Analysis of the acquisition of verbal operants in a child from 17 months to 2 years of age.
Given these relationships, Skinner provided categories for the analysis of verbal behavior, called verbal operants. Each of the verbal operants was addressed separately. However, this can only be justified by didactic purposes, once multiple variables control a response. Verbal responses are a function of a combination of variables that act simultaneously on a response. In natural situations, verbal operants are not likely emitted separately. According to Serio and Andery (2002), assuming that any of these types of verbal operants can be found in its pure form in any instance of verbal behavior would be a mistake. Like any behavior, verbal behavior is controlled by multiple factors. What can be identified are verbal responses with multiple functions, combining different operants in the same response. This assumption reveals the complexity involved in the analysis of verbal behavior, beginning with the identification of verbal operants. It is the basis for further and more complex analysis.
The complexity involved in the analysis of verbal behavior was recognized by Skinner (1979). According to him, this is one of the reasons why most of the analysis in his book did not have empirical support. The book Verbal Behavior attempts to explain a complex phenomenon, and many assumptions are only interpretations that still require further investigation. Although the book was written in 1957, behavioral analysis still needs this empirical support because of the extent of the interpretations and innovation of its proposal.
Data from other areas sometimes do not contribute to a broader understanding of verbal behavior from Skinner's perspective of analysis. Traditionally, the unit of analysis adopted in studies of language acquisition is based on grammatical aspects, such as nouns, verbs, adjectives, pronouns, prepositions, and suffixes. According to Skinner's proposal, a noun, for example, may have different functions, depending on the relationships that these responses have with their controlling variables. For example, a child can emit the verbal response "ball" in the presence of the object of a ball under the control of social reinforcement received when this response occurs. These relationships characterize this response as a tact.
The same child can emit a verbal response with the same topography as "ball," but under control of the reinforcement of receiving the ball in his hands when this response is emitted. These relationships characterize this response as a mand. When verbal responses are analyzed according to grammatical aspects, this kind of analysis does not describe the relationships that responses maintain with their controlling variables. For Skinner (1957), the meaning of words or sentences is not in the words or sentences themselves but rather in the relationships with their controlling variables. To this extent, identifying the relationships between a verbal response, the conditions under which the response is emitted, and the consequences produced by this response are essential for a complete analysis of verbal behavior.
According to Michael (1984), the extensive literature on language acquisition does not use the concepts, terms, or analyses presented in Verbal Behavior (Skinner, 1957). Even when the terms that describe elementary verbal relationships, such as mands, tacts, and others, are used, they are not associated with an important proposal or used as the unit of analysis, and research could be conducted without the benefit of these distinctions. According to this author, this may have discouraged the interest of behavioral analysts in this literature. Skinner's proposal could provide major contributions to the analysis of language acquisition. According to Fonai and Serio (2007), by using verbal operants as tools for the analysis of verbal behavior, we can overcome some of the dilemmas encountered when such an analysis is made, based on traditional classification systems that emphasize the topography of the response. One example that shows this difficulty is when someone is able to read a foreign language but does not know how to speak the language. Another example is when someone cannot say the name of a particular object while standing before it, but can ask for this object when it is needed. These apparent dilemmas are clarified when dealing with verbal operants and recognizing the uniqueness of each of these relationships and especially the specific history of reinforcement that is necessary to generate each of them (Fonai & Serio, 2007, p. 352)
More studies on the acquisition of verbal behavior that adopt verbal operants as the unit of analysis are needed. Another issue draws attention to research on verbal behavior. According to Partington and Bailey (1993), few studies have utilized the concepts from Skinner's book to analyze verbal behavior in typically developing children. According to these studies, because of the large number of behavioral analysts who work in the applied area with children with developmental delays, the research area has focused on defective verbal behavior. Few studies have investigated the normal development of verbal behavior. The rapid and simultaneous acquisition of multiple verbal operants by typically developing children can make the acquisition and development of verbal behavior in these children different from what occurs in children with atypical development (Partington & Bailey, 1993).
Lee (1981) indicated that psycholinguists do not deny that the techniques described by behavioral analysts produce appropriate verbal behavior, but they argue that these training studies reveal nothing about how language is acquired in the natural environment. You can undoubtedly teach verbal behavior with reinforcement techniques. The question is whether language acquisition in the natural environment occurs in this way. Behavior analysis, given the characteristics of the area, focuses on manipulative research that is more experimental than descriptive. To understand language acquisition in typically developing children in the natural environment, more descriptive longitudinal studies are necessary.
Ribes-Inesta and Quintana (2003) stated that despite the fact that child development is a crucial field for understanding the origins and functions of processes related to complex human behavior, behavior analysis has neglected longitudinal studies of behavioral development in infants and children. According to these authors, research on language development in behavior analysis has focused on two aspects: training acquisition procedures related to the verbal operants proposed by Skinner as mands and tacts and the identification and demonstration of reinforcement control over generalized operant classes, such as imitation, grammatical or syntactic responses, the following of instruction, and verbal/nonverbal correspondence. Few studies have investigated the behavioral interactions between children and mothers or caregivers, and only some studies have dealt with language acquisition and development since its early stages. According to these authors, behavior analysis efforts have rather been directed to experimental or applied studies, in which particular settings and selected behaviors are extracted from the developmental process.
Longitudinal studies that investigate the acquisition of verbal behavior in typically developing children are lacking. Moerk (1976) argued that experimental studies are able to demonstrate what can be potentially taught by parents, but not what is actually taught by them. According to Moerk (1976), to understand the process of language acquisition, investigating in detail the contingencies between the behavior of the mother and child is necessary, and this approach has been used in very few studies.
Ernst Moerk is an author who must be highlighted, considering several works he published on language acquisition (1976, 1978, 1980, 1983, 1985, 1994, 1999). Most of these studies were reanalyses of the data from longitudinal studies performed by Roger Brown (Brown, 1973; Brown & Bellugi, 1964). Brown's works reported the interactions between three pairs of mothers and children.
Overall, the results analyzed by Moerk demonstrate the various responses found in daily mother and child interactions and the decisive effect of the mother on promoting the acquisition and extension of a child's verbal repertoire. The results showed how mothers use these interactions in an instructionally sophisticated way. For example, the results showed that after the mother asked a question and realized that the child could not answer it, the mother expanded part of the question itself and thus provided the necessary information for the child to answer the question. Generally, mothers were very versatile and creative in establishing interaction opportunities with instructional purposes. The results showed that mothers provided on average more than five instances of corrective feedback per hour of interaction with their children, which were adopted by the child. Another interesting phenomenon was that the children became more dominant in the interactions and began to bombard the mothers with questions, even when they knew the answer. It is interesting that the mothers attempted to reply to the questions with new questions for the children to answer.
In a study published in 1999, Moerk discussed the importance of adopting a methodology of analysis that includes the delayed effects of reinforcement contingencies. Every behavior is the product of a history of reinforcement, and the interactions between a mother and child often do not have an immediate effect; the effects may be seen after many repetitions after an interval of time. The negligence of these considerations can lead to misinterpretations of the data, analyzing only the immediate antecedents and consequences of a particular response, and the entire process can be lost.
The purpose of the present study was to longitudinally investigate the acquisition of verbal behavior in a typically developing child age 17 months to 2 years, adopting categories based on the verbal operants proposed by Skinner (1957) as the units of analysis. The research attempted to analyze verbal responses emitted by a child and his caregivers in natural situations, trying to identify relationships between the emission of the categories of the child and the caregiver and investigate possible patterns of interaction in the acquisition of verbal behavior.
The participants were one 17-month-old boy and his caregivers, including his parents; his 2-year, 7-month-old brother; his nanny; and other family members. The parents signed an informed consent form according to the criteria of 196/96 resolution of Brazilian National Health Council. The research was approved by the ethics committee on Human Research of the institute of Psychology of the University of Sao Paulo.
A Sony DCR-HC96 camcorder and USB memory stick were used to record the sessions. The camera did not stay fixed in one position throughout the sessions; it moved to follow the movements of the child, always trying to show the child from the front and allowing verbal responses to be recorded more accurately.
The sessions were conducted in a natural environment, usually at the child's house.
Thirty-four sessions were recorded. They were recorded once per week, with no fixed schedule. The recordings comprised 7 months of data collection, beginning when the child was 17 months old and ending 1 week after he reached 2 years of age. Each session lasted an average of 15 min. The child was accompanied by different relatives at the sessions (i.e., his brother, mother, father, and others were present).
Transcriptions of the sessions were made, registering the child's verbal responses, their antecedents and consequences, and describing the interaction between the child and caregivers. The register form was a table with three columns that corresponded to a three-term contingency (antecedents, child's verbal response, consequences) describing the contingencies of reinforcement of the child's verbal behavior. This was an attempt to identify functional relations of the child's verbal behavior. An interpretative analysis of the transcriptions was made categorizing the child's verbal behavior according to Skinner's verbal operants. The same analysis was made for the caregiver's verbal responses.
The following categories applied to both the analysis of the child's responses and the caregivers' responses. The categories that applied only to the child's or caregivers' responses are described in the following section. The categories were inspired by the categories proposed by Skinner's book Verbal Behavior (Skinner, 1957). However, other categories and subcategories were created to add specific features described by language acquisition research. According to Michael, Palmer, and Sundberg (2011), pure forms of the respective verbal operants are rare outside the laboratory or instructional contexts. Because verbal behavior is mostly under multiple control and responses are generally under the control of more than one stimulus simultaneously, the categories are not exclusive. An attempt was made, in some cases, to exclude other possibilities of interpretation.
Categories of the Analysis of the Child's and Caregivers' Verbal Responses 1. Vocalizations (only for the analysis of the child's verbal responses) Vocal responses with sound patterns with no apparent meaning, for which no antecedents or consequences that would characterize them as a specific verbal operant. 2. Listener responses Specific responses modeled by a verbal community subsequent to the verbal response of a speaker. For example, after the verbal stimulus "give me the ball" is emitted by the speaker, the listener emits the response of throwing the ball. The responses categorized as those emitted by the listener were only motor responses. Oral responses emitted after verbal stimuli were categorized according to their antecedents and consequences and interpreted as belonging to other categories. 3. Other responses Nonvocal responses directed to the child or caregiver, such as pointing, gesturing, approaching, and smiling at the speaker during the emission of verbal responses. 4. Imitative responses Nonvocal imitative responses. The child emits a response with a topography similar to a previous response emitted by the caregiver. For example, after his brother climbed on the couch, the child climbed on the couch in the same way. When his brother stepped in a puddle of water, the child emitted a response with the same topography. 5. Mand Responses emitted under a condition of deprivation or aversive stimulation and reinforced by a specific consequence, often named by the response itself. For example: "Listen," "Look," "Run," "Stop," "More juice," and "Give it to me." 5.1. Mands of the caregiver, in which the specific reinforcer was the emission of verbal responses by the child Caregivers' mands, in which the specific reinforcer was the emission of verbal responses by the child, were specifically recorded, such as "What is it called?", "What is the name of this?", "What is that?", and "Tell me what happened." 6. Echoic Vocal responses with the same topography as the vocal verbal stimulus that preceded it. 6.1. Imperfect echoic (exclusively for the analysis of the child's verbal responses) Verbal responses that are reinforced because of the similarity to the antecedent verbal stimulus but do not maintain a point-by-point correspondence with the previous stimulus. For example, after the preceding verbal stimulus "toy," the child emits the verbal response "oy." 6.2. Expanded echoic (exclusively for the analysis of the caregivers' verbal responses) Echoic responses of the caretaker that expand or complement the verbal response of the child. For example, the child emits the verbal response "oy," and the caregiver emits an echoic response, "toy." 6.3. Echoic that confirms or investigates the response emitted by the child (exclusively for the analysis of the caregivers' verbal responses) Echoic responses that are reinforced by confirming or investigating which response was emitted earlier by the child. For example, the child emits the verbal response "ball," and the caregiver says "ball?" Notably, in this case, the caregiver's echoic response could also function as a mand, in which the specific reinforcer is the repetition of the verbal response emitted earlier by the child. In this Case, these responses were categorized only as belonging to this category. 6.4. Echoic with possible reinforcing function (exclusively for the analysis of the caregivers' verbal responses) Echoic responses of the caregiver that are likely to function as reinforcers for the child's responses. For example, the child says "ball," and the caregiver emits the response "ball!!!", possibly reinforcing the response that was emitted by the child. 7. Intraverbal Verbal responses controlled by previous verbal stimuli that do not maintain a point-to-point correspondence. For example, saying the alphabet, counting, singing, and so forth. 8. Tacts Verbal responses under the control of previous nonverbal stimuli and social reinforcement. 8.1. Tacts occasioned by aspects of the physical and social environment For example, "The ball is blue." 8.2. Tacts occasioned by the child's behavior For example, "You caught the ball." 8.3. Tacts occasioned by other people's behavior For example, "Mommy caught the ball."
Categories That Relate Verbal Operants Emitted by the Child and Caregivers
These relationships were made to identify conditions established by the caregivers that may facilitate verbal operant acquisition by the child. This was an attempt to identify operants emitted by the caregivers that preceded the emission of specific operants by the child. At first, the following categories may appear to disagree with Skinner's verbal operant definitions. However, what allows this analysis is the assumption that a response is not controlled by a single stimulus but is under the control of multiple factors. As explained in the following paragraph, a response that was interpreted as a tact, accounting for its strong relationship with a previous nonverbal stimulus and generalized reinforcers, may also have been preceded by other stimuli. Because the purpose of this analysis is to identify the conditions under which these operants are learned, the identification of other stimuli that precede the response may be important to understand these conditions. The categories applied only to operants emitted by the child.
1. Echoic 1.1. Echoic preceded by mand A mand of the caregiver, such as "Say mommy" preceding an echoic of the child, such as "mommy." In some situations the mand of the caregiver specified the echoic response of the participant, but in others situations it did not. For example, the caregiver emits the mand "Give me the ball," and the child echoes "ball." 1.2. Echoic preceded by tact The caregiver emits a tact (e.g., "This is Mike"), and the child echoes "Mike." 1.3. Echoic preceded by intraverbal The caregiver emits an intraverbal, such as "Happy birthday to you," and the child emits the echoic response "birthday." 1.4. Echoic preceded by echoic The caregiver emits an echoic response, and so does the child. For example, the child says "mom," the caregiver emits the echoic "mommy," and the child emits the echoic "mommy." 2. Tact 2.1. Tact preceded by tact The caregiver emits the tact "the airplane," and after a while the child looks at the aircraft and emits the tact "the airplane." The child's response is a tact in accounting for its relationship of control with a nonverbal stimulus that was emitted at the sight of a passing airplane in the sky and not before the emission of the verbal response of the caregiver. However, the emission of the tact by the child occurred moments after the caregiver emitted a tact. In another example, the caregiver emits the response "I found the cow," picking up a toy cow, and the child emits the response "I found the pig" by picking up a toy pig. 2.2. Tact preceded by mand The caregiver emits a mand, such as "What is it?", and the child emits the tact "the ball," which is followed by generalized reinforcers, such as "That's right." In this case, the child responds under the control of a verbal stimulus emitted by the caregiver and a nonverbal stimulus, such as the object present at that time. These responses were categorized as tacts because they were under the control of a nonverbal stimulus and generalized reinforcers, but it was preceded by a mand of the caregiver. 2.3. Tact without any previous verbal stimulus The child emits a tact under the control of an object or event without any previous verbal stimulus. For example, seeing the ball and then saying "the ball." 3. Mand 3.1. Mand preceded by mand The child emits a mand that was preceded by a mand emitted by the caregiver. For example, the child points to a ball, the caregiver holds it and says, "Say what do you want," and gives the ball to the child just after the emission of a verbal response, such as "ball" or "give me." In this case, the child's response was interpreted as a mand because it was under the control of a particular reinforcer, such as the ball. However, the mand "ball" was emitted by the child after a mand emitted by the caregiver. 3.2. Mand preceded by tact The caregiver emits a tact, such as "the ball," and the child emits a mand, "give me." that is reinforced by delivering the ball to the child. The child's verbal response was interpreted as a mand because it was under control of the reinforcer stimulus (i.e., ball) but was also preceded by the emission of a tact by the caregiver. 3.3. Mand without any previous verbal stimulus The child emits a 'nand under the control of a specific reinforcer without any previous verbal stimulus. For example, the child sees the ball and says "give me," and the caregiver gives it to the child. 3.4. Mand preceded by echoic The child emits a mand, such as "the pig," the caregiver emits an echoic "the pig?", and the child emits another mand, "give me," so he caregiver hands the rubber pig to the child. In this case, the second mand emitted by the child was preceded by an echoic emitted by the caregiver. 4. Intraverbal 4.1. Intraverbal preceded by intraverbal The emission of an intraverbal by the child is preceded by the emission of an intraverbal by the caregiver, such as the caregiver counting "One, two ..." and the child says, "Go!" 4.2. Intraverbal preceded by mand The caregiver emits a mand, such as "Wait there," and the child emits the intraverbal "Ok." 4.3. Intraverbal preceded by tact The caregiver emits a tact, such as "this is a telephone," and the child emits an intraverbal "hello," or after the emission of the tact "lion" emitted by the caregiver, the child responds by saying "Grrr."
An overview of the results showed that the cumulative frequency of the mand, tact, echoic, and intraverbal categories emitted by the child increased gradually and simultaneously until approximately 20 months of age (see Figure 1). From that point, the curves sloped, indicating a rapid and significant increase in the emission of these categories until the last session. This increase was more prominent in the tact and mand categories.
The cumulative frequency of the mand category began to increase from the beginning of the 20th month and around 21 months went through an even more intense increase. Tacts went through this sudden increase in their cumulative frequency in the middle of the 20th month, and echoics went through this sudden increase in their cumulative frequency at the end of this month. The increase in the cumulative frequency of intraverbals appeared to occur a few sessions after the increase in the tact, mand, and echoic categories at approximately the 21st month.
The echoic category also increased rapidly between the 20th and 21st month. Unlike the mand, tact, and intraverbal categories, the cumulative frequency of echoics reverted to slower growth after 21 months.
The vocalizations category presented a different pattern. The frequency increased from 17 to 20 months and from that point onward decelerated. Afterward, the emission of vocalizations gradually decreases, whereas the frequency of the emission of other operants continues to increase until 28 months, when it stabilizes. Figure 1 shows that the frequency of the emission of vocalizations was the highest of all of the categories until 20 months, equaling the emission of tacts and mands at that point, and being exceeded by these categories at the beginning of the 21st month.
Figure 2 shows the cumulative frequency of categories emitted by caregivers during the 34 sessions. The cumulative frequency of mands, tacts, and echoics significantly increased over the sessions. The cumulative frequency of mands increased more intensely after 20 months, similar to the cumulative frequency of the mands of the child (see Figure 1). The cumulative frequency of the tacts of the caregivers increased more abruptly in the middle of the 20th month, simultaneously with the increase in the cumulative frequency of the child's category of tacts (see Figure 1). The cumulative frequency of the category of echoics of the caregivers increased more intensely between the 20th and 21st month, again coinciding with the increased frequency of the same category of the child (see Figure 1). These data appear to indicate a relationship between the emission of verbal operants by the caregiver and child.
Another interesting finding was that the most frequently emitted category by the caregiver was mands, prevailing over the other categories since the first session. Unlike the results from the categories emitted by the child, in which the most frequently emitted category was tacts (see Figure 1). The tact and echoic categories of the caregivers were the ones emitted with the second and third largest cumulative frequencies.
Figure 3 shows the cumulative frequency of mands and mands that were reinforced by the child's verbal responses emitted by the caregiver. The results showed that the majority of the mands emitted by the caregiver were mands that were reinforced by the child's verbal responses (e.g., "Tell mama what you're doing," "Tell daddy what you saw," "What is this?", and "Who are you playing with?"). More than 60% of the mands were reinforced by the child's verbal responses, and more than 70% of the mands were of that kind in 12 sessions. The results suggest that parents' mands may play an important role in installing the child's verbal repertoire.
The majority of the tacts emitted by both the caregivers and the child were tacts occasioned by aspects of the environment (see Figure 4), followed by tacts occasioned by the child's behavior and tacts occasioned by the behavior of others.
Figure 5 shows the cumulative frequency of echoics emitted by the caregiver compared with the frequency of echoics and imperfect echoics emitted by the child. The results indicate that the caregivers emitted a higher frequency of echoics than the child.
Another interesting finding is presented in Figure 5. Initially, almost all of the echoics emitted by the child were imperfect echoics. As the sessions continued, the cumulative frequency of imperfect echoics was progressively differentiated from the cumulative frequency of echoics, indicating that the child acquired a generalized echoic repertoire and became able to properly echo new words that he heard.
The results showed in Figure 1 are similar to others in the literature. Researchers who used different approaches agreed that one of the most representative phenomena in language acquisition is the fast learning of words related to objects, occurring between 18 and 24 months of age (Oliveira& Gil, 2007). Hart and Risley (1999) showed that children's vocabulary increased gradually until 19 months of age, undergoing a rapid acceleration beginning in the 20th month. According to these authors, early in the language acquisition process, between 7 and 10 months of age, parents spoke to their children without waiting for verbal responses from the child. A general interaction occurred, with parents directing and talking to their children unilaterally. When the children began to utter their first words, at an average of 11 months, a gradual change in verbal interactions between the parents and children occurred. The number of verbalizations and imitation of verbal responses increased. This period was characterized by changes in the pattern of interaction, changing from unilateral to reciprocal between the parents and children. Interactions, however, were still characterized by a predominance of parental speech. From 11 to 19 months of age, the children began to interact more verbally and, in return, parents asked more and more questions. At 20 months of age, the children were able to say recognizable words in more than half of their utterances. The vocabulary growth curve created by Hart and Risley (1999) to illustrate the average growth of vocabulary by age in months is very similar to the curve of cumulative frequency of the tact category.
A qualitative analysis of the sessions occurred between 20 and 22 months suggested that a general echoic repertoire may have emerged at the beginning of the 21st month. The child emitted echoic responses of more complex words that were not recorded in previous sessions. In one session, the father, older brother, and child were looking at pictures of animals. While they were shown the figures, the father and brother emitted various tacts, such as "dolphin," "giraffe," "butterfly," and "eagle." After the emission of the father's and brother's tacts, the child emitted imperfect echoics, such as "phin," "affe," "fly," and "gle." During this session, some images of the animals that had complex names arose, and the child emitted imperfect echoics of words that were not observed previously in his repertoire. This significant increase in tacts and echoics appeared to occur simultaneously with a decrease in the acceleration of mands at between 20 and 21 months.
During this session, an interaction pattern between the caregivers and child was observed. A caregiver emitted a tact, the child emitted an imperfect echoic, and the caregiver emitted an expanded echoic. For example, after being shown the picture of the dolphin, the caregiver said "dolphin," the child emitted an imperfect echoic "phin," and the caregiver emitted an expanded echoic "dolphin," emphasizing the sounds that were omitted by the child.
Similar data were also presented by Moerk (1976). The author highlighted two patterns of interaction in verbal episodes initiated by the mother. One of them is based on the categories emitted by the mother: "models from picture book," "describes an object on hand," "describes an observed event," "models a phrase not from a picture book," "describes own action," "labels an object," "provides information," and "describes child's act." These categories were followed more frequently by the child categories "imitating" and "expanding an utterance," with "imitates" the most frequent category. These categories emitted by the mother can be interpreted as tacts (e.g., those that describe objects, events, and actions, model from picture books, and label objects). The category "imitates" was emitted more frequently by the child and could be interpreted as echoic behavior.
This pattern changed across the sessions that followed. The child began to emit tacts, other than echoic, such as saying "phin" after seeing the picture of a dolphin, and the caregiver emitted an expanded echoic "dolphin." Verbal responses that were previously echoics because they were emitted under the control of a previous oral stimulus that had a formal correspondence with the response and generalized reinforcers began to be emitted as tacts. They were no longer under the control of a previous oral stimulus emitted by the caregiver but rather under the control of the picture. Perhaps this change in the interaction pattern could explain why the cumulative frequency of echoics decelerated after 22 months, whereas tacts continued to grow at a high rate. Echoics may have an important function in tact acquisition.
A transfer of stimulus control appeared to occur. The child's verbal response was emitted in the presence of two stimuli, the oral stimulus emitted by the caregivers and the picture. At first, the child's response was emitted under the control of the oral stimulus emitted by others, and then the picture acquired control over this response. These data indicate that echoics can play a facilitating role in the acquisition of other operants. Skinner (1957) suggested this in his book. In the example cited by the author, a child learned to say "alligator" first as an echoic, and then "alligator" was emitted in the presence of an alligator. The alligator as a stimulus acquired control over the response (Skinner, 1957, p. 62).
According to Michael et al. (2011), formal control is important because a response can be emitted independent of what is conventionally called its meaning. Thus, we can echo words we do not know. An elementary echoic repertoire permits a response to occur in a single trial, passing the long process of shaping, but once emitted, a response originally under formal control can be followed by important consequences and come under control of other relevant stimuli. The authors emphasize that the importance of this point cannot be overstated. However, these issues require further empirical investigation.
These data can also be compared with Moerk (1976). His research indicated that the most frequent interaction pattern in verbal episodes initiated by the child consisted of the child categories "to describe a picture," "to describe an object," "to describe an event," "to describe his own action," and "name an object," followed by the mother categories "to give corrective feedback," "to expand verbalization," and "ask questions." The author also reported that the child became more dominant in the interactions throughout the sessions. Compared with the data presented in Figure 1, we notice that the category most often emitted by the child was tacts, corresponding to the categories "to describe a picture," "to describe an object," "to describe an event," "to describe his own action," and "to name an object."
The interaction pattern changed again a third time. Responses that were emitted as tacts began to be emitted as intraverbals. Responses emitted by the child under the control of nonverbal antecedent stimuli, such as the verbal response "dolphin" after seeing the picture of a dolphin, began to be emitted after a verbal stimulus. For example, the mother asked the child, "What were you seeing with Daddy?" and the child replied, "Dolphin, zebra, spider, fish, toad...," no longer in the presence of the pictures. Again, a transfer in stimulus control occurred. This may be a possible explanation for the abrupt increase that occurred slightly late in the cumulative frequency of intraverbals after 21 months.
This transfer of stimulus control that possibly facilitated the acquisition of new verbal operants was possibly attributable to the multiple control of verbal behavior. A response can be affected by more than one variable, and stimuli that were present at the time of emission of the response but did not have discriminative control over it slowly began to exert control over the response. Thus, a verbal response that initially had the function of a specific verbal operant acquired new functions and became a new verbal operant (e.g., from tact to intraverbal).
The pattern of vocalizations category showed in Figure 1 can also be compared with the results of Hart and Risley (1999). According to these authors, after the emission of the first word at approximately 11 months, the frequency of emission of recognized words increases stably without the occurrence of a decrease in the emission of vocalizations that are not words, such as babbling, the emission of meaningless sounds, and screams. By the age of 18 or 19 months, the frequency of the emission of vocalizations equals the frequency of the emission of words.
Results of the cumulative frequency of categories emitted by caregivers showed in Figure 2 are consistent with Moerk (1976), who described that the mother asks the child too many questions, including questions for which she already knows the answer and she knows that the child is able to respond. This indicates that the questions had the function of a mand and had as a reinforcer the emission of a verbal response by the child.
Results showed in Figure 3 can also be compared with Hart and Risley (1999). The authors reported that many parents asked the children many questions, such as "How?", "Where?", "Who?", "When?", and "Which one?", encouraging them to name, describe, and remember what happened to them and around them. According to the authors, the number of questions asked by the parents increased until 24 or 25 months of age. Thereafter, the number of questions began to decrease as the child gradually began to speak up and start a conversation, no longer requiring questions. These results may indicate that the mands of the caregiver play an important role in language acquisition, especially in early stages.
Figure 4 indicates that the majority of the tacts emitted by both the caregivers and the child were tacts occasioned by aspects of the environment. Hart and Risley (1999) reported that the first words spoken by children were the names of toys, mommy, daddy, baby bottles, and so forth. It is interesting that the cumulative frequency of tacts occasioned by aspects of the environment emitted by the child surpassed the frequency of the same category of the caregivers after 21 months.
Results presented in Figure 5 indicate that the caregivers emitted a higher frequency of echoics than did the child. Hart and Risley (1999) found similar data. According to the authors, parents imitated more the words emitted by the children than otherwise (Hart & Risley, 1999). Furthermore, the parents asked the children to imitate them.
Moerk (1990) found evidence that two types of response delivered by the mother functioned as reinforcers: agreement--that was a category defined as the mother saying "yes," "right," or an equivalent; and expansions--that was a category defined as the mother repeating the child's utterances but adding items that were omitted by the child or that were incorrect. The expansions functioned as a reinforcer and also as a corrective feedback. These results may suggest that echoics are used by parents as teaching strategies to develop children's verbal repertoire. Echoics seems to have a double function: reinforcer and discriminative stimuli for other responses.
Generally, the results indicate that caregivers arrange contingencies for the acquisition of verbal operants in the repertoire of the child. These results can contribute to the discussion about how verbal behavior is learned and to the debate of nativism and environmental factors (Schoneberger, 2010). Mands and echoics appear to be important operants used by the caregivers as teaching strategies. Echoics apparently play an important role as a repertoire that facilitates the acquisition of other operants by the child.
Longitudinal studies can be valuable for describing the acquisition of verbal repertoires in a natural environment. The adoption of verbal operants proposed by Skinner (1957) as the unit of analysis has provided new data on the knowledge that evolved in his book, such as the frequency of the emission of different operants during the acquisition of verbal behavior, changes in the repertoire of the child during this acquisition, and verbal interaction patterns that may be established between the child and caregivers.
The book Verbal Behavior (Skinner, 1957) proposed a theory for understanding language. Research that investigates and contributes to this comprehension is necessary. Developing and improving the methodology to study verbal behavior are also essential and should involve discussions of the categories of analysis and how to investigate this phenomenon, considering multiple control and the effects of reinforcement history.
This research was supported by grants from CNPq (141541/2005-9). This research adheres to relevant ethical guidelines (e.g., Ethical Principles of Psychologists and Code of Conduct, 2002), including adherence to the legal requirements of the study country. The participants signed an informed consent form according to the criteria of 196/96 resolution of Brazilian National Health Council. The research was approved by the ethics committee on Human Research of the Institute of Psychology of University of Sao Paulo as stated in page 5.
Correspondence concerning this article should be addressed to Adriana Cunha Cruvinel, Rua Campanha 127/1701, Belo Horizonte, MG, 30310-770, Brazil. E-mail: email@example.com
BROWN, R. (1973). A first language: The early stages. Cambridge, MA: Harvard University Press.
BROWN, R., & BELLUGI, U. (1964). Three processes in the child's acquisition of syntax. Harvard Educational Review, 35, 133-151.
FONAI, A. C. V., & SERI, T. M. A. P. (2007). 0 conceito de audiencia e os multiplos controles do comportamento verbal [The concept of audience and verbal behavior's multiple control]. Revista Brasileira de Terapia Comportamental e Cognitiva [Brazilian Journal of Behavioral and Cognitive Therapy], 9(2), 349-360.
HART, B., & RISLEY, T. R. (1999). The social world of children learning to talk. Baltimore, MD: Brookes.
LEE, V. L. (1981). Terminological and conceptual revision in the experimental analysis of language development: Why? Behaviorism, 9,25-53.
MICHAEL, J. (1984). Verbal behavior. Journal of the Experimental Analysis of Behavior, 42(3), 363-376.
MICHAEL, J., PALMER, D. C., & SUNDBERG, M. L. (2011). The multiple control of verbal behavior. The Analysis of Verbal Behavior, 27, 3-22.
MOERK, E. L. (1976) Processes of language teaching and training in the interactions of mother--child dyads. Child Development, 47, 1064-1078.
MOERK, E. L. (1978). Determiners and consequences of verbal behaviors of young children and their mothers. Developmental Psychology, 14, 537-545.
MOERK, E. L. (1980). Relantionships between parental input frequencies and children's language acquisition: A reanalysis of Browns's data. Journal of Chisd Language, 7, 105-118.
MOERK, E. L. (1983). A behavior analysis of controversial topics in first language acquisition: Reinforcements, corrections, modeling, input frequencies, and the three term contingency pattern. Journal of Psycholingistic Research, 12, 129-155.
MOERK, E. L. (1985). Analytic, synthetic, abstracting, and word class defining aspects of verbal mother child interactions. Journal of Psycholinguistic Research, 14, 263-287.
MOERK, E. L. (1990). Three-term contingency patterns in mother-child verbal-interactions during first-language acquisition. Journal of the Experimental Analysis of Behavior, 54, 293-305.
MOERK, E. L. (1994). Corrections in first language acquisition: Theoretical controversies and factual evidence. International Journal of Psycholinguistics, 2, 43-72.
MOERK, E. L. (1999). Sequential analyses, multiple controlling stimuli, and temporal patterning in first-language transmission. Analysis of Verbal Behavior, 16,63-80.
OLIVEIRA, T. P., & GIL, M. S. C. (2007). Elementos fundamentais para a aquisicao de operantes verbais por bebes: Analise comportamental da "atencao compartilhada [Fundamental elements for infant's verbal operant acquisition: behavioral analysis of joint attention]. Revista Brasileira de Terapia Comportamental e Cognitiva [Brazilian Journal of Behavioral and Cognitive Therapy], 9(2), 217-225.
PARTINGTON, J. W., & BAILEY, J. S. (1993). Teaching intraverbal behavior to preschool children. Analysis of Verbal Behavior, 11, 9-18.
RIBES-INESTA, E., & QUINTANA, C. (2003). Mother-child linguistic interactions and behavioral development: A multidimensional observational. Behavior Analyst Today, 3,442-454.
SCHONEBERGER, T. (2010). Three myths from the language acquisition literature. The Analysis of Verbal Behavior, 26,107-131.
SERIO, T. M. A. P., & ANDERY, M. A. (2002). Comportamento verbal Verbal behavior. In T. M. A. P. Serio, M. A. Andery, P. S. Gioia, & N. Micheletto (Eds.), Controle de estimulos e comportamento operante-uma (nova) introducao[Stimulus control and operant behavior--A (new) introduction] (pp. 113-137). Sao Paulo, Brazil: EDUC.
SKINNER, B. F. (1957) Verbal behavior. Englewood Cliffs, NJ: Prentice Hall.
SKINNER, B. F. (1979). The shaping of a behaviorist. New York, NY: Knopf.
Adriana Cunha Cruvinel and Maria Martha Costa Hubner
University of Sao Paulo, Brazil
|Printer friendly Cite/link Email Feedback|
|Author:||Cruvinel, Adriana Cunha; Hubner, Maria Martha Costa|
|Publication:||The Psychological Record|
|Date:||Sep 22, 2013|
|Previous Article:||Reinforced behavioral variability in humans.|
|Next Article:||Constructing relational sentences: establishing arbitrarily applicable comparative relations with the relational completion procedure.|