3E5.» WI1WIHIIWIWIIWHHIHIIHINIWHIWIIHI ‘Em'a'if‘fidzt 5‘33333‘3flf'w tit-e49 3-1;:sznsg-wh €31.th I ,3 21. ~ 1' ,7 n g"- . bias-thwam This is to certify that the thesis entitled INDIRECT REQUEST COMPREHENSION ABILITIES OF HEARING IMPAIRED CHILDREN presented by Marilyn Leigh Park has been accepted towards fulfillment of the requirements for FLA- degree in _Audjnjpgy_and Speech Sciences Major professor Date May 1986 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES m RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. INDIRECT REQUEST COMPREHENSION ABILITIES OF HEARING IMPAIRED CHILDREN By Marilyn Leigh Park A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Audiology and Speech Sciences 1986 ABSTRACT INDIRECT REQUEST COMPREHENSION ABILITIES OF HEARING IMPAIRED CHILDREN By Marilyn Leigh Park This study investigated young hearing impaired children's ability to comprehend various types of indirect requests. Subjects included 17 moderately-severe to profound hearing impaired children between the ages of 5 and 12 who were placed into one of four groups, depending on grade placement. Subjects each received a red and a blue crayon along with a response booklet that contained 40 sheets of paper. Each sheet had a blank circle drawn on it. The general design was one in which children were to listen to an indirect request and color the circle according to what they thought had been requested. Forty sentences representing a variety of syntactic forms were presented without supporting linguistic or nonlinguistic contextual information. Results indicated a signifi- cant developmental effect with older children exhibiting greater comprehension of indirect requests than the younger children. ACKNOWLEDGMENTS Without the interest, patience, and support of a number of individuals, this manuscript would not have reached completion. It is to those individuals that I wish to express my gratitude. To Dr. Michael N. Casby, thesis director, for his unfailing support and enthusiasm for this research; furthermore, for the friendship and empathy he offered to this "reentry student" during my undergraduate and graduate studies. To Dr. Leo V. Deal, academic advisor and thesis committee members, for showing me in our first interview that the impossible dream, wasn't. To Dr. Linda Lou Smith, thesis committee member, for her interest in and support of this research when I was all but a stranger to her. To the administrators, staff, teachers, parents, and students of the Grand Rapids Oral Deaf Program, for their assistance and interest in this research. To Susan and Karen, my "reentry friends," for making returning to school seem normal, healthy and fun. We've come a long way baby! To my parents, Russ and Jean, for their invaluable love and assistance. This time, I'll accept being a statistic! ii To my uncle, John Godfrey, for his emotional and financial assistance during my undergraduate studies. To my sister, Lynne, and brother-in-law, Dennis, for sharing the laughs and tears of the past six years. To my husband, Joe, for never doubting my ability or my determination to finish school. It was much easier with your support. To my children, Kiersten, Ryan, and Marisa, for making this all worthwhile. This one's for you, guys! TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES INTRODUCTION GRAMMATICAL COMPETENCE . Phonology Morphology Syntax Semantics Pragmatics PRAGMATIC COMPETENCE: INDIRECT REQUESTS . Indirect Requests . . Indirect Requests and Language Impairment . METHOD Stimuli Subjects . Procedure . Temporal Reliability RESULTS . Comparison of Children's Comprehension of Positively Conveyed and Negatively Conveyed Indirect Requests Expressed in Different Sentence Modalities Correlational Analysis . . . DISCUSSION . REFERENCES . iv Page 31 34 34 39 46 Table LIST OF TABLES Pairs of Sentences Used as Experimental Stimuli with Polarity of Conveyed Meaning Noted . . . . Individual Subject Table and Data for the Following Variables: Grade, Chronological Age (CA), Hearing Age (HA), Unaided Hearing Level (UAHL), Aided Hearing Level (AHL), Language Age Receptive (LAR), Language Age Expressive (LAE), Positive Score in Experimental Task (POS), Negative Score in Experi- mental Task (TOT) . . . . . Means and Standard Deviations of Group and Combined Performance for the Positively Conveyed, Negatively Conveyed Items and Total Score . . . . Percentage of Children's Correct Comprehension of Indirect Requests Relative to Different Sentence Modalities . . . . . . . . . . Spearman Rank Order Correlations for Subject Variables . . . . . . . . Page 21 25 32 35 36 LIST OF FIGURES Figure Page 1. Group Means for Children's Responses to Positively and Negatively Conveyed Indirect Requests . . . . 33 vi INTRODUCTION Language involves the comprehension and production of phonology, morphology, syntax, semantics and pragmatics. In general, develop- ment of these aspects is predictable across children, and those with normal abilities have little difficulty learning them early through exposure to their linguistic environment. However, at a time when they are most ready, children with hearing losses greater than 56+ dB HL (ANSI, 1969) typically miss effective oral language exposure. Consequently, differences between normal hearing children and this group of hearing impaired children are seen across the five components of oral language. Many of these differences are noted below. GRAMMATICAL COMPETENCE Phonology Phonology concerns the specific phonemes and combinations of sequences of phonemes that are acceptable within a given language. Nith decreased auditory sensitivity, even children with hearing losses classified as moderate to mild can suffer deleterious effects in phonology. Ling (1972) reported that some children who have fluctuating conductive hearing losses experience spelling and read- ing fluency difficulties due to phonologic delays. Oller, Jensen, and LaFayette (1978) reported that hearing impaired children may employ phonological processes similar to younger hearing children. For example, the phonetic preferences of the hearing impaired child in their study included reductions of consonant clusters, deletions of final consonants, deletion or devoicing of final voiced conso- nants, avoidance of final velar plosives, fronting of consonants, and fricatives replaced by stops. All of these preferences have been viewed in normal children but at much earlier ages. Morphology As hearing children develop language, they learn to combine sounds into meaningful parts of words. Hearing impaired children have been found to be inferior to hearing children in their ability to generate morphological forms. Garber (1967) noted that while the ability to use noun inflections was age related, verb and possessive inflections were not, thereby confirming the difficulties hearing impaired children have in learning verb forms of spoken English. When Laubscher (1969) analyzed spoken language samples of 5 through 10 year old hearing impaired children for the development of correct usage of determiners, plural inflections, occurrence of verb tenses and verb inflections, he found that the present verb tense was acquired earliest, followed by present progressive, past, and past progressive. Further data analysis revealed that the subjects infrequently used plural nouns or third person singular verb inflections and showed slow increases with age. He noted strong similarities in the developing morphological trends of normal hearing and hearing impaired youngsters but that the hearing impaired developed these features at a later age and took longer to do so. Syntax After the introduction of Chomsky's (1957, 1965) theory of generative grammar, research on syntax, or sentence structure, of hearing impaired individuals proliferated (Brannon & Murray, 1966; Sarachan-Deily & Love, 1974; Thomas, 1972; Goda, 1964). Simmons (1962) reported that hearing impaired children were restricted in their choice of words. By developing type/token ratios (TTR) for each part of the speech, she found that hearing impaired children's TTRs in all categories were lower than their normal hearing counter- parts. Data revealed that the hearing impaired subjects used more determiners, nouns, and verbs but fewer conjunctions and auxiliaries than hearing children. They also used fewer different words within each category. For example, the hearing impaired subjects almost exclusively referred to a child in one picture as "boy," whereas the hearing children employed synonyms such as "young man" and "the kid." Furthermore, Simmons (1962) found that hearing impaired subjects tended to use an inflexible word order, exemplified, for instance, by their consistent use of adjectives in a predicate- adjective form. She characterized this as the result of "rubber stamping" by teachers as children became overly dependent on phrases which were taught in a specific form. Assessing the presence of 24 transformational rules in the spoken language of hearing impaired subjects, Holmes (1972) reported that hearing impaired subjects acquire the same syntactical rules by age 12 that normal hearing subjects do by age 4. By studying the acquisition and development of syntax in the spoken language of congenitally hearing impaired 5 through 13 year olds, Pressnell (1973) reported clear similarities between the hearing impaired and normally hearing in their order of syntactical acquisition in spoken utterances. Results she gained from the receptive and expressive portions of the Northwestern Syntax Screening Test_(NSST) (Lee, 1969) revealed that older hearing impaired children performed somewhat better than younger hearing impaired children; however, syntactical development began at a later age for hearing impaired children than it did for normal hearing youngsters and continued well into adoles- cence. Additionally, they required a longer period of time than hearing children did to complete a given stage of development. This was particularly true for the acquisition and usage of verb construc- tions. Pressnell postulated that these differences may arise from the fact that her hearing impaired subjects were typically taught verbs in a different order from hearing children and that their learning sequence was influenced by visual-auditory cues inherent in their more structured learning environment. Semantics Although the development of semantics has been widely studied among normal subjects, less attention has been given to it in the hearing impaired population. Applying a socio-linguistic approach to investigate the semantic component in spontaneous communication among hearing impaired preschoolers, Skarakis and Prutting (1977) reported that although their subjects' exhibited certain semantic functions in their spontaneous communication, they predominantly employed prelinguistic semantic behaviors found in younger hearing children. The most frequently occurring of these were performative, indicative object, and volition/negative volition. 0f the linguistic functions noted, action/state of agent and action/state of object occurred most frequently. Since it can be argued that prelinguistic semantic functions are precursors to formal linguistic development and that they provide a basic foundation upon which later language develops, these researchers suggested a continuum for the hearing impaired that parallels the semantic development of normal children as reported by Greenfield and Smith (1976). Curtiss, Prutting, and Lowell (1979) reported on 12 hearing impaired children ranging in age from 2 to 4 years that were expres- sing prelinguistic semantic functions through both verbalizations and made-up gestures. By expanding upon categories drawn from Greenfield and Smith (1976), these researchers found that although there was a general increase in hearing impaired children's expression of semantics as they grew older, specific semantic func- tions developed more slowly than normal hearing children across all age groups. 0f 13 selected semantic categories, only locate/name and performative were expressed more than 2% of the time by five child- ren in the 2 year old group, whereas eight such categories that included action or state of an agent, agent, aspect, locate/name, locative, object, performative, and volitional object were found in the 3 year old group. The 4 year old group expressed the same categories as the 3 year olds with the addition of negation. The hearing impaired children with a higher mean length of utterance (MLU) exhibited a wider range of semantic functions as they combined two or more categories into utterances that expressed semantic relationships between categories. while Greenfield and Smith's (1976) normal hearing children employed performatives ("the act that the speaker intends to carry out with his sentenceN--declaring, commanding, promising, asking questions, etc.,~Bates et al. [1977]) as their first semantic expression these hearing impaired subjects did not. It is possible that since young hearing impaired children often use pointing gestures to supply a semantic feature missing from a verbal utterance, they may pass through the normal stages of learning to speak using gestures for performatives instead of verbalizations. However, once performatives did emerge in their communication, the hearing impaired subjects used them much more fre- quently than any other category. They also relied on location more than Greenfield and Smith's (1976) normal hearing subjects. This might be explained by the fact that hearing impaired youngsters are tuned in to visual space and focus on location earlier develop- mentally or to a greater extent than hearing children (Curtiss et al. 1979). Pragmatics Again, as with semantic development among the hearing impaired, research into their pragmatic development is rare compared to that involving normal children (Bates, 1975; Halliday, 1975; Dore, 1973, 1974, 1975; Greenfield & Smith, 1976), Skarakis and Prutting (1977), in their socio-linguistic approach to describing the pragmatic component of four hearing impaired children's spontaneous communi- cation, found their subjects using the same communicative intentions as Greenfield and Smith's (1976) and Dore's (1974) younger hearing subjects. The intentions request/demand, description, attention and response occurred most frequently in communication as these subjects expressed intentions with a multiplicity of behaviors that included motor activity, gesture, combined gesture and vocalization or verbal- ization and verbalization alone. Curtiss et al. (1979) analyzed the speech acts, comprised of' simple gestures or one word utterances, of 12 hearing impaired preschoolers. All were found to exhibit the 16 pragmatic intentions drawn from a modified version of Dore's (1973, 1974, 1975) categories (e.g., demand, command, question, labelling, response to a question, response to a summons, response to a command, imitation, repetition, summons, description, protes- tation, ritual, request for approval, request for confirmation, and acknowledgment). This ability to display a wide range of pragmatic intentions by age 2 is found in hearing as well as hearing impaired children, thereby making a strong case for the human need to communi- cate . PRAGMATIC COMPETENCE: INDIRECT REQUESTS Within the field of child language research, the shifting emphasis from syntax to semantics to pragmatics has created some important changes in how we view the nature and source of communi- cation abilities. This view is especially important when we try to understand the communicative develOpment of the hearing impaired, hence the need for more research in this largely uncharted area. The term "pragmatics” was originally used by American pragmatist philosopher Charles Pierce (1932) and elaborated further by C. W. Morris (1946, p. 217) who defined it as "the relationship between signs and their human users." The principles of pragmatic analysis were viewed as a means of explaining use of language in context by Bates (1976). Contemporary interest came through the realization that structural and semantic analyses of language could not provide an adequate account of language development. Within the notion of pragmatics lies the identification and description of factors and rules that affect content and structure of the linguistic code. The first is that of function, or the intended purpose language serves (i.e., illocutionary force). Intentions fall into such general categories defined as declarative (tell), interrogative (ask), request (order), expressive (feel), and commissive (promise) (cf., Searle, 1969). Along with the intention 10 of the utterance, pragmatic analysis also looks at the locutionary form of the utterance and the effect that it has upon the listener (i.e., prelocutionary effect). In their pragmatic analysis of hear- ing impaired children's early communicative development, Curtiss et al. (1979) took form and effect into consideration by coding every identifiable communicative act performed by each child in their study. All communicative behaviors (e.g., utterance, gesture, facial expression, body movement, and vocalization) were able to be analyzed and placed into 16 categories (Dore, 1973, 1974, 1975) that were previously mentioned. Another pragmatic factor is context. It refers to the environ- mental or individual factors that influence the form, content, and intent of language. Environmental factors may include the social and/or situational variables such as who the speakers and listeners are, what goal the communication is supposed to accomplish, roles of the participants, presuppositions the participants have, time and place of the communication, and the events that occurred prior to the communication (Hopper & Naremore, 1978; Hymes, 1967). Individuals change the content and form of language as influenced by these variables. Indirect Requests Learning the rules for the use of language in context constitutes a major task in the psycho-socio-linguistic development of children. In order to develop "communicative competence," children must be able to smoothly communicate their intentions to others and to 11 interpret communication directed toward them. There must be a shared understanding of certain conversational postulates. One such postulate involves comprehension of the difference between the syntactic structure of an utterance and what is meant pragmatically (Dore, 1977). The distinction between these two aspects of communi- cative competence is perhaps best illustrated by the difference between direct and indirect requests. Direct requests reflect a surface structure that allows for clear interpretation of the speech act, whereas indirect requests reflect a situation where contextual cues such as intonational patterns of facial expressions are critical in the interpretation process. These cues are necessary because the surface structure of the sentence follows one form, but it is intended to be interpreted as another. An example of this would be the difference between the direct request “Open the door" and the indirect request "Wouldn't you like to open the door?9 The surface structure of the second seems to suggest an illocutionary force of an interrogative or asking; but the intent, if one is wanting to air out the room, is rather an imperative or request that someone open the door. According to Searle (1975) and Gordon and Lakoff (1975), individuals comprehend indirect requests in the following manner. The listener constructs the literal meaning of the utterance, checks it for plausibility and, if finding it implausible, applies a rule of conversation to obtain the conveyed meaning. Using a sentence/picture verification format with ten basic sentence pairs 12 that represented a number of syntactic/semantic/pragmatic categories, Clark and Lucy (1975) explored this notion with adults. Given a sentence on the left and a colored circle on the right, subjects were instructed to regard each sentence as a request to color the circle a certain color and to consider the circle on the right as a response to that request. Then, they had to decide whether the circle on the right had been colored according to the request and indicate their decision by pushing a “yes' to "no" button as quickly as possible. Thus, the basic measure was response latency time or the elapsed time between the onset of the sentence/picture diSplay to pressing the button. Error rate was used as a secondary . measure. By this method, Clark and Lucy (1975) determined that the longer the elapsed time for a response, the more difficult the request was to comprehend. Not only did they find evidence to support Searle's (1975) and Gordon and Lakoff's (1975) theory, but they also discovered that certain syntactic forms were most easily interpreted than others for their conveyed, indirect request meaning. For instance, verification of the interrogative types took longer and were more difficult to process than declarative types; and negative requests were more difficult to interpret than their corresponding positive requests. Although Clark and Lucy's (1975) adults con- firmed Searle's (1975) and Gordon and Lakoff's (1975) outline for comprehension of pragmatic meanings, developmental data from young children have not. Studies have shown that children seem to acquire the ability to comprehend indirect requests before the age at which 13 they are believed to comprehend the literal meanings of such (Bates, 1975; Shatz, 1974; Ervin-Tripp, 1977). Shatz (1975) reported child- ren responded accordingly to indirect requests as young as 2 years of age. She argued that the children did so because they were following an action—oriented process in which they "act out or act on what can be identified from the speech stream" (p. 101) not because of actual linguistic-pragmatic comprehension. For example, since children begin actively manipulating the world around them early on, their first response, upon hearing someone say "Can you shut the door?” would be to act out the words they understand and shut the door, thus giving the impression of comprehending an indirect response. Shatz (1975) explains that this stragegy recruits action responses to language unless some linguistic or nonlinguistic element indicates not to do so. Shata (1978a) further tested this hypothesis by examining videotapes of five mothers and their normal children aged 19-24 months while they talked about a toy they were playing with in a natural setting. Results revealed that the children responded with action to their mothers' requests for action when expressed in both the direct imperative form and the indirect directive form. Later Shatz (1978b) examined the responses of normal 19-34 month old children to sentences containing more than one interpretation. In one instance, the sentences were spoken in as neutral a linguistic context as possible; that is, the experi- menter presented the sentences in varied order so as to eliminate contextual cues. In the other experimental situation, contextual 14 information preceded the test sentences. For instance, the test sentence would be preceded by three or four direct imperatives to foster a directive interpretation or several informational ques- tions or statements to support an informational interpretation. Although context did affect the subjects' responses to test sen- tences, action responses were still common even in contexts that supported informational interpretations. The more linguistically sophisticated subjects showed more sensitivity to context and pro- duced less action ”errors" that the less sophisticated subjects. The results of these studies support Shatz's view that children gain early entrance into communicative interactions by responding to language with action and that development progresses as they learn to recognize contextual markers to stop action and integrate them into the process of interpretation. Using Clark and Lucy's (1975) procedures for the study of adult comprehension of different types of indirect requests, Carrell (1981) investigated how well normal 4 to 7 year old children under- stood the same requests. In a systematic, experimental approach, her study omitted any form of contextual cues and subjects relied only on linguistic forms for correct interpretation. As with adults in Clark and Lucy's (1975) report, Carrell found that children understood certain syntactic forms more easily than others. While the total percentage of correct responses (72.2%) indicated that children performed better than chance overall, closer inspection revealed that the above chance performance was not attained for all 15 of the indirect requests. Except for the request types "Can You?" and "Why not?", interrogative forms were more problematic for child- ren than declarative forms such as "It doesn't need." Children correctly averaged 82.5% for declaratives as compared to 66.7% for interrogatives. Additionally, children were especially influenced and confused by the surface or literal polarity of a conveyed request. For example, when conveyed meaning differed from literal meaning (e.g., Must you?, Why?, Shouldn't you?, Should you?, Doesn't it need?, Does it need?), children's responses consistently fell below 75% correct into the 50-60% chance range. Furthermore, positively conveyed requests (Please do) were generally more easily interpreted than their negative counterpart (Please don't) through all grade levels; and certain request types produced a greater number of errors than others. For example, while the request HI'll be happy unlesss . . .V gave children the most difficulty, its counterpart request "I'll be sad unless . . ." was correctly interpreted far more often. While this confusion was obvious across the four grade levels involved, a general developmental pattern of acquisition was apparent as subjects' performances improved with age. Indirect Requests and Language Impairment While Shatz's (1975) study investigated normal children's comprehension of positive indirect requests (Can you shut the door?), Leonard et al. (1978) studied language impaired children's compre- hension of indirect requests that were negative in nature (Can't you answer the phone?) and those with affirmative syntactic 16 construction but negative intention (Must you play the piano?). Sixty children ranging in age from 4 to 6 years who were judged to have language problems watched video taped interactions of indirect requests between two adults. After watching and listening to the interaction, the children were to decide whether the listener had acted appropriately according to the speakers' indirect requests. Results revealed that subjects had no more difficulty comprehending indirect requests involving negative syntactic constructions than those coded affirmatively. However, indirect requests using positive syntactic construction but negative intention did pose problems for the subjects. Judgments of 4 and 5 year olds were no higher than chance, whereas 6 year olds (who evidenced difficulties as well) seemed to better understand that certain indirect requests contained information in the predicate that specified an alteration in behavior. Following a modification of Shatz (1978b), Shatz, Shulman, and Bernstein (1980) examined language impaired children's responses to indirect directives in varying linguistic contexts. Just as with the normal 2 year olds, language impaired children had a tendency to employ action responses to indirect request. Qualitatively, they followed the same course of development but quantitatively there were some differences. The language impaired children had more difficulty producing informing responses than action ones even when they knew that an informing response was called for. For example, by experi- menting with a sequence of contextual sentences that preceded each 17 test sentence, these researchers determined that their language impaired subjects also had more difficulty than normals in utilizing prior linguistic context when deciding upon an interpretation. Because of the high preportion of primitive action strategies employed in response to an ambiguous context, it was hypothesized that this latter difficulty reflected problems in their ability to consistently process multiple input sentences over time. Prinz and Ferrier (1983) focused their investigation on the requesting abilities of language impaired children between the ages of 3 1/2 and 9 years. Using a speech model that investigated pur- pose, directness and surface form, these abilities were examined under three situations: (1) a role playing dyad, (2) production of requests in an experimental procedure using hand puppets, and (3) perception of requests using hand puppets. The experimental procedure was designed to assess the comprehension, production and judgment of polite requests. For assessment of production, children were shown a small chest which contained candies and told that if they asked a puppet nicely, they would receive a piece of candy. Each time they made a request, they were asked to say it Feven nicer" until a total of five requests were made to two different puppets. To assess judgment of polite forms, children were required to deter- mine which of two puppets made the most polite request. Finally, to measure comprehension, the children's ability to understand and comply with a series of requests to tidy up test materials was assessed. Subjects revealed a predominant usage of direct forms 18 with the older group using indirect forms only somewhat more. Overall results determined that language impaired children's pragmatic and syntactic abilities are closely linked to each other in the comprehension, production, and judgment of polite requests. The majority of these subjects were unable to effectively vary their type of polite requests and there was only a slight increase in their production of polite forms across the ages. However, they seemed to compensate by frequently using structures they had already acquired. In this respect, they may have recognized the need to be more polite but lacked the means to produce more indirect forms. In general, they Operated on a pragmatic level two or more years below chronological age. As demonstrated by the previous review, young normally develOp- ing children demonstrate communicative competency with indirect requests at an early age. Similarly, language impaired children follow normal developmental language milestones albeit at an older age and over a longer period of time. Although a significant amount of research has been concerned with grammatical competency in hear- ing impaired youngsters and a few studies (Skarakis & Prutting, 1977; Curtis et al., 1979) have investigated their overall produc- tive pragmatic competency, none have specifically focused on the subtle aspects of indirect request comprehension. The present study, as Carrell (1981) did with normally hearing children, was designed to focus on how hearing impaired children comprehend various types of indirect requests in a systematic, experimental mode, 19 relying only on linguistic forms and not on linguistic or non- linguistic contextual cues for information. Rationale for this approach was based on convincing data reporting that children's linguistic skills seem better than they actually are because of the naturalistic contextual cues provided in spontaneous communication (Bates, Camaioni, & Volterra, 1975; Shatz, 1974). The purpose of this investigation was to answer the following questions: How well do young hearing impaired children between the ages of 5 and 12 comprehend various syntactic forms containing indirect requests; and How does the polarity of indirect requests influence hearing impaired children's comprehension of them? METHOD Using the sentence protocol followed by Carrell (1981) and originally tested by Clark and Lucy (1975), ten different sentence pairs that represented a number of different indirect requests were presented to the children (see Table 1). The indirect requests varied according to sentence modality (i.e., declarative, impera- tive, and interrogative), as well as polarity of conveyed meaning (i.e., negative vs. positive). Stimuli From the ten basic sentence pairs shown in Table 1, 40 different sentences were constructed. Forty sentences were utilized in order to insert the word "blue" or "red" in each. This was done so that the children had equal opportunity to color a circle blue and to color a circle red, thereby ruling out any built-in systematic bias. These sentences were originally tested by Clark and Lucy (1975) and Carrell (1981). In terms of polarity of the conveyed meaning of each request, sentence pairs consisted of a positive and corresponding negative member. The first member, or (a) of the pair conveyed a positive request to color the circle the named color. The second member, or (b), conveyed a negative request to color the circle the opposite color. Rationale for choosing such pairs is based on data about explicit positive and negative English 20 Table 1 21 Pairs of Sentences Used as Experimental Stimuli with Polarity of Conveyed Meaning Noted 1(a) (b) 2(a) (b) 5(a) (b) 6(a) (b) 7(a) (b) Please color the circle blue. (red) Please don't color the circle blue. Can you make the circle blue? Must you make the circle blue? Why not color the circle blue? Why color the circle blue? (red) (red) (red) (red) (red) I would love to see the circle colored blue. (red) I would hate to see the circle colored blue. (red) You should color the circle blue. (red) You shouldn't color the circle blue. (red) Shouldn't you color the circle blue? (red) Should you color the circle blue? (red) The circle really needs to be colored blue. (red) The circle doesn't really need to be colored blue. (red) Positive Negative Positive Negative Positive Negative Positive Negative Positive Negative Positive Negative Positive Negative 22 Table 1 Continued 8(a) (b) 9(a) (b) 10(a) (b) Doesn't the circle really need to be colored blue? (red) Does the circle really need to be colored blue? (red) I'll be very happy if you make the circle blue. (red) I'll be very sad if you make the circle blue. (red) I'll be very sad unless you make the circle blue. (red) I'll be very happy unless you make the circle blue. (red) Positive Negative Positive Negative Positive Negative 23 sentences by Clark and Chase (1972). By including corresponding positive and negative indirect requests, it was possible to test for any differences in the children's response to the different polarity of conveyed meaning. With regard to syntactic form, (1a,b) and (5a,b) were imperatives; (2a,b), (3a,b), (6a,b), (8a,b) were interrogatives; and (4a,b), (7a,b), (9a,b), and (10a,b) were declaratives. Subjects Subjects included 17 moderately-severe to profound hearing impaired youngsters (i.e., 56 to 100 dB HL in the better ear; [ANSI, 1969]) as determined by hearing tests given within six months of the experiment. These children were placed into one of four groups, depending on grade placement. Group I consisted of four kindergardeners, Group II had three first and one second graders, Group III consisted of four third and one fourth graders, while Group IV had two fifth and two sixth graders. All of the subjects came from homes where English was the only language spoken, and while they were attending school all were required to wear FM audi- tory trainers. The subjects were enrolled in an oral education program of a metropolitan Michigan city which serves children from diverse socio-economic backgrounds. In order to be enrolled for the oral education program, the children were required to demonstrate normal nonverbal intelligence by their performance on a range of intelligence tests, including the Leiter International Performance Scale (Leiter, 1969) and the WISC-R (Wechsler, 1974). 24 This information was used to confirm the child's normal nonverbal intelligence for participation in this research project. Further- more, subjects exhibited no additional known handicaps (e.g., visual impairment, cerebral palsy, learning disability). For the purposes of obtaining current information regarding each child's oral language performance, the Test of Language Develop- ment (TOLD) (Newcomer & Hamill, 1982) was administered prior to par- ticipation in the study. The TOLD is a multifaceted language test from which a receptive language and an expressive langauge age can be derived. From the results of the children's responses to the “TOLD, each child was assigned a receptive language age (LAR) and an expressive language age (LAE). In addition to chronological age, hearing age was also calcu- lated for each child. Hearing age was defined as the amount of time between the date the child received a hearing aid and the date of participation in the study. For example, Subject 1 had a chrono- logical age of 4 years, 9 months (DOB 4-20-81). She did not receive her hearing aid until 2-85, thereby having a hearing age of 1 year as of the time of her participation in the experiment. A descriptive summary of the subjects is presented in Table 2. Specific characteristics of each group were as follows: Group I consisted of four children--two males and two females. All were enrolled in kindergarden and ranged in age from 4;9 to 5;11 with a mean of 5;3 (SD 6.29 month). The mean hearing age of the children in Group I was 2:8 (SD 21.39 months). Their mean receptive language 25 HN.m N¢.m Hm. oo.mH oo.¢H me.o Hm.mH mN.m mm.H om oo.mN oo.m oo.mH mwm mmo He om mmm mHN m mN m NH mmm mmo cm ooH mam mmu N e «m 0H NH Nun emw mm on va mm“ H m oN N mH mac Nmo mm on mmm man H N HN e NH mmm Hum me om Nmo HHH H H HH azocw mH.N em.¢ m¢.e mm. m.N mm.oN eo.mH mm.HN mN.m om oo.0N mN.m mN.oH Nmm emm Hm mm wHN mmm m MN N mH mum Nmm om om mum HHmm x q 0H o mH mum Nnm me mm one mmm x m NH w oH mmm Nam mN mm «HN one x N ON 0 ON Hmm Nmm oe om oHH one x H H azoco Ho» umz mom NHpmmwz .Hmoav xmmp HmpcmeHLmQxN :H mcoum m>HHHmoa .HNHmmmgqmu mm< mmmzmcmg .HN<49 m>Humuumm mm< mmzmmcmH .HH:m4 mcwcmmz umcm< .HHI<=v Hm>mH mcHmez umuwmcz .H mcmonHou mgu Loo name one «Ham» Humnnzm Hm=UH>HucH N mHan m¢.e mm.m mm. Hm.NH eN.m NN.m m¢.NH ow.mN mH.m om mN.om oo.mH mN.NH mum mmN mN mN onN oHHHH m Nm NH mH mmm HHN mH om Nae NHNH o a NN HH NH mmm mmm mN mN oHHN mHHH m m NN oH NH mmm mnN om om HHmm mHHH m N Hm mH NH Hum NHN co co mum mmHH m H >H aaogw eo.H mo.m om.N «N.mH mo.NH oo.N Nm.o ¢¢.mH NN.oH om ON.mN om.¢ mm.NH Nue mmm cw om HHHm mww m mN m oH mmm mmm mm om mmm mHoH c m mN m ON cum Nmm mm om HHN onw m c NN N ON mmm mHN om ooH mHN mmm m m 0N HH mH mmm mmo mm ow Nwo Nmm m N NN m NH mac mme me om oHH¢ NHN m H HHH aaocu HoN wmz mo; «<4 mqg HI< Hz<= <1 cou HHm>Hummmz new HHm>HuHmoa op mmmcoammm m.:mLcHH;u com memo: Quota .H mgzmwm >_ 38.0 =. 95.0 __ 95.6). 986 . — m 0—. 0—. ON 91003 III-IIIIIIIIII 3 mo: uIIJmoa 34 was 38.7 (df . 1, p < .001). Only Group IV revealed a nonsignificant difference in the children's comprehension of positively versus negatively conveyed indirect requests (x2 = 2.63; df - 1, p < .05). Children's Comprehension of Indirect Requests Expressed in Different Sentence Modalities As previously mentioned, the sentence modalities of indirect requests used in this study were imperative, declarative, and interrogative. The following analysis was competed to investigate the children's comprehension of indirect requests relative to these sentence modalities. The results of subjects' comprehension of imperative, declarative and interrogative indirect requests are displayed in Table 4. In short, the children found imperative requests the easiest to comprehend (89% total), followed by declarative types (66% total) and finally by interrogative indirect requests (49% total). The individual groups followed the same pattern of comprehension. Correlational Analysis A Spearman Rank Order correlation was performed to investigate the relationships among the variables of chronological age, hearing age, unaided hearing level, aided hearing level, receptive lan- guage age, expressive language age, positive score on the experi- mental task, negative score on experimental task, and total score on experimental task. Results of these correlations are found in Table 5. 35 Table 4 Percentage of Children's Correct Comprehension of Indirect Requests Relative to Different Sentence Modalities Group Imperative Declarative Interrogative I 56% 50% 45% II 100% 71% 45% III 100% 59% 47% IV 100% 87% 58% Total 89% 66% 49% 36 moo. v a««* Ho. v are mo. v or 11 Ho» rrrow. .. uwz No.1 *om.f 11 mo; «x5mm. «««NN. 0H.1 11 N Homwnzm com mcoHpmHmLcoo gouge xcmm cmscmmam m anmh 37 The correlational analysis indicated that the relationship between chronological age and hearing age was .67 (p < .005). The relationship between chronological age and receptive language age was .70 (p < .005). The variables of chronological age and expres- sive language age demonstrated a relationship of .66 (p < .005). While the relationship between chronological age and negative score was .48 (p < .05) and the relationship between chronological age and total score was .73 (p < .005), there was a nonsignificant relation- ship of .14 between chronological age and positive score. These results indicate that of all the above-mentioned variables, only the subjects' score on the positively conveyed indirect requests was not influenced by advancing age. As previously noted, hearing age was not significantly related to any of the other variables, except chronological age. Of particular interest was the finding that hear- ing age was not significantly related to either receptive language age (r = .41) or expressive language age (r = .21). Correlational analysis also demonstrated that the relationship between unaided hearing level, receptive language age, expressive language age, as well as the subjects' positive, negative and total score was nonsignificant. For instance, the relationship between unaided hearing level and subjects' score on negatively conveyed items was -.43 and -.17 for unaided hearing level and total score; neither were significant. Results did, however, indicate a signifi- cant positive relationship between unaided hearing level and aided hearing level (r = .81; p < .005). This relationship indicates that the improvement in hearing due to presence of a hearing aid is 38 dependent upon subjects' initial unaided hearing level. Results also indicated that a significant relationship existed between subjects' aided hearing levels and their score on the negative stimuli. This relationship was -.61 (p < .01). The relationship between aided hearing level and variables other than unaided hear- ing level was nonsignificant. A positive relationship existed between receptive language age and the subjects' negative scores (r = .68; p < .005) as well as a positive relationship between receptive language age and the subjects' total score (r = .86; p < .005). However, a nonsignifi- cant relationship of .01 was found between receptive language age and positive score. The children's expressive language age was significantly related to both their score on the negatively conveyed indirect requests (r = .77; p < .005) and their total score (r = .83; p < .005). A nonsignificant relationship of -.16 was found between expressive language age and score on positively conveyed items. The correlational analysis also found a high positive relation- ship between receptive and expressive language performance (r = .91; p < .005). Finally, the relationship between the children's nega- tive score and their total score was .80 (p < .005), whereas it was -.02 (nonsignificant) for the positive and total score values. DISCUSSION This study was conducted to investigate young hearing impaired children's ability to comprehend various types of indirect requests. Although there was a small number of subjects, results indicate that there was a definite developmental effect, with the older hearing impaired children exhibiting greater comprehension of indirect requets than the younger ones. As a group, all 17 of the subjects tended to comprehend positively conveyed requests more readily than negatively conveyed requests. They also experienced greater difficulty inter- preting interrogative indirect request forms than they did imperative or declarative forms. These findings are consistent with Carrell's (1981) normally devel0ping, low hearing impaired subjects. Her data demonstrated that older children were better able to comprehend indirect requests than were younger children. This developmental trend was noted across the grade level span of preschool to second grade (approxi- mate age range was from four to seven years), Carrell's subjects, however, exhibited greater comprehension at earlier chronological ages than did the hearing impaired children in this study. Whereas normally hearing children achieved a 78% comprehension level by Grade 1, the hearing impaired children did not reach a comparable level (76.8%) until Grades 5 and 6. The overall mean comprehension 39 40 score for the normal hearing subjects in Carrell's study was 77.2%, whereas it was 62.5% for the hearing impaired children of the present research. None of the groups of hearing impaired subjects ever reached the 92% comprehension level that Carrell's oldest group of second graders achieved. A slightly different picture emerges when one compares the sub- 5 jects of the present study to those of Carrell's on the basis of hearing age. While the mean chronological age for Group I was 5;3, their mean hearing age was only 2;8. This young hearing age may very .- . '4. l;*|:‘\“‘_ well account for the group's low rate of only 50% correct on the L. experimental task as compared to Carrell's youngest group of subjects (CA = 4 years) who attained a correct response rate of 64.5%. This difference in percent correct between the hearing impaired children and Carrell's hearing children was 7.5% for Group II, 20% for Group II, and 16% for Group IV, thereby demonstrating a tendency for the difference to become greater as the hearing impaired children became older. However, even when one compares the results of the two studies on the basis of hearing age, the hearing impaired child- ren yet performed below Carrell's normal hearing subjects. The present finding that negatively conveyed requests are more difficult for hearing impaired children to process than are posi- tively conveyed requests, compares favorably to other research. Quite recently Paul and Cohen (1985) demonstated a similar pattern in persistent developmentally delayed and mentally retarded sub- jects. Their investigation was also modeled after Carrell's who 41 reported that normal developing children also comprehended posi- tively conveyed requests more so than they did the negative counter- parts. The correlational results of the present study support the view that the more difficult to comprehend negatively conveyed indirect requests are a more sensitive measure of children's ability to comprehend indirect requests. As noted earlier in the results section, the children's score on comprehension of positively con- veyed indirect requests was not significantly related to other vari- ables such as chronological age of language performance. However, their comprehension of the negatively conveyed indirect requests was significantly related to other developmental variables such as chronological age and language performance. The ability to compre- hend positively conveyed indirect requests appears to be achieved quite early. Normal hearing, developmentally delayed and hearing impaired subjects tend to reach a ceiling of understanding of posi- tively conveyed indirect requests early in development. 0n the other hand, effective comprehension of negatively conveyed requests requires a longer amount of time. Similar to Paul and Cohen's (1985) developmentally delayed and Carrell's (1981) young normal subjects, the hearing impaired subjects of the present investigation also demonstrated a rank ordering of comprehension of indirect requests relative to different sentence modalities. Like the normal hearing and developmentally delayed subjects of the previously noted research, 42 these hearing impaired children found it easier to comprehend indirect requests that were of an imperative nature than those couched in a declarative form. Interrogative indirect requests presented the most comprehension difficulty for subjects. In light of research into other areas of language abilities in hearing impaired children, the results of this study were not unexpected. Investigations have consistently pointed to evidence that hearing impaired children acquire language in a similar develop- mental pattern as normal hearing children but at a slower rate and over a protracted period of time. For example, Laubscher (1969) reported this finding in his study with hearing impaired children's developing morphological trends. Holmes (1972 and Pressnell (1973) also noted this phenomenon in their investigations into hearing impaired children's acquisition of syntactic rules. Finally, Skarakis and Prutting (1977) as well as Curtiss et al. (1979) noted that hearing impaired children's pragmatic abilities develop along normal patterns albeit at a slower rate. Although it was expected that this research might find delayed comprehension of indirect requests as well as a developmental pro- gression among hearing impaired children, it was, at first, somewhat puzzling as to why Group III (consisting of third and fourth graders) performed lower than Group 11 (consisting of first and second graders). Since the total mean score for Group II was higher than the total mean score for Group I and the total mean score for Group IV was higher than the total mean scores for Groups I, II, and 43 111, it did not follow that Group 11 would perform better than their older counterparts in Group III. Closer examination of the raw data revealed that scores for subject three in Group II might con- tain a clue. This child, whose chronological age was 7;5, had a receptive language age of 8;4. He achieved a score of 18 for the positively conveyed items and on negatively conveyed indirect requests, a score of 16. His total score of 34 on the experimental task. This total score was exceeded only by one other subject whose total score was 37. This child was in Group IV. His chronologial age was 12;8 and his receptive language age was 8;1. Upon consultation with the teacher of subject three in Group II, it was determined that this child was extremely bright and his presence in Group 11 seems to have contributed to the absence of a developmental progression between Group II and Group 111. If one only considers the total scores for subjects 1, 2, and 4 in Group II, it is evident that there is a developmental pattern between Groups I, II, and III. The results from this study have several implications. First, the data indicate that this type of experimental task is sensitive to the language growth and development of hearing impaired children; with the negatively conveyed indirect requests providing the most useful information. Speech-language professionals as well as edu- catores should be aware that many children, regardless of hearing ability, have difficulty comprehending certain types of indirect requests. These professionals may need to monitor their own produc- tion of indirect requests in order to assure that children are 44 comprehending the desired communicative intent. Carrell (1981) suggests that there may actually be too many teachers that give "coaching questions" or "gently reminder" questions (e.g., "why not work on your math?" or "should you be sitting there?") that are intended to keep students on task. These types of questions may actually be confusing to the children in their day—to-day classroom performance. Research is needed in which “teacher talk" is analyzed for the frequency and type of indirect requests actually used. This line of inquiry should also note children's comprehension of such indirect requests in context. This phenomenon needs to be investi- gated with classroom teachers of regular children as well as class- room teachers of the hearing impaired. In conclusion, this research demonstrated that hearing impaired children's ability to comprehend indirect requests progresses sig- nificantly across the grades of kindergarden through six. However, it appears that their comprehension of indirect requests is not completed by grades five and six. In fact, they are not demonstrating comprehension of indirect requests as highly developed as normal hearing second graders. This result supports the need for further research into the indirect request comprehension abilities of older hearing impaired children. It would be of interest to discover how highly developed these abilities are by the end of their high school years. Finally, in comparing these data with that of Carrell (1981), it appears as if hearing impaired children between grades kindergarden through six are following a similar developmental 45 pattern as normal hearing children between preschool and second grade. REFERENCES 46 REFERENCES American National Standards Institute. (1970). Specifications for audiometers, ANSI S3.6-1969. New York: American NatiOnal Standards Institute, Inc. Bates, E. (1976). Pragmatics and sociolinguistics in child language. In M. Morehead & Morehead (Eds.), Language deficiengy in children. Baltimore: University Park Press. Bates, E., Camaroni, L., & Volterra, V. (1975). The acquisition of performatives prior to speech. Merrill-Palmer Quarterly, 21, 205-226. Bates, E., Benigni, L., Bretherton, I., Camaioni, L., & Volterra, V. (1977). From gesture to the first word: On cognitive and social prerequisites. In M. Lewis & L. Rosenblum (Eds.), Interation, conversation, and the development of language. New York: John Wiley & Sons. Brannon, J. B., Jr., & Murray, T. (1969). Linguistic word classes in the spoken language of normal, hard-of-hearing and deaf child- ren. Journal of Speech and Hearing Research, 2, 504-610. Carrell, P. (1981). Children's understanding of indirect requests: Comparing child and adult comprehension. Journal of Child Language, 8, 329-345. Chomsky, N. (1957). Syntactic structures. The Hague: Mouton. Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge: MIT Press. Clark, H. H., & Chase, W. G. (1972). On the process of comparing sentences against pictures. Cognitive Psychology, 2, 472-517. Clark, H. H., & Lucy, P. (1975). Understanding what is meant from what is said: A study in conversationally conveyed requests. Journal of Verbal Learning and Verbal Behavior, 13, 56-72. Curtiss, S., Prutting, C., &Lowell, E. (1979). Pragmatic and semantic development in young children with impaired hearing. Journal of Speech and Hearing Research, 22, 534-552. 47 48 Dore, J. (1973). A develOpmental theory of speech act production. Transactions of the New York Academy of Sciences 11, 25, 623-630. Dore, J. (1974). A pragmatic description of early language develop- ment. Journal of Psycholingoistic Research, 2, 343-350. Dore, J. (1975). Holophrases, speech acts, and language universals. Journal of Child Langyage, 2, 21-40. Dore, J. (1977). Children's illocutionary acts. In R. Freedle (Ed.), Discourse production and comprehension. Norwood, J.J.: Ablex. Ervin-TripP. S. (1977). Wait for me, roller-skate. In S. Ervin- Tripp & Mitchell-Kernan (Eds.), Child discourse. New York: Academic Press. Ferguson, G. (1976). Statistical analysis in psychology and educa- tion. New York: McGraw-Hill. Garber, C. (1967). An analysis of English morphological abilities of deaf and hearing children. Unpublished doctoral dissertation, Ohio State University, Columbus, Ohio. Goda, S. (1964). Spoken syntax of normal, deaf, retarded adolescents. Journal of Verbal Learning & Verbal Behavior, 2, 401-405. Gordon, 0., & Lakoff, G. (1975). Conversational postulates. In P. Cole 5 J. Morgan (Eds.), Syntax and semantics: Speech acts. New York: Academic Press. Greenfield, P., & Smith, J. (1976). Communication and the beginnings of language: The development of semantic structures in one-word §peech and ooyond. New York: Academic Press. Halliday, M. A. K. (1975). Learning how to mean: Explorations in the development of language. London: Edward Arnold. Holmes, 0. (1972). The use of structured imitation in the assess- ment of deaf children's syntax. Unpublished doctoral disserta- tion, Syracuse University, Syracuse, New York. Hopper, R., & Naremore, R. C. (1978). Children's speech: A practical introduction to communication development. New York: Harper & Row. Hymes, D. H. (1967). The functions of speech. In J. P. DeCecco (Ed.), Thepsychology of language, thought and instruction. New York: Holt, Rinehart & Winston. 49 Kretschmer, R., & Kretschmer, L. (1978). Language development and intervention with the hearing impaired. Baltimore: University Park Press. Laubscher, A. (1969). Language acquisition of a group of ten hear- ing impaired children. Unpublished manuscript, Central Institute for the Deaf, St. Louis, MO. Lee, L. (1969). Northwestern syntax screening test. Evanston, IL: Northwestern University Press. Leonard, L., Wilxoc, M. J., Fulmer, K., & Davis, G. A. (1978). Understanding indirect requests: An investigation of children's comprehension of pragmatic meanings. Journal of Speech and HearingResearch, 2;, 528-537. Leiter, R. (1969). Leiter International Performance Scale. Los Angeles: Western Psychological Services. Ling, D. (1972). Auditory discrimination of speech transposed by a sample-and-hold process. In G. Fant (Ed.), Speech communication ability and profound deafness. Washington, 0.0.: A. G. Bell Association for the Deaf. Morris, C. W. (1946). Signs, language and behavior. Englewood Cliffs: Prentice-Hall, 1946. Newcomer, P., & Hammill, D. (1977). Test of language development. Austin, TX: Empiric Press. Oller, D. K., Jensen, H., & Lafayette, R. (1978). The relatedness of phonological processes of a hearing-impaired child. Journal of Communication Disorders,_ll, 97-105. Paul, l2., 3 Cohen, 0. (1985). Comprehension of indirect requests in adults with autistic disorders and mental retardation. Journal of Speech and Hearing Research, 22, 475-479. Pierce, C. (1932). Collected papers of Charles Sanders Pierce. In C. Hartshorne & P. Weiss (Eds.), Elements of logic, Vol. II. Cambridge, MA: Harvard University Press. Pressnell, L. (1973). Hearing impaired children's comprehension and production of syntax in oral language. Journal of Speech and Hearing Research, lg, 12-21. Prinz, P., & Ferrier, L. (1983). "Can you give me that one?“: The comprehension production and judgment of directives in language- impaired children. Journal of Speech and Hearing Disorders, 48, 44-54. SO Sarachan-Deily, A., & Love, R. (1974). Underlying grammatical rule Structure in the deaf. Journal of Speech and Hearing Research, 17, 89-698. "’ Searle, J. (1969). Speech acts. Cambridge, England: Cambridge University. Searle, J. (1975). Indirect speech acts. In M. Cole & J. Morgan (Eds.), Syntax and semantics. New York: Academic Press. Shatz, M. (1974). The comprehension of indirect directives. Can two-year-olds shut the door? Paper presented at Summer LSA Meeting, Amherst, Massachusetts. Shatz, M. (1975). How young children respond to language: Pro- cedures for answering. Papers and Reports on Child Language Development, 10. Shatz, M. (1978a). Children's comprehension of their mother's directives. Journal of Child Language, 5, 39-46. Shatz, M. (1978b). On the development of communicative understand- ings: An early strategy for interpreting and responding to messages. Cognitive Psychology, 19, 271-302. Shatz, M., Shulman, M., & Bernstein, D. (1980). The responses of language disordered children to indirect directives in varying contexts. Applied Psycholinggistics, 2, 295-306. Siegel, S. (1956). Nonparametric statistics. New York: McGraw- Hill. Simmons, A. A. (1962). A comparison of the type-token ratio of spoken and written language of deaf and hearing children. Volta Review, o4, 417-421. Skarakis, E., & Prutting, C. (1977). Early communication: Semantic functions and communicative intentions in the communication of the preschool child with impaired hearing. American Annals of the Deaf, 382-391. Thomas, S. (1972). A compariative study of the development of seven verb phrase features in hearing-impaired and normal- hearing preschool children. Unpublished master's thesis, University of North Carolina at Chapel Hill. Wechsler, D. (1974). Wechsler Intelligence Scale for Children. New York: Psychological Corp.