Working memory and comprehension in children with specific language impairment: what we know so far

Many children with specific language impairments (SLI) demonstrate deficits in the areas of verbal working memory and language learning/processing. In this article, evidence is reviewed suggesting that the lexical/morphological learning and sentence comprehension/processing problems of many of these children are associated with their deficient working memory functioning. Evidence is also reviewed for the possibility that deficient working memory provides a clinical marker of SLI. A number of potentially useful assessment and intervention techniques are offered, as well as several directions for future research.The reader will be introduced to two prominent models of verbal working memory (phonological working memory model, functional working memory) and how each model potentially relates to (a) various language abilities in typically developing children, (b) the morphological and lexical learning abilities in children with specific language impairment (SLI), and (c) the sentence comprehension of children with SLI. The reader will also be provided a variety of clinical suggestions on how to assess and treat the working memory and language processing problems of children with SLI. Finally, some suggestions for future research will also be offered.     

Beyond capacity limitations: determinants of word recall performance on verbal working memory span tasks in children with SLI.

Reduced verbal working memory capacity has been proposed as a possible account of language impairments in specific language impairment (SLI). Studies have shown, however, that differences in strength of linguistic representations in the form of word frequency affect list recall and performance on verbal working memory tasks. This suggests that verbal memory capacity and long-term linguistic knowledge may not be distinct constructs. It has been suggested that linguistic representations in SLI are weak in ways that result in a breakdown in language processing on tasks that require manipulation of unfamiliar material. In this study, the effects of word frequency, long-term linguistic knowledge, and serial order position on recall performance in the competing language processing task (CLPT) were investigated in 10 children with SLI and 10 age-matched peers (age 8 years 6 months to 12 years 4 months). The children with SLI recalled significantly fewer target words on the CLPT as compared with their age-matched controls. The SLI group did not differ, however, in their ability to recall target words having high word frequency but were significantly poorer in their ability to recall words on the CLPT having low word frequency. Differences in receptive and expressive language abilities also appeared closely related to performance on the CLPT, suggesting that working memory capacity is not distinct from language knowledge and that degraded linguistic representations may have an effect on performance on verbal working memory span tasks in children with SLI.     

Grammatical tense deficits in children with SLI and nonspecific language impairment: relationships with nonverbal IQ over time.

The relationship between children's language acquisition and their nonverbal intelligence has a long tradition of scientific inquiry. Current attention focuses on the use of nonverbal IQ level as an exclusionary criterion in the definition of specific language impairment (SLI). Grammatical tense deficits are known as a clinical marker of SLI, but the relationship with nonverbal intelligence below the normal range has not previously been systematically studied. This study documents the levels of grammatical tense acquisition (for third-person singular -s, regular and irregular past tense morphology) in a large, epidemiologically ascertained sample of kindergarten children that comprises 4 groups: 130 children with SLI, 100 children with nonspecific language impairments (NLI), 73 children with low cognitive levels but language within normal limits (LC), and 117 unaffected control children. The study also documents the longitudinal course of acquisition for the SLI and NLI children between the ages of 6 and 10 years. The LC group did not differ from the unaffected controls at kindergarten, showing a dissociation of nonverbal intelligence and grammatical tense marking, so that low levels of nonverbal intelligence did not necessarily yield low levels of grammatical tense. The NLI group's level of performance was lower than that of the SLI group and showed a greater delay in resolution of the overgeneralization phase of irregular past tense mastery, indicating qualitative differences in growth. Implications for clinical groupings for research and clinical purposes are discussed.     

Visuospatial immediate memory in specific language impairment.

PURPOSE: Investigations of the cognitive processes underlying specific language impairment (SLI) have implicated deficits in verbal short-term and working memory and in particular the storage and processing of phonological information. This study investigated short-term and working memory for visuospatial material for a group of children with SLI, to test whether the verbal memory impairments already established extend to the visuospatial domain. METHOD: Fifteen children with SLI and control groups of children matched on chronological age and language age completed tests of visuospatial short-term and working memory. RESULTS: The SLI group performed comparably with age-matched control children on all measures and at a higher level than the language-age control group on several measures. CONCLUSIONS: The visuospatial short-term and working memory abilities were at age-appropriate levels in this SLI group. This contrasts markedly with their impairments on tests of verbal short-term and working memory.     

The comprehension of wh-questions in children with specific language impairment.

Current theories of specific language impairment (SLI) in children fall into 2 general classes: those that attribute SLI to processing limitations and those that attribute the disorder to deficits in grammatical knowledge. In this study, the authors examined children's comprehension of subject and object Wh-questions because they offer the means of determining the relative contribution to performance of knowledge and of processing. Comprehension of subject and object Wh-questions presumably requires knowledge of grammatical movement. However, through manipulation of the length of the questions, it is also possible to vary the processing demands of the questions. If a simple deficit in knowledge of movement is involved, children should show poorer comprehension on object questions than on subject questions, regardless of sentence length. However, if processing limitations are involved, length should affect comprehension of object questions but not subject questions. Children with SLI and typically developing (TD) children matched on receptive vocabulary test scores participated in a comprehension task consisting of short and long subject and object Wh-questions. The two groups performed similarly on short questions, each showing high accuracy in both subject and object conditions. However, the children with SLI showed poorer performance on long object questions compared to long subject questions. They were also less accurate on long object questions than were children in the TD group. We argue that demands on linguistic processing abilities play an important role in the difficulties experienced by children with SLI.     

Specific language impairment in Swedish: the status of verb morphology and word order.

Several competing proposals have been offered to explain the grammatical difficulties experienced by children with specific language impairment (SLI). In this study, the grammatical abilities of Swedish-speaking children with SLI were examined for the purpose of evaluating these proposals and offering new findings that might be used in the development of alternative accounts. A group of preschoolers with SLI showed lower percentages of use of present tense copula forms and regular past tense inflections than normally developing peers matched for age and younger normally developing children matched for mean length of utterance (MLU). Word order errors, too, were more frequent in the speech of the children with SLI. However, these children performed as well as MLU-matched children in the use of present tense inflections and irregular past forms. In addition, the majority of their sentences containing word order errors showed appropriate use of verb morphology. None of the competing accounts of SLI could accommodate all of the findings. In particular, these accounts--or new alternatives-must develop provisions to explain both the earlier acquisition of present tense inflections than past tense inflections and word order errors that seem unrelated to verb morphology.     

Morphophonological salience as a compensatory means for deficits in the acquisition of past tense in SLI

The aim of this study is threefold: Firstly, to describe the acquisition patterns of Greek past tense by children with specific language impairment (SLI); secondly, to investigate the relationship between the phonological salience of past tense in Greek and its acquisition by children of typical and atypical language development; thirdly, to establish an account on the nature of specific language impairment by comparing the acquisition patterns exhibited by children with SLI to those presented by typically developing children.The performance of 10 children with SLI in elicited past tense production is compared to that of chronological age matched (CA) and language matched (LM) controls. Based on the claim that morphophonological salience aids acquisition (or learning), it is predicted that children with SLI will perform better in the production of past forms of higher salience than less salient forms. The results confirm this prediction, providing support for the claim that salience differences do account for better performances in tense marking in Greek SLI.Learning outcomes: By reading this article, the reader will be able to: (1) describe the problems children with SLI face in Greek tense marking, (2) explain the reasons why SLI performances differ from one language to the next based on the Interpretability Hypothesis, (3) discuss whether children with SLI differ qualitatively from TD controls in past tense marking and explain how.     

The effect of temporal adverbials on past tense production by children with specific language impairment.

PURPOSE: Children with specific language impairment (SLI) often fail to produce past tense forms in obligatory contexts, although the factors affecting such inconsistency are not well understood. This study examined the influence of accompanying temporal adverbials (e.g., just, already) on the past tense production of these children. METHOD: Fifteen preschool-aged children with SLI, 15 typically developing children matched for age (TD-A) and 15 younger typically developing children matched for mean length of utterance (TD-MLU) participated in the study. The children responded to probes that obligated the use of past tense forms. The verbal context provided by the experimenter for half of the items included a temporal adverbial. RESULTS: Overall, the SLI and TD-MLU groups produced past tense less frequently than the children in the TD-A group, and there were no significant differences between the SLI and the TD-MLU groups. However, both the SLI and the TD-MLU participants produced past tense forms less frequently when temporal adverbials were included than when they were absent. CONCLUSIONS: These findings suggest that the tendency to use past tense by the children with SLI and their younger MLU-matched peers may not have been independent of other types of temporal information.     

Language-specific effects of task demands on the manifestation of specific language impairment: a comparison of English and Icelandic

PURPOSE: Previous research has indicated that the manifestation of specific language impairment (SLI) varies according to factors such as language, age, and task. This study examined the effect of task demands on language production in children with SLI cross-linguistically. METHOD: Icelandic- and English-speaking school-age children with SLI and normal language (NL) peers (n = 42) were administered measures of verbal working memory. Spontaneous language samples were collected in contexts that vary in task demands: conversation, narration, and expository discourse. The effect of the context-related task demands on the accuracy of grammatical inflections was examined. RESULTS: Children with SLI in both language groups scored significantly lower than their NL peers in verbal working memory. Nonword repetition scores correlated with morphological accuracy. In both languages, mean length of utterance (MLU) varied systematically across sampling contexts. Context exerted a significant effect on the accuracy of grammatical inflection in English only. Error rates were higher overall in English than in Icelandic, but whether the difference was significant depended on the sampling context. Errors in Icelandic involved verb and noun phrase inflection to a similar extent. CONCLUSIONS: The production of grammatical morphology appears to be more taxing for children with SLI who speak English than for those who speak Icelandic. Thus, whereas children with SLI in both language groups evidence deficits in language processing, cross-linguistic differences are seen in which linguistic structures are vulnerable when processing load is increased. Future research should carefully consider the effect of context on children's language performance.     

Verbal working memory and sentence comprehension in children with specific language impairment.

In this study we examined the influence of verbal working memory on sentence comprehension in children with SLI. Twelve children with SLI, 12 normally developing children matched for age (CA), and 12 children matched for receptive vocabulary (VM) completed two tasks. In the verbal working memory task, children recalled as many real words as possible under three processing load conditions (i.e., no-load condition; single-load condition, where words were recalled according to physical size of word referents; and dual-load condition, where words were recalled by semantic category and physical size of word referents). In the sentence comprehension task, children listened to linguistically nonredundant (shorter) and linguistically redundant (longer) sentences. Results of the memory task showed that the children with SLI recalled fewer words in the dual-load condition than their CA peers, who showed no condition effect. The SLI and VM groups performed similarly overall, but both groups showed poorer recall in the dual-load condition than in the other conditions. On the sentence comprehension task, children with SLI comprehended fewer sentences of both types than the CA children and fewer redundant sentences relative to themselves and to the VM children. Results were interpreted to suggest that children with SLI (a) have less functional verbal working memory capacity (i.e., ability to coordinate both storage and processing functions) than their CA peers and (b) have greater difficulty managing both their working memory abilities and general processing resources than both age peers and younger children when performing a "complex" off-line sentence processing task.     

Nonword repetition: a comparison of tests.

PURPOSE: This study compared performance of children on 2 tests of nonword repetition to investigate the factors that may contribute to the well-documented nonword repetition deficit in specific language impairment (SLI). METHOD: Twelve children with SLI age 7 to 11 years, 12 age-matched control children, and 12 control children matched for language ability completed 2 tests of nonword repetition: the Children's Test of Nonword Repetition (CNRep) and the Nonword Repetition Test (NRT). RESULTS: The children with SLI performed significantly more poorly on both tests than typically developing children of the same age. The SLI group was impaired on the CNRep but not the NRT relative to younger children with similar language abilities when adjustments were made for differences in general cognitive ability. The children with SLI repeated the lengthiest nonwords and the nonwords containing consonant clusters significantly less accurately than the control groups. CONCLUSION: The evidence suggests that the nonword repetition deficit in SLI may arise from a number of factors, including verbal short-term memory, lexical knowledge, and output processes.     

The onset of tense marking in children at risk for specific language impairment.

PURPOSE: The purpose of this study was to develop and evaluate measures reflecting the onset of tense marking for children between the ages of 2;0 (years;months) and 3;0. METHOD: The validity of 4 cumulative measures of tense marker emergence and productivity was evaluated relative to existing measures of early grammatical development in a sample of 20 children followed longitudinally. Fourteen children were at risk for specific language impairment (AR-SLI group), and 6 children had low average language abilities (LA group). RESULTS: All measures of onset were highly correlated with the traditional measures; however, children's progress toward mastery of grammatical tense marking was best explained by the productivity of their tense marking systems. Finally, the onset measures imposing productivity requirements best differentiated children in the LA group from those in the AR-SLI group. CONCLUSIONS: The clinical implications for using the late onset of tense marking to improve the early identification of SLI are discussed.     

French-English bilingual children with SLI: how do they compare with their monolingual peers?

The goal of this study was to determine whether bilingual children with specific language impairment (SLI) are similar to monolingual age mates with SLI, in each language. Eight French-English bilingual children with SLI were compared to age-matched monolingual children with SLI, both English and French speaking, with respect to their use of morphosyntax in language production. Specifically, using the extended optional infinitive (EOI) framework, the authors examined the children's use of tense-bearing and non-tense-bearing morphemes in obligatory context in spontaneous speech. Analyses revealed that the patterns predicted by the EOI framework were borne out for both the monolingual and bilingual children with SLI: The bilingual and monolingual children with SLI showed greater accuracy with non-tense than with tense morphemes. Furthermore, the bilingual and monolingual children with SLI had similar mean accuracy scores for tense morphemes, indicating that the bilingual children did not exhibit more profound deficits in the use of these grammatical morphemes than their monolingual peers. In sum, the bilingual children with SLI in this study appeared similar to their monolingual peers for the aspects of grammatical morphology examined in each language. These bilingual-monolingual similarities point to the possibility that SLI may not be an impediment to learning two languages, at least in the domain of grammatical morphology.     

Longitudinal relationships between lexical and grammatical development in typical and late-talking children.

PURPOSE: This study examined the longitudinal relationships between lexical and grammatical development in typically developing (TD) and late-talking children for the purposes of testing the single-mechanism account of language acquisition and comparing the developmental trajectories of lexical and grammatical development in late-talking and TD children. METHOD: Participants included 30 children identified as late talkers (LTs) at 2;0 (years;months), and 30 TD children matched on age, nonverbal cognition, socioeconomic status, and gender. Data were collected at 5 points between 2;0 and 5;6. RESULTS: Cross-lagged correlational analyses indicated that TD children showed evidence of bidirectional bootstrapping between lexical and grammatical development between 2;0 and 3;6. Compared with the TD group, LTs exhibited less evidence of syntactic bootstrapping. Linear mixed-effects modeling of language sample data suggested that the relationship between lexical and grammatical growth was similar for the 2 groups. CONCLUSION: Lexical and grammatical development were strongly related in both groups, consistent with the single-mechanism account of language acquisition. The results were mixed in terms of finding longitudinal differences in lexical-grammatical relationships between the TD and late-talking children; however, several analyses suggested that for late-talking children, syntactic growth may be less facilitative of lexical development.     

Profiles of grammatical morphology and sentence imitation in children with specific language impairment and Down syndrome.

The purpose of the present study was to examine the grammatical morphology and sentence imitation performance of two different groups of children with language impairment and to compare their performance with that of children learning language typically. Expressive use of tense-bearing and non-tense-related grammatical morphemes was explored. Children with specific language impairment (SLI), with Down syndrome (DS), and with typical language development (TL) were matched on mean length of utterance (MW). Performance was compared primarily on composite measures of tense, tense inflections, and non-tense morphemes, as well as on the Sentences subtest of the Wechsler Preschool and Primary Scale of Intelligence-Revised (WPPSI-R; D. Wechsler, 1989). Exploratory analyses were completed on a set of 11 individual grammatical morphemes as a follow-up to the principal analyses. As predicted, the children with SLI performed significantly more poorly than the children with TL on all three composite measures. In addition, the DS group exhibited significantly weaker performance than did the TL group on the tense inflections and non-tense morpheme composites. Although there were no statistically reliable differences between the SLI and DS groups on any morpheme measure, the groups were not comparably weak in their use of the regular post, -ed; the irregular third person singular morphemes (e.g., has, does); the present progressive, -ing; or the use of modals. The SLI and DS groups both performed more poorly than did the TL group on the sentence imitation task.     

A functional magnetic resonance imaging investigation of verbal working memory in adolescents with specific language impairment.

This study used neuroimaging and behavioral techniques to examine the claim that processing capacity limitations underlie specific language impairment (SLI). Functional magnetic resonance imaging (fMRI) was used to investigate verbal working memory in adolescents with SLI and normal language (NL) controls. The experimental task involved a modified listening span measure that included sentence encoding and recognition of final words in prior sets of sentences. The SLI group performed significantly poorer than the NL group for both encoding and recognition and displayed slower reaction times for correct responses on high complexity encoding items. fMRI results revealed that the SLI group exhibited significant hypoactivation during encoding in regions that have been implicated in attentional and memory processes, as well as hypoactivation during recognition in regions associated with language processing. Correlational analyses indicated that adolescents with SLI exhibited different patterns of coordinating activation among brain regions relative to controls for both encoding and recognition, suggesting reliance on a less functional network. These findings are interpreted as supporting the notion that constraints in nonlinguistic systems play a role in SLI.     

An examination of verbal working memory capacity in children with specific language impairment.

This study investigated verbal working memory capacity in children with specific language impairment (SLI). The task employed in this study was the Competing Language Processing Task (CLPT) developed by Gaulin and Campbell (1994). A total of 40 school-age children participated in this investigation, including 20 with SLI and 20 normal language (NL) age-matched controls. Results indicated that the SLI and NL groups performed similarly in terms of true/false comprehension items, but that the children with SLI evidenced significantly poorer word recall than the NL controls, even when differences in nonverbal cognitive scores were statistically controlled. Distinct patterns of word-recall errors were observed for the SLI and NL groups, as well as different patterns of associations between CLPT word recall and performance on nonverbal cognitive and language measures. The findings are interpreted within the framework of a limited-capacity model of language processing.     

Lexical aspect and the use of verb morphology by children with specific language impairment.

PURPOSE: Many typically developing children first use inflections such as -ed with verb predicates whose meanings are compatible with the functions of the inflection (e.g., using -ed when describing events of brief duration with clear end points, such as dropped). This tendency is assumed to be beneficial for development. In this study, the authors examine whether preschool-aged children with specific language impairment (SLI) show a similar tendency. METHOD: Sixteen children in each of three groups participated-children with SLI, typically developing children matched for age (TD-A), and younger typically developing children matched for mean length of utterance (TD-MLU). The children described actions in contexts that promoted either past tense -ed or progressive aspect -ing in past contexts. Half of the verb predicates referred to events of brief duration with distinct endpoints (e.g., drop), and half referred to events of considerable duration with less distinct points of termination (e.g., play). RESULTS: Both the TD-A children and the TD-MLU children used -ed with verb predicates of the first type more consistently than they did with verb predicates of the second type. They showed the reverse pattern for -ing. The children with SLI did not show any effects according to the verb predicate type. However, although the children with SLI made less overall use of -ed than did both groups of TD children, they differed only from the TD-A children in their overall use of -ing. CONCLUSION: Difficulties with tense-related morphology may be compounded in children with SLI if they fail to make use of associations between the lexical aspect of verb predicates and the grammatical function of the accompanying inflections. The authors argue that the advantages of using these associations as a starting point in acquisition may be especially important in the case of -ed. Additional studies of children with SLI are clearly needed, including those that employ longitudinal, naturalistic data.     

Simulating SLI: general cognitive processing stressors can produce a specific linguistic profile.

This study attempted to model specific language impairment (SLI) in a group of 6-year-old children with typically developing language by introducing cognitive stress factors into a grammaticality judgment task. At normal speech rate, all children had near-perfect performance. When the speech signal was compressed to 50% of its original rate, to simulate reduced speed of processing, children displayed the same pattern of errors that is reported in SLI: good performance on noun morphology (plural -s) and very poor performance on verb morphology (past tense -ed and 3rd-person singular -s). A similar pattern was found when memory load was increased by adding redundant verbiage to sentence stimuli. The finding that an SLI-like pattern of performance can be induced in children with intact linguistic systems by increasing cognitive processing demands supports the idea that a processing deficit may underlie the profile of language difficulty that characterizes SLI.     

Verbal and spatial information processing constraints in children with specific language impairment.

A dual-processing paradigm was used to investigate information processing limitations underlying specific language impairment (SLI). School-age children with and without SLI were asked to recall verbal and spatial stimuli in situations that varied the number of tasks that were required and the speed at which stimuli were presented. Children recalled digits or locations of X's that were presented on a computer screen. In some conditions, they were asked to name or point to the color of the stimuli before completing the recall task. In comparison to their typically developing peers, children with SLI had generally poorer recall of digits and locations across all conditions. Typically developing children derived greater benefit than the children with SLI under conditions that enabled them to disperse processing efforts across verbal and spatial response modalities. It appears that limitations in general cognitive capacity and central executive functions in working memory work synergistically with response modality to constrain information processing in children with SLI.