Clinical Guide to Assessment & Treatment of Communication Disorders PDF

Chapter 6 Children with Specific Language Impairment Introduction Speci!c language impairment (SLI) has been described as a signi!cant language impairment that has no obvious cause and that cannot be attributed to anatomical, physical, or intellectual problems (Owens, 2010). Although it is a preva...

Chapter 6 Children with Specific Language Impairment Introduction Speci!c language impairment (SLI) has been described as a signi!cant language impairment that has no obvious cause and that cannot be attributed to anatomical, physical, or intellectual problems (Owens, 2010). Although it is a prevalent disor- der in childhood, it often goes unrecognized or masquerades as inattention or something worse (Leonard, 2014). Many labels have been used to describe the condition including “aphasia,” “dysphagia,” “developmental aphasia,” “infantile speech,” “delayed language,” “deviant language,” “developmental language impairment,” “speci!c language de!cit,” “language disorder,” “expressive lan- guage disorder,” “expressive-receptive language disorder,” “language-learning dis- ability,” “language-learning impairment,” and “primary language impairment.” The purpose of this chapter is to describe the nature, possible causes, assess- ment, and treatment of the condition that we will refer to as “speci!c language impairment” or SLI. The term is not without its challenges. Even if weaknesses in other areas are “subclinical,” impairment is rarely “speci!c” to the domain of lan- guage. On the other hand, SLI is the most widely adopted term at present, it is not mistaken for conditions such as autism or intellectual disability, and it avoids the impression that the weakness in language is minor or temporary (Leonard, 2014). SLI Described Diagnostic Criteria According to the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5; American Psychiatric Association, 2013) (who uses the term “language disorder” to distinguish this condition from “social communication disorder” and “speech disorder”), SLI is characterized by © Springer International Publishing AG, part of Springer Nature 2018 53 P. A. Prelock, T. L. Hutchins, Clinical Guide to Assessment and Treatment of Communication Disorders, Best Practices in Child and Adolescent Behavioral Health Care, https://doi.org/10.1007/978-3-319-93203-3_6 54 6 Children with Specific Language Impairment Table 6.1 Exclusionary factors required for a diagnosis of SLI Exclusionary factor Criterion Nonverbal intelligence Nonverbal IQ of at least 85 or score that remains above 70 after test error is accounted for Hearing sensitivity No indication of hearing impairment; hearing tests are passed at conventional levels Recurrent otitis media This factor is unlikely to be a cause of SLI, but to interpret language status, there should not be recent evidence of repeated episodes Oral structure and No structural or functional abnormalities function Interaction with people No symptoms or impaired reciprocal social interaction or restriction of and objects activities Neurological function No evidence of neurological disorder Adapted from Kadervek (2011) and Leonard (2014) dif!culties in acquiring and using language across modalities (spoken, written, sign language, or other). These dif!culties are apparent in early development, due to de!cits in language comprehension or production, and include reduced vocab- ulary (word knowledge and use), limited sentence structure (ability to put words and word endings together to form sentences based on the rules of grammar and morphology), and impairments in discourse (ability to use vocabulary and con- nect sentences to explain or describe a topic or series of events or have a conversation). Finally, clinical identi!cation of SLI is contingent on the absence of any known causal factors. In other words, alternative explanations of the language impairment must be ruled out before a diagnosis of SLI is appropriate. The traditional exclu- sionary factors for SLI are presented in Table 6.1. Language Profiles As stated above, the DSM characterizes SLI by persistent dif- !culties in the acquisition and use of language due to de!cits in comprehension or production that include vocabulary, limited sentence structure, and discourse. Although these criteria indicate de!cits in comprehension or production, children with SLI tend to show one of two patterns: good comprehension relative to impaired production or depressed scores in both areas. Note also that the effects of SLI extend beyond the areas (vocabulary, sentence structure, discourse) indicated in the DSM- 5. Functioning in these areas will be described !rst but will be followed by a discus- sion of additional language de!cits characteristic of SLI. Although SLI is known to affect individuals differently, the modal pattern reveals areas of particular language weaknesses. With regard to vocabulary, children with SLI show a slower pattern of vocabulary growth, tend to rely heavily on a small handful of all-purpose nouns (e.g., “thing”) and verbs (e.g., “go,” “do,” “make”), and have special dif!culty acquiring verbs more generally. Children with SLI tend to be slower and less accurate in naming tasks, some exhibit word-!nding problems or anomia (Newman & German, 2002), and the word descriptions of children with SLI often lack detail when compared to those of their typically developing (TD) SLI Described 55 peers (Mainela-Arnold, Evans, & Coady, 2010). These vocabulary de!cits sometimes result in circumlocution, a clumsy and roundabout way of talking. Bernstein-Ratner (2013, p. 293) offers the following examples of circum- locutions occurring in the context of naming tasks: “something round and English” (for an English muf!n), “on my brother’s pants” (zipper), and “you eat breakfast with it” (spoon). With regard to sentence structure, de!cits in morphology (putting words together) and syntax (putting sentences together) are considered the primary de!cits of SLI. As it was with vocabulary, children with SLI are delayed in morphosyntactic development, and although they generally acquire morphosyntactic features in the same developmental sequence as TD children, certain structures prove to be espe- cially problematic. A list of morphosyntactic features that are dif!cult for English- speaking children with SLI is presented in Table 6.2. As seen in Table 6.2, many errors appear as errors of omission (e.g., leaving out past tense -ed marking of verbs), whereas others appear as errors of substitution (e.g., swapping “me” for “I”). However, it is important to note that these patterns are not always consistent and can be influenced by other elements (e.g., sentence or word length and complexity, sound production requirements, familiarity of terms) that occur in a sentence (Krantz & Leonard, 2007). Discourse includes narrative discourse (the ability to construct stories) and conversational discourse (the ability to negotiate conversations). Successful Table 6.2 Common morphosyntactic de!cits in English-speaking children with SLI Morphosyntactic feature Example of error -ing (present progressive verb) “Dog eat him food.” (The dog is eating his food”) -s (plural s) “She have two book” (She has two books) -ed (past tense) “I walk already” (I walked already) Wh- questions “What do you think what is in here?” (What do you think is in here?) “What we can make?” (What can we make?) Prepositions “in” and “on” “Mommy put table, my book” (Mommy put my book on the table”) Demonstrative (this, that, these, those) “This mine!” (This book is mine) without paired noun Pronouns “Him going” (He is going) Auxiliary (AKA “helping”) verbs (e.g., is, “Patty do it” (Patty can do it) do, can) “Be” verbs “Me Batman today” (I am Batman today) “She baking” (She is baking) Articles (a, the) “I want sticker” (I want a sticker) Possessive ‘s “That Mommy coat” (That is Mommy’s coat) Third-person singular verbs “Daddy !x cars” (Daddy !xes cars) Complementizers (to) “I need go now” (I need to go now) Three element noun phrases (determiner + “The girl here” “The girl big” (The big girl is here) adjective + noun) Adapted from Kadervek (2011) and Owens (2010) 56 6 Children with Specific Language Impairment discourse is dif!cult as it places extreme demands on working memory (e.g., planning and sequencing events) and requires integration of vocabulary and morphosyntactic knowledge, the understanding of conventional story structures and elements, and a sense of audience. The acquisition of discourse skills is extremely delayed in children with SLI. Compared to those of their TD peers, the narratives of children with SLI are less complete, less coherent, and more confusing. With regard to conversation, children with SLI may offer inappropri- ate responses to a topic and have trouble securing conversational turns and repairing conversational breakdowns (Owens, 2010). One might expect that children who have trouble acquiring words, construct- ing words and sentences, telling stories, and negotiating conversations will also have problems in many social situations. It is not surprising then that many chil- dren with SLI show marked impairments in pragmatics (the social use of lan- guage). Pragmatic de!cits “seem to directly reflect impairments in language knowledge and use” (Bernstein-Ratner, 2013, p. 295). As a result, children with SLI may act or otherwise seem like younger TD children. Although they use the same pragmatic functions (e.g., requesting, informing) as their TD peers, they tend to express and respond to them less effectively. They may display less sensi- tivity to a partner’s need for information or clari!cation and be less able to adapt language to suit the needs of the listener. Some children with SLI demonstrate a tendency to interpret language literally, which can have serious social conse- quences. As Bernstein-Ratner (2013) illustrated: “One child with SLI responded to the subtle indirect request for sharing implied by ‘Your snack looks good’ by responding ‘Yes, and it tastes good too!’. Such a tendency toward literal interpre- tation will lead a child to ignore the intent behind many conversational gambits… [and] may alienate peers unintentionally” (p. 295). Finally, phonology (mastering sound patterns in a language) is impaired in approximately 40% of children with SLI (Beitchman, Nair, Clegg, Ferguson, & Patel, 1986). As toddlers, children with SLI tend to vocalize less and have less varied and less mature syllable structures (see Chap. 2 for description and exam- ples). Children with SLI often show phonological processes (see Chap. 2) similar to those of younger TD children, but the phonological problems in SLI continue to be problematic in later years and result in reduced speech intelligibility. Not surprisingly, children with SLI also show de!cits in phonological awareness (how sounds of the language match to letters in the alphabetic reading and writing system) which, when combined with de!cits in higher level language skills, leads to reading and writing problems and academic failure. Children with SLI can also be very disfluent and may repeat sounds and words so often that they are mistak- enly labeled as stutterers (Finneran, Leonard, & Miller, 2009; see Chap. 9 for more information about stuttering). One particularly dif!cult aspect of phonol- ogy in SLI is poor nonword repetition (e.g., “Repeat after me: flimik”). Challenges in nonword repetition have proven so persistent that some have described it as one of the best diagnostic indicators of SLI (e.g., Owens, 2010; Weismer & Thordardottir, 2002). Associated Problems Although diagnosis of SLI excludes children with obvious motor or neurological impairment, research has found evidence of neurological soft SLI Described 57 signs (Kadervek 2011). Neurological soft signs are not detectable via brain scan but are instead evident in a higher incidence of dif!culty with chewing and sucking (Whitehurst & Fischel, 1994). These signs may be interpreted as symptoms of oral- motor weakness or slight motor differences which might be described as “clumsi- ness” or “accident prone” (Kadervek, 2011). Depressed social skills can be seen in SLI early in development. Compared to TD school-aged children, children with SLI are less successful at initiating play (Liiva & Cleave, 2005) and show more reticence to engage in cooperative play (Hart, Fujiki, Brinton, & Hart 2004). They are also perceived as having poor emo- tion regulation (Fujiki, Brinton, & Clarke, 2002), are less likely to be selected as preferred classmates (Gertner, Rice, & Hadley 1994), and are three times more likely to be victimized by their TD peers (Conti-Ramsden & Botting, 2004). A prac- tical result is that individuals with SLI experience higher rates of anxiety and depression, have markedly lowered self-esteem (Jerome, Fujiki, Brinton, & James, 2002), and are at increased risk for social maladjustment into adolescence (Conti- Ramsden & Botting, 2008). Developmental Course When it comes to productive language, between 25 and 50% of toddlers who are “late bloomers” will go on to have long-term language impairments (see Chap. 6 for a discussion of the predictors and outcomes of late talkers). These children constitute the core of those with SLI (Owens, 2010). SLI in early and middle childhood is characterized by limited vocabulary and impaired narrative performance, but the most obvious signs are the failure to produce simple morphological structures (e.g., omission of past tense -ed or plural -s) and sentence structures (e.g., three element noun phrases like “the big girl”). As children with SLI grow into late childhood, their problems with simple morphology become less severe, while problems with complex sentences, narratives, and !gurative language become more obvious (Leonard, 2014). It is at this point that SLI begins to seriously affect school achievement, as de!cits in language affect the ability to master read- ing, writing, and discourse (Bernstein-Ratner, 2013). Thus, persons with SLI are at risk for reading problems, academic failure, and even dyslexia, which is a distinct condition that can co-occur with SLI. Notably, most children with SLI show gains in language ability over time, but the language de!cit proves to be longstanding, and many adults with a history of SLI continue to be less able linguistically compared to their peers (Tomblin, Freese, & Records, 1992). Thus, the language signs of SLI tend to change over the life span, with more obvious problems evident in early development (especially in vocabulary and simple morphology) and more subtle – but consequential – impairments in later years (especially in complex syntax and higher-order language skills related to aca- demic achievement). Recall that social and emotional well-being is also a concern in SLI. Yet it is encouraging that depression and withdrawal, though detectable, often occur at subclinical levels and symptoms tend to attenuate after adolescence (Leonard, 2014; Records, Tomblin, & Freese, 1992). Although some research has found that adoles- cents and young adults with SLI may be less independent in daily living (Conti- Ramsden & Durkin, 2008) and may feel less in control of their daily lives (Records 58 6 Children with Specific Language Impairment et al., 1992), they also tend to hold generally positive attitudes and report levels of personal happiness and life satisfaction no different from their TD peers (Records et al., 1992). Cultural and Linguistic Considerations A dialect is a regional or social variety of a language that is distinguished by pronunciation, grammar, or vocabulary. Dialectical and language variation are important to the discussion of SLI for at least two reasons. First, when identifying SLI, professionals must consider whether the child is presenting with a language disorder or a language difference, a distinction that is not always clear. For exam- ple, in African American English (AAE), instead of “She is baking,” it is accept- able to drop the auxiliary verb and say “She baking.” “She baking” is an error that would be consistent with SLI in a child speaker of standard English, but a child speaking AAE would be producing language in a way that is consistent with her home dialect, and she should not be wrongly identi!ed as having SLI. The second reason to consider dialect and language is the striking cross-linguistic differences among children with SLI. As Leonard (2009) explained: A common dif!culty seen in English-speaking children with SLI is an extraordinary dif!- culty with tense and agreement morphemes, with omission by far the most common error… In contrast, in languages such as Spanish and Italian, tense and agreement are relatively accurate in the speech of children with SLI, although function words such as articles…can be problematic. (p. 169) In summary, the morphosyntactic signs of SLI look different across languages. This fact has the potential to complicate assessment (discussed more fully below). One recommendation for meeting this challenge is that professionals consult sources devoted to the description of SLI in the language of interest if they exist (see Leonard (2014) for a review). Another is that professionals consider language pro!les at the most general level. “If there is a universal feature of SLI apart from generally show and poor language learning, it is well hidden” (Leonard, 2014, p. 150). Regardless of language then, SLI should be associated with abundant evidence for late emer- gence of language and protracted development that is suggestive of a delay. Moreover, the errors that are made will generally resemble those seen in younger TD children. Prevalence and Common Concomitant Disorders Determining the prevalence rate of SLI has been especially dif!cult due to great variability in how SLI has been de!ned. At present, the most trustworthy data indi- cate that the prevalence rate of SLI is 7.4% (8% for boys and 6% for girls; Tomblin, Smith, & Zhang, 1997) making SLI one of the most common disorders of Causes 59 childhood. As noted previously, SLI is strongly associated with reading problems (about 80% of children with SLI experience reading problems; Botting, Simkin, & Conti-Ramsden, 2006) and may co-occur with dyslexia. It also frequently co- occurs with attention de!cit hyperactivity disorder (58%; Beitchman, Hood, Rochon, and Peterson 1989) and autism spectrum disorder (3.9% which is about 10 times what would be expected from the general population; Conti-Ramsden, Simkin, & Botting, 2006). Causes Some researchers have proposed that SLI represents the low end of the normal distribution of variation in language ability. However, recent research has led to a broader consensus that SLI reflects underlying brain dysfunction at some level, even though it is not grossly manifest (Bernstein-Ratner, 2013). In most cases, there is evidence of a familial, hereditary component to SLI. Of course, this also means that in a minority of cases, the child with SLI comes from families with no evidence of a language disorder. As such, the basis of SLI is complex, likely involves multiple genes and multiple biological factors, and may interact with environmental factors (Bernstein-Ratner, 2013; Leonard, 2009, 2014). Although the quality of the lan- guage environment plays a vital role in the treatment of SLI (discussed in more detail below), and environmental contributions to SLI are certainly possible, inad- equate language input usually is not considered a causal factor in SLI (see Leonard for a review of the evidence, 2014). A number of models of SLI have been proposed as explanatory mechanisms (for a comprehensive description of all of the theories and evidence for various models, see Leonard, 2014). Some of the more influential models include the surface hypothesis which proposes that children with SLI have dif!culty processing those pieces of language that lack prominence due to their sound structure (e.g., unstressed sounds that are brief in duration like plural -s and past tense -ed). Although this hypothesis predicts many of the error patterns seen in the productive language of children with SLI, it cannot explain all patterns, and acoustically enhancing speech to increase the saliency of these forms has not been found to improve performance. One well-known proposition is the auditory processing hypothesis. This hypothesis proposes that children with SLI suffer from de!cits in the temporal processing of auditory stimuli (AKA auditory processing disorder where the indi- vidual has dif!culty identifying, segmenting, sequencing, and integrating auditory stimuli). This hypothesis has led to the development of interventions (commer- cially available as “Fast ForWord”) that have yielded impressive but highly con- troversial results (Bernstein-Ratner, 2013). In short, a series of studies by Tallal and colleagues (see Agocs, Burns, De Ley, Miller, & Calhoun, 2006; Tallal, 2003, for reviews by authors who have contributed to the Fast ForWord approach) reported large gains on language measures, but other researchers have been unable 60 6 Children with Specific Language Impairment to replicate these !ndings. Moreover, the literature on auditory processing in SLI is mixed with some studies indicating weakness in this area (Corriveau, Pasquini, & Goswani, 2007), while others show little evidence of atypical auditory process- ing (Bishop, Adams, Nation, & Rosen, 2005). Another model of SLI is known as the generalized slowing hypothesis. This model proposes that children with SLI have generally slowed information process- ing which leads to problems that include, but go beyond, language. This hypothe- sis suggests that children with SLI need about one third more time to perform a range of perceptual and motor functions (Bernstein-Ratner, 2013). “Such slowing might contribute to, or interact with, other proposals that suggest that SLI is the outcomes of the limited processing capacity in some children. The underlying limit may be one of slowed capacity…or of limits of processing ‘space’ or vulner- ability to competing demands on the system” (Bernstein-Ratner, 2013, p. 299). Additional processing systems that have been implicated include, but are not lim- ited to, de!cits in working memory (particularly phonological working memory) and executive function. Furthermore, Ullman and Pierpoint (2005) have argued that SLI is the result of an impaired procedural processing system, which underlies the learning and perfor- mance of skills involving sequences. Much enthusiasm for the procedural deficit hypothesis has accrued in recent years as it seems well-suited to explain the hetero- geneity seen across persons with SLI. It predicts broad impairment in functions that rely on processing of sequential information (e.g., grammar, rapid naming, working memory, auditory processing) and appears to be consistent with the neurological evidence for SLI (Leonard, 2014). Of course, the actual mechanisms operating in SLI are still a matter of debate. What is clear is that if we can identify the causal factors of SLI, we can develop more appropriate methods for intervention with a focus on early identi!cation. Assessment SLI needs to be distinguished from normal variations in language ability. For this reason, the usual recommendation is to avoid a diagnosis of SLI before age 4 years (American Psychiatric Association, 2013). Around this period of development, fam- ily members and others close to the child may begin to recognize a language prob- lem although they may lack con!dence in their ability to describe it adequately. The child may appear shy, reticent to talk, and prefer to communicate only with family members (American Psychiatric Association, 2013). A positive family history of language disorder (the best-known predictor of SLI) is common, and observation should reveal general language delay with particular weakness in morphosyntax. In fact, the vast majority of English-speaking children with SLI are not identi!ed by their limited vocabularies, poor narrative performance, or possible accompanying speech disorder. Rather, they are identi!ed by their failure to achieve normal syntac- tic production whether or not they have accompanying de!cits in comprehension (Bernstein-Ratner, 2013). Treatment 61 Until recently, low scores on standardized language tests played the primary role in establishing a diagnosis of SLI (see Chap. 3 for a description of several tests that may be appropriate for this purpose). Generally 1.25 standard deviations below the mean (about 80) on a language test would suggest SLI. Some professionals have advocated for a criterion that requires a low score on a test of language com- prehension as well as a low score of a test of language production, but this criterion has not been applied consistently (Leonard, 2014). For exclusionary criteria (e.g., nonverbal intelligence), performance in the average range is typically de!ned as falling above 85. Of course, the “test score-only” approach should not be considered a comprehen- sive evaluation regardless of the number of tests. When possible, assessment should also include observation (see Kadervek , for a description of parent-child toy play and book reading observation assessments and assessment tools) and language sample analysis. An added drawback of the “test score-only” approach is that it neglects clinical signi!cance. Indeed, diagnosing SLI using a cut-score of 85 on a test of nonverbal intelligence has not proven helpful because children with scores just above and below the cut-score have been shown to have similar language pro- !les (Tommerdahl & Drew, 2008). Recall that diagnostic criteria (American Psychiatric Association, 2013) require evidence that the observed language limitation constitutes a quanti!able and func- tional impairment. As such, some have described the value of test scores combined with clinical judgment and have argued that these factors in combination be consid- ered the gold standard of assessment (Leonard, 2014; Tomblin, Records, & Zhang 1996). More recently, placement into language intervention as an indicator of parent or teacher concern about the child’s language development is seen as a gold stan- dard (Leonard, 2014). Another has been curriculum-based assessment where a stu- dent’s academic performance is evaluated in such a way that reveals functioning across language domains and results in meaningful intervention goals. Treatment Productive language de!cits are a de!ning feature of SLI, and, as noted above, whether comprehension de!cits should be an inclusion criterion remains a matter of debate. With regard to prognosis, children with comprehension de!cits have a poorer prognosis and tend to be more resistant to treatment (American Psychiatric Association, 2013). When gains are achieved, the current evidence suggests modest success when morphosyntax is the target of intervention and success is somewhat greater when treatment targets vocabulary (Leonard, 2009). For treatment of SLI, general language stimulation is not as effective as treatment that focuses on the speci!c linguistic skills that the child needs to master (Leonard, 2014). Some com- mon intervention techniques that are designed to address the morphosyntactic de!- cits of SLI are presented in Table 6.3. Of course, in practice, these techniques are often used in ways that can also promote vocabulary knowledge. 62 6 Children with Specific Language Impairment Table 6.3 Some approaches to treating the morphosyntactic de!cits of SLI Approach Description Example Conversational Adult responds to child’s spontaneous Child: “This green clay” recasting language by rephrasing it to include Clinician: “This is green clay” the target form Expansion The adult responds to the child’s Child: “This clay no good” spontaneous language by including Clinician: “This clay isn’t any good. It additional information isn’t. It is too dry” Imitation Child is asked to repeat model Clinician: “I am rolling the clay. You presented by therapist are too. Say I am rolling” Child: “I am rolling” Focused Child exposed to large number of Clinician: “Here is green clay. Let’s stimulation exemplars of the target form or work; make vegetables. Lettuce is green. child may then be asked questions Cabbage is green. A cucumber is requiring use of that form green. Here you make a tomato” (hands child green clay) Child: “No, Tomato is red” Modeling Adult models a target form; child can Clinician: “I am rolling the clay. I am be asked to produce form through use rolling. What are you doing?” of a question prompt Child: “I am rolling” Scaffolding Adult provides structure for the Clinician: “This snake is very big. This child’s attempts. Gradually, this one is very small. And this one is… structure is faded to allow the child to Child: “Skinny!” produce the target on his own Clinician: “Right. This one is skinny. Any this one… Child: “is fat” Sentence Adult gives child two or more simple Clinician: “The boy is running. The combining sentences and asks child to combine boy wears a red hat. The boy is going them into one longer sentence to the store” Child: “The boy in the red hat is running to the store” Sentence Adult gives core sentence and child Clinician: “The dog is sleeping. Now expansion asked to elaborate on it and make it expand the sentence and tell me why” longer Child: “The dog is sleeping because he stayed up all night” Time delay or Slowing the pace of conversation and Adult: “All these cans of clay are slowed waiting for the child to supply a mine!” (pulls the materials toward her presentation required form and waits) Child: “No, these are mine!” Adapted from Bernstein-Ratner (2013) In each case, the intervention target is italicized Most of these strategies were developed from what we know about the kind of caregiver-child interactions that facilitate language growth in TD children, and all of them have received some support in the literature for use with children with SLI. Although conversational recasting has been particularly lauded, it is impor- tant to remember that SLI is not a monolithic condition, no SLI intervention is universally effective, and treatment decisions must consider priorities for interven- tion as well as children’s initial ability level and learning style. One common theme from the research, however, is that due to their status as slow language learners, children with SLI require more contact with language targets than their TD peers.