by relying on consistent word orders. This appeared to be in lieu of the grammatical inflections that had not yet emerged. For example, without the possessive inflection /'s/ available, the children appeared to rely on a set word order to "express possession," as the traditional formulation goes. Any number of different word pairs placed in similar order could achieve the same result (e.g., mommy sock, baby shoe, daddy chair). Of course, without the possessive inflection, they still had to rely on adults to interpret the word orders based on the situational context. Brown coined the term "case frames" for these consistent word orders (1973, p. 135). This is an interesting correlate to Skinner's (1957) concept of autoclitic frames (p. 361) and syntactic frames (p. 405) which refer to similar consistent word orders used to comment on relationships among the primary verbal behaviors.
Another important correlate in the literature that illustrates the relevance of Skinner's model appeared not long after Verbal Behavior (1957) was published. Berko (1958) published what became a landmark study illustrating a phenomenon in child language called overregularization. It has been observed that young children initially learn to correctly produce irregular past tense verbs (e.g., ran, sat, ate) and irregular plural nouns (e.g., men, women, children). However, at the approximate time that they begin to produce the regular forms using the grammatical inflections /-ed/ and /-s/, they appear to "unlearn" the irregular forms and produce overregularized forms such as eated, sitted, runned.
This phenomenon has been explained through a variety of mentalistic models, including single mechanism, connectionist, or network mechanisms (Maslen, Theakston, Lieven, & Tomasello, 2004). In contrast, Skinner's model explains this phenomenon in a more objective and scientifically elegant manner relying on the behavioral principles of stimulus generalization and response induction. For example, in the case of past tense the controlling stimulus for both regular and irregular forms is the past time frame for an event. The irregular verb forms learned earlier by children relate to common everyday events-sitsat, eat-ate, run-ran, and so forth. When the child begins to learn the regular past tense inflection, it will naturally occur in response to events that also occur in the past time frame. Because this is the controlling stimulus common to both, stimulus generalization occurs and response induction causes the regular form to generalize to all occurrences of past tense, causing the child to produce sitted, eated, runned, and so on.
The ultimate test of any analysis is whether it offers a better and more parsimonious explanation and whether if finds application. An important dispositio n in the conduct of science is that it is better to go without an explanation than to settle for an inadequate one (Bachrach, 1972; Hegde, 2003). Skinner's analysis of verbal behavior offers a parsimonious explanation based on experimentally manipulable empirical relations as against unobservable mentalistic or cognitive structures (see Schilnger in the current issue). When so many had settled prematurely on an inadequate explanation (e.g., the innate mechanisms, grammatical universals, the LAD and the LAS), we can now see the debt of gratitude owed to so many who were committed to arriving at an empirically based understanding of language learning.
Although more research is needed, applied research based on Skinner's Verbal Behavior has increased to a significant extent, as evidenced in publications in such journals as The Analysis of Verbal Behavior, The Behavior Analyst, Journal of Applied Behavior Analysis, and several other national and international journals. Professionals in a number of fields have used the tools of change described by Skinner-possibly in many cases without recognizing it. Much of the evidenced-based treatment procedures in speech-language pathology are behavioral procedures that are based on Skinner's operant conditioning, currently generally described as applied behavioral analysis (Hegde, 1998). That this evidence has such good fit is vindicating of the insights provided by Skinner's analysis. It is now time for speech-language pathologists to see the inconsistency of holding up the LAD or other innate mechanisms as the basis for language learning while applying behavioral principles in actually changing language behaviors. That most clinicians manipulate the environmental variables, including their own models, contexts, stimuli, and reinforcement in their everyday clinical practice is not only vindicating of Skinner's analysis, but compelling of clinicians to understand and adopt that analysis.
Bachrach, A. J. (1972). Psychological research: In introduction. (3rd ed.). New York, NY: Random House.
Bateson, M. (1971). The interpersonal context of infant vocalizations. Quarterly Progress Report, Research Laboratory of Electronics, 100, 170-176.
Berko Gleason, J. (2005). The development of language (6th ed.). Boston: Allyn & Bacon.
Berko, J. (1958). The child's learning of English morphology. Word, 14, 150-177.
Bloom, L. (1973). One word at a time: The use of single-word utterances before syntax. The Hague: Mouton.
Bloom, L., & Lahey, M. (1978). Language development and language disorders. New York, NY: John Wiley & Sons.
Brown, R. (1973). A first language: The early stages. Cambridge, MA: Harvard University Press.
Brown, R. (1958). Words and things. New York, NY: Free Press.
Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA: M.I.T. Press.
Chomsky, N. (1959). Review of Skinner's Verbal behavior. Language , 26-58.
Chomsky, N. (1957). Syntactic structures. The Hague: Mouton.
Clark, E. V. (1973). What's in a word? On the child's acquisition of semantics in his first language. In T. E. Moore (Ed.), Cognitive development and the acquisition of language. (pp. 65-110). New York, NY: Academic Press.
deBoysson-Bardies, B., Sagart, L., & Durand, C. (1984). Discernible differences in the babbling of infants according to target language. Journal of Child Language , 11, 1-15.
Dore, J. (1974). A pragmatic description of early langugae development. Journal of Psycholinguistic Research, 3, 343-350.
Dore, J. (1986). The development of conversational competence. In R. Scheifelbusch (Ed.), Language competence: Assessment and intervention. (pp. 3-60). San Diego, CA: College-Hill Press.
Dunham, P., Dunham, F., Hurshman, A., & Alexander, T. (1989). Social contingency effects on subsequent perceptual-cognitive tasks in young infants. Child Development, 60, 1486-1496.
Folger, J., & Chapman, R. (1978). A pragmatic analysis of spontaneous imitations. Journal of Child Language, 5, 23-38.
Formby, D. (1967). Maternal recognition of infant's cry. Developmental Medicine and Child Neurology , 9, 293-298.
Gartstein, M. A., Crawford, J., & Robertson, C. D. (2008). Early markers of languge and attention: Mutual contributions and the impact of parent-infant interactions. . Child Psychiatry and Human Development , 9-26.
Gentner, D. (1982). Why nouns are learned before verbs: Linguistic relativity versus natural partitioning. In S. Kuczaj (Ed.), Language development: Language, thought, and culture. (Vol. 2, pp. 301333). Hillsdale, NJ: Erlbaum.
Gibbs, R. W. (1987). Linguistic factors in children's understanding of idioms. Journal of Child Language ,14, 569-586.
Goldfield, B. (1993). Noun bias in maternal speech to one-year-olds. Journal of Child Language, 20, 8599.
Goldstein, M. H., & Schwade, J. A. (20008). Social feedback to infants' babbling facilitates rapid phonological learning. Psychological Science , 19, 515-523.
Goldstein, M. H., Schwade, J. A., & Bornstein, M. H. (2009). The value of vocalizing: Five-month-old infants associate their own noncry vocalizations with responses from caregivers. Child Development , 636-644.
Hart, B., & Risley, T. R. (1999). The social world of children learning to talk. Baltimore, MD: Paul H. Brookes .
Hegde, M. N. (2003). Clinical research in communication disorders: Principles and strategies. (3rd ed.). Autsin, TX, CA: Pro-Ed.
Hegde, M. N. (1998). Treatment procedures in communicative disorders (3rd ed.). Austin, TX: Pro-Ed.
Hulit, L. M. (2006). Born to talk: An introduction to speech and language develpment. Boston: Allyn & Bacon.
Keller, H., & Scholmerich, A. (1987). Infant vocalizations and parental reactions during the first 4 months of life. Developmental Psychology , 62-67.
Kennan, E. (1974). Conversational competence in children. Journal of Child Language , 1 , 163-184.
Kuhl, P. K. (2004). Early languge acquisition: Cracking the speech code. Nature, 5, 831-843.
Kuhl, P. K., Tsao, F.-M., & Liu, H.-M. (2003). Foreign-language experience in infancy: effects of shortterm exposure and social interaction on phonetic learning. Proceedings of the National Academy of Sciences , 9096-9101.
Leonard, L. B., Schwartz, R. G., Folger, M. K., Newhoff, M., & Wilcox, M. J. (1979). Children's imitations of lexical items. Child Development, 50, 19-27.
Locke, J. L. (1993). The child's path to spoken language. Cambridge MA: Harvard University Press.
MacCorquodale, K. (1970). On Chomsky's review of Skinner's Verbal Behavior. Journal of the Experimental Analysis of Behavior , 83-99.
Maslen, R. J., Theakston, A. L., Lieven, E. V., & Tomasello, M. (2004). A dense corpus study of past tense and plural overregularization in English. Journal of Speech, Language, and Hearing Research, 47, 1319-1333.
McLaughlin, S. (2006). Introduction to language development (2nd ed.). Clifton Park NY: Thomson Delmar Learning.
Moerk, E. L. (1992). First language: Taught and learned. Baltimore, MD: Paul H. Brookes.
Mowrer, O. H. (1952). Speech development in the young child: 1. The autism theory of speech development and some clinical applications. Journal of Speech and Hearing Disorders , 263-268.
Nelson, K. (1974). Concept, word, and sentence: Interrelations in acquisition and development. Psychological Review, 81, 11-56.
Nelson, K. E. (1977). Facilitating children's syntax acquisition. Developmental Psychology , 13, 101-107.
Nelson, K. (1973). Structure and strategy in learning to talk. Monographs of the Society for Research in Child Development , 38, 1-136.
Owens, R. E. (2005). Language development: An introduction (6th ed.). Boston: Allyn & Bacon.
Rheingold, K. S., Gewirtz, J., & Ross, H. (1959). Social conditioning of vocalizations in the infant. Journal of Comparative and Physiological Psychology , 52, 68-73.
Routh, D. K. (1969). Conditioning of vocal response differentiation in infants. Developmental Psychology, 1, 219-226.
Searle, J. R. (1969). Speech acts. Cambridge, England: Cambridge University Press.
Skinner, B. F. (1974). About behaviorism. New York: Alfred A. Knopf.
Skinner, B. F. (1988). Skinner's reply to Catania. In A. C. Catania, & S. Harnad (Eds.), The selection of behavior: The operant behaviorism of B. F. Skinner: Comments and consequences (pp. 483-488). Cambridge, UK: Cambridge University Press.
Skinner, B. F. (1938). The behavior of organisms. New York: Appleton.
Skinner, B. F. (1957). Verbal behavior. New York: Appleton.
Tamis-Lamonda, C. S., Bornstein, M. H., & Baumwell, L. (2001). Maternal responsiveness and children's achievement of language milestones. Child Development , 72, 748-767.
Tardif, T., Liang, W., Zhang, Z., Fletcher, P., & Kaciroti, N. (2008). Baby's first 10 words. Developmental Psychology , 44, 929-938.
Todd, G., & Palmer, B. (1968). Social reinforcement of infant babbling. Child Development, 39, 591596.
Wahler, R. G. (1969). Infant social development: Some experimental analsyses of an infant-mother interaction during the first year of life. Journal of Experimental Child Psychology, 7, 101-113.
Weisberg, P. (1963). Social and nonsocial conditioning of infant vocalization. Child Development, 39, 377-388.
Winokur, S. (1976). A primer of verbal behavior. Englewood Cliffs, NJ: Prencice-Hall.
Author Contact Information:
Dr. Scott McLaughlin
University of Central Oklahoma
100 N. University Drive
Edmond, OK 73034
|Printer friendly Cite/link Email Feedback|
|Author:||McLaughlin, Scott F.|
|Publication:||The Journal of Speech-Language Pathology and Applied Behavior Analysis|
|Date:||May 13, 2010|
|Previous Article:||Language and grammar: a behavioral analysis.|
|Next Article:||The bases for language repertoires: functional stimulus-response relations.|