Skip to main content
NCBI home page
Journal of Speech, Language, and Hearing Research : JSLHR logoLink to Journal of Speech, Language, and Hearing Research : JSLHR
. 2019 Sep 12;62(9):3462–3469. doi: 10.1044/2019_JSLHR-L-18-0331

Not All Nonverbal Tasks Are Equally Nonverbal: Comparing Two Tasks in Bilingual Kindergartners With and Without Developmental Language Disorder

Kathleen Durant a,✉, Elizabeth Peña b, Anna Peña c, Lisa M Bedore d, María R Muñoz e
PMCID: PMC6808348  PMID: 31518170

Abstract

Purpose

This study investigates the interaction of language ability status, cultural experience, and nonverbal cognitive skill performance in Spanish–English bilinguals with typical development (TD) and developmental language disorder (DLD).

Method

One hundred sixty-nine Spanish–English bilingual kindergartener's scores on the Symbolic Memory and Cube Design subtests from the Universal Nonverbal Intelligence Test (Bracken & McCallum, 1998) were analyzed by language ability (TD vs. DLD).

Results

t tests and analysis of variance showed bilingual children with TD and DLD performed comparably to the Universal Nonverbal Intelligence Test norming sample on the cube design task, while children with DLD had significantly lower performance on the symbolic memory task.

Conclusion

These results suggest that cultural experience minimally impacted performance for bilingual children with typically developing language. Bilingual children with DLD were differentially impacted on symbolic memory, a task that is verbally mediated despite nonverbal administration and performance. Findings are discussed within the Cattell–Horn–Carroll theory of cognitive abilities.

Developmental language disorder (DLD) is defined as a significant delay or difficulty in the area of oral language not resulting from differences in learning opportunities or environment and unassociated with neurological, perceptual, developmental, or cognitive delays (Leonard, 2014; Tallal & Stark, 1981; Tomblin, Records, & Zhang, 1996). As a result, assessment of DLD often involves quantification of language performance and cognitive skills. In language development research, nonverbal IQ testing is often used as inclusionary or exclusionary criteria to identify populations of interest. Clinically, when children are referred for special education, nonverbal IQ testing is part of a comprehensive battery that determines strengths and needs across multiple cognitive domains.

Verbal testing of cognitive skills inappropriately penalizes children with DLD for language difficulties (Botting, 2005). For this reason, nonverbal cognitive testing is intimately related to a specific and reliable diagnosis of children with DLD. For sequential Spanish–English bilingual children suspected of having DLD, nonverbal cognitive assessment is recommended so as not to penalize children for incomplete second language (L2) acquisition while assessing underlying cognitive abilities (Saar, Levänen, & Komulainen, 2018). Nonverbal IQ testing is thought to evaluate children's capacity to perceive, attend to, maintain, and manipulate information and, by extension, their learning potential across domains of cognition (Flanagan & McGrew, 1998). Administration and interpretation of nonverbal IQ tasks reflect mainstream cultural norms that may negatively impact the performance of Spanish-speaking children who live in Latino households compared to their monolingual peers (Ortiz, 2001).

Cognition and language share a complex bidirectional relationship embedded in cultural norms. How different facets of cognitive performance interact with language performance in typical and disordered language development in bilingual children is an open question (Earle, Gallinat, Grela, Lehto, & Spaulding, 2017; Saar et al., 2018). The current study investigates the relationships between language ability and performance on two nonverbal cognitive tasks—symbolic memory and cube design—for kindergarten-age Spanish-speaking English language learners with typical development (TD) or DLD from Latino households. In an effort to reduce verbal task demands, tasks are completed nonverbally through gestures. Yet, some tasks may provoke more inner speech than others, and there are cultural differences in how concepts are encoded in gestures. Thus, interpretation of nonverbal performance for children with DLD can be affected by cognition, verbal mediation, and culture.

Cognition and Language

Task performance in children with DLD varies considerably within comprehensive IQ batteries that target different aspects of cognition. Cognitive abilities can be expressed both verbally and nonverbally but are not distinctly verbal or nonverbal, reflected in findings of robust correlations between nonverbal IQ and language scores in kindergartners (Tomblin & Nippold, 2014). There is mounting evidence that processing and cognitive deficits are associated with DLD in monolingual English-speaking children, as reflected by consistently lower scores on both experimental and standardized nonverbal cognitive assessments (Earle et al., 2017; Gallinat & Spaulding, 2018; Newton, Roberts, & Donlan, 2010; Saar et al., 2018). Gallinat and Spaulding's (2018) meta-analysis of 131 studies found DLD children's scores were on average 0.69 of a standard deviation below their peers with TD, placing their performance within the typical range (i.e., above a standard score of 85 but below 100) yet significantly lower than their peers with TD.

Nonverbal IQ task development has been heavily influenced by the Cattell–Horn–Carroll (CHC) theory of cognitive abilities. Intelligence, within the CHC model, is thought to be a multidimensional construct with a general ability factor contributing to broad abilities (e.g., psychomotor abilities) as well as narrow abilities (e.g., visuospatial processing; Flanagan & McGrew, 1998). General, broad, and narrow abilities contribute both directly and indirectly to specific goal-orientated behavior. Additionally, the CHC framework can be used to classify the tasks used to assess intelligence. For example, psychomotor abilities (i.e., physical movements related to conscious cognitive processing) can directly support cube design performance and indirectly through visuospatial processing (i.e., making use of simulated mental images, in combination with perceived images, to accomplish a goal). Cognition is additionally hypothesized to be operationalized through fluid and crystallized intelligence.

Fluid intelligence encompasses the ability to use inductive reasoning to accomplish novel tasks. Crystallized intelligence refers to the knowledge and skills gained and retained through learning opportunities and experiences (Cattell, 1963). In an attempt not to penalize examinees for possible differences in learning opportunities or cultural practices, nonverbal IQ tests assess fluid intelligence to measure a person's capacity to learn, as opposed to measuring the crystallized knowledge and skills a person has already learned (DeThorne & Schaefer, 2004). While language knowledge (e.g., vocabulary) is theorized to represent crystallized intelligence, the capacity for learning language reflects fluid intelligence. L2 learning strategies, such as analogical reasoning, are related to fluid intelligence (Robinson, 2001). For bilingual children with DLD, differences in L2 learning outcomes (i.e., crystallized language intelligence) may reflect differences in mechanisms of learning (i.e., domain general fluid intelligence). Both crystallized and fluid intelligence reflect information processing subsumed by domain-general cognitive abilities, such as short-term memory. Fluid language learning tasks, such as nonword repetition, can reliably classify children with TD and DLD but have been theorized to reflect phonological working memory deficits that are constrained to the language system (Archibald & Gathercole, 2007; Dollaghan & Campbell, 1998). The evidence of domain-general fluid intelligence deficits, including short-term memory and processing deficits, is ambiguous in monolinguals with DLD and scarce in bilinguals with and without DLD (Vugs, Hendriks, Cuperus, & Verhoeven, 2014). Nevertheless, one study demonstrated that bilingual Spanish-Catalan–speaking preschoolers with DLD had lower performance on processing tasks than their peers with TD (Aguilar-Mediavilla, Buil-Legaz, Pérez-Castelló, Rigo-Carratalà, & Adrover-Roig, 2014).

Lower nonverbal cognitive test scores for children with DLD, compared to those with TD, may additionally reflect differences in problem-solving strategies, such as the use of language to manipulate information during goal-oriented problem solving. Inner speech plays a role in attending and keeping novel information activated in short-term memory through strategic verbal mediation (Sokolov, 2012). Verbal mediation has been found to facilitate problem solving on a range of nonverbal tasks across CHC domains, such as spatial localization, mental arithmetic, and card-sorting tasks, even when verbalization is not required (Perrone-Bertolotti, Rapin, Lachaux, Baciu, & Lœvenbruck, 2014). Indeed, the Universal Nonverbal Intelligence Test (UNIT; Bracken & McCallum, 1998) symbolic memory task was specifically designed to assess children's ability to use inner speech to label, organize, and categorize symbols in order to predict academic achievement. During task administration, gesture (e.g., pointing and hand waving) is used to direct children's attention to the stimulus card displaying a sequence of symbols depicting people of different ages and genders and response cards with the same symbols. Using inner speech to label the presented symbolic sequence (e.g., “baby, girl, boy, woman, man”) supports encoding and maintaining the sequence in short-term memory while a response is generated (DeThorne & Schaefer, 2004). Consequently, children with DLD may have lower performance on symbolic memory tasks due to difficulties with verbal mediation and insufficient short-term memory.

On the UNIT, cube design tasks draw on spatial visualization and orientation abilities in addition to processing speed. Linguistic encoding has not been found to support adult bilingual performance on a spatial orientation task, where participants were asked to decide if an inanimate object had changed orientation in space from a previously presented referent (Bosse & Papafragou, 2018). Participants can refer to the visual stimulus while completing the UNIT cube design task, however, reducing short-term working memory load compared to the spatial orientation confrontation task (Bracken & McCallum, 1998). Children with DLD are therefore hypothesized to score similarly to children with TD on spatial orientation tasks that do not require verbal mediation or tax short-term memory, such as cube design.

Nonverbal Assessment of Cognition

A recent meta-analysis by Gallinat and Spaulding (2018) identified the most frequently used IQ tests in DLD research, summarized in Table 1. Most IQ batteries used in research of DLD populations are comprehensive IQ batteries with both verbal and nonverbal subscales. Nonverbal IQ assessments, such as the UNIT and Test of Nonverbal Intelligence (TONI; Brown, Sherbenou, & Johnsen, 1982), do not require verbal responses and primarily focus on visuospatial skills and fluid problem solving (DeThorne & Schaefer, 2004). Of the completely nonverbal IQ assessment batteries, only the UNIT and TONI meet recommendations of psychometric criteria. The UNIT is the only completely nonverbal IQ test that tests a broad range of cognitive abilities (DeThorne & Schaefer, 2004).

Table 1.

Selected characteristics of the most frequently used nonverbal IQ assessments in development language disorder research.

Test Nonverbal only? Nonverbal subtests
Columbia Mental Maturity Scale (Burgemeister et al., 1972) No Pick the picture that does not belong with the others
Kaufman Assessment Battery for Children, Second Edition (Kaufman & Kaufman, 2004) No Face Recognition
Hand Movement
Triangles
Matrix Analogies
Spatial Memory
Photo Series
Leiter International Performance Scale–Third Edition (Roid, Miller, Pomplun, & Koch, 2013) No Figure Ground
Design Analogies
Form Completion
Sequential Order
Repeated Patterns
Classification
Paper Folding
Raven's Progressive Matrices (Raven & Court, 1998) Yes Progressive Matrices
Test of Nonverbal Intelligence–Third Edition (Brown et al., 1997) Yes Abstract figure problem solving
Wechsler Intelligence Scale for Children–Fourth Edition (Wechsler et al., 2004) No Object Assembly
Block Design
Matrix Reasoning
Picture Concepts
Universal Nonverbal Intelligence Test–Second Edition (Bracken & McCallum, 2016) Yes Symbolic Memory
Spatial Memory
Cube Design
Analogical Reasoning
Open in a new tab

The UNIT's four subtests target multiple CHC cognitive ability factors, while the TONI's one task focuses on fluid intelligence (i.e., the ability to solve novel problems independent from previous knowledge). If the symbolic memory and cube design tasks both assess fluid intelligence in the visuospatial cognitive domain, the difference between the tasks is the domain-general cognitive abilities controlled for through task demands. For symbolic memory, removing the stimulus requires students to attend to their representation of the stimulus in short-term memory in order to accurately complete the task. In contrast, the cube design stimuli are available for reference, and processing speed (i.e., the time required to complete the task from stimulus presentation to accurate cube completion) is the cognitive ability targeted (Bracken & McCallum, 1998; DeThorne & Schaefer, 2004; Flanagan & McGrew, 1998). In addition to task demands, the relationship of cultural practices and cognition may also impact the performance of bilingual children with and without DLD.

Cultural Implications of Nonverbal Cognitive Assessment

Cognitive tasks can be viewed as measuring skills gained through experiences within a cultural context. It is possible to identify cognitive skills that are embedded in culturally specific situations and consequently tied to social norms (Ortiz, 2001). An example of a cognitive test with high cultural loading is the UNIT Analogical Reasoning subtest, where participants complete a geometric or conceptual matrix by pointing to one of four possible responses depicting geometric shapes or common objects. In addition to potential bias through expectations of exposure to cultural content, nonverbal cognitive tests may also be biased by patterns of conceptual encoding.

Nonverbal directions are posited to be more culturally specific and, consequently, culturally biased than verbal instructions (Ortiz, 2001). Gestures reflect culturally specific encoding of abstract concepts (Slobin, 2003). A difference encountered for young Spanish–English bilinguals is that English verbs encode manner of movement (e.g., move up), while in Spanish, verbs encode path of motion but not manner (e.g., subir). Divergent patterns of encoding abstract conceptual knowledge are reflected in differential patterns of gestures during L2 use by adult Spanish-speaking English language learners when compared to monolingual English speakers (Negueruela, Lantolf, Jordan, & Gelabert, 2004). Consequently, nonverbal gestured directions from the UNIT intended to encode action could be culturally biased toward English cultural abstractions. Examples of potentially problematic gestures on the Symbolic Reasoning and Cube Design subtests on the UNIT include, but are not limited to, (a) palm rolling (i.e., gesturing with palms up and figures together by rotating wrists toward body in small circles) to indicate “go ahead” or “you try it now” (Bracken & McCallum, 1998, p. 48) and (b) hand waving (e.g., moving an open hand horizontally, with palms up) to indicate that stimuli should be considered as a group from which the examinee should choose (Bracken & McCallum, 1998, p. 48).

Summary and Current Research Questions

Two nonverbal IQ tasks, symbolic memory and cube design, require children to draw on visuospatial skills, short-term memory, and processing speed. Tasks requiring nonverbal gestures to convey specific manner of action (e.g., symbolic memory) may be especially difficult for Spanish-speaking English language learners due to different culturally specific encoding of manner of motion. Furthermore, tasks with higher verbal mediation are expected to be more difficult for children with DLD (e.g., symbolic memory) than tasks with lower verbal mediation (e.g., cube design).

Due to absent comparable literature in the area of nonverbal cognition and DLD in young bilinguals, the CHC theoretical framework provides testable hypotheses regarding the performance of bilingual children with TD and DLD on nonverbal cognitive tasks. Specifically, based on the CHC, we predict that children with TD and DLD would have similar performance on tasks more dependent on processing speed (e.g., cube design) and dissimilar performance on tasks dependent on short-term memory (e.g., symbolic memory), with children with DLD underperforming their peers with TD. Therefore, we ask the following questions:

  1. Is performance on cube design and symbolic memory similar for bilinguals and the monolingual UNIT norming sample?

  2. Is performance on cube design and symbolic memory similar for bilinguals with TD and DLD?

Method

Participants

Children who participated in this study were a subset from a larger study, the Diagnostic Markers study (see Bohman, Bedore, Peña, Mendez-Perez, & Gillam, 2010; Gillam, Peña, Bedore, Bohman, & Mendez-Perez, 2013; Peña, Gillam, Bedore, & Bohman, 2011). Participants were recruited from two districts in Central Texas and one school district in Northern Utah that served a large proportion of bilingual Latino children. Parents granted informed consent and completed demographic questionnaires at the time of recruitment.

A total of 169 Spanish–English bilingual children (146 with TD and 23 with DLD) between the ages of 5;4 and 6;2 (years;months) participated in the study. Children who were included in the study were Hispanic, spoke English and Spanish at least 20% of the time, and completed the Bilingual English–Spanish Oral Screener (Peña, Gutiérrez-Clellen, Iglesias, Goldstein, & Bedore, 2018) before beginning kindergarten. Children completed nonverbal cognitive testing during the kindergarten year and language testing in both kindergarten and first grade. Selected demographic characteristics are presented in Table 2. In the DLD group, there were 12 boys. In the TD group, there were 73 boys. Sixteen of the children with DLD and 113 of the TD group received free or reduced lunch. There was no statistical significance between group differences for the reported demographic variables.

Table 2.

Selected demographics of Spanish–English bilingual kindergartners with typical development (TD) and developmental language disorder (DLD).

Demographic measures TD
M (SD)		 DLD
M (SD)
Age at kindergarten in months 69.43 (4.8) 68.38 (4.8)
Age of first English exposure in years 2.1 (1.75) 2.19 (1.69)
% Input–output ratings in English 48.98 (20.59) 54.31 (20.97)
Mother's education, Hollingshead 2.72 (1.66) 2.35 (1.60)
Open in a new tab

Language Ability Status

Children were identified with DLD on the basis of three expert bilingual speech-language pathologists who rated (a) English and Spanish semantics and morphosyntax responses on the Bilingual English–Spanish Assessment (Peña et al., 2018), (b) elicited narratives in Spanish and English, (c) English responses from the Test of Language Development–Primary: Third Edition (Newcomer & Hammill, 1997), (d) English narratives from the Test of Narrative Language (Gillam & Pearson, 2004), and (e) parent ratings from the Instrument to Assess Language Knowledge (ITALK; Peña et al., 2018). Overall judgment of impairment was based on a 6-point scale (0 = severe/profound, 1 = moderate, 2 = mild, 3 = low normal, 4 = normal, and 5 = above normal). Children were identified with DLD if two of the three raters assigned a rating of 2 or less for both Spanish and English. See Gillam et al. (2013) for additional information about the classification process. Determination of ability status occurred after testing was completed; therefore, testers were blind to the ability status of children.

Measures

Table 3 summarizes information about the measures used to assess language performance to determine DLD status for this study, including descriptive statistics for the TD and DLD groups. The Symbolic Memory and Cube Design subtests of the UNIT (Bracken & McCallum, 1998) were administered to assess nonverbal cognitive performance. In the Symbolic Memory subtest, the child is presented with a series of simple pictures (e.g., boy, girl, baby) in two different colors (green and black) arranged in a specific predetermined order and then asked to replicate the sequence from short-term memory using plastic chips that have the same pictures drawn on them. In the Cube Design subtest, the child is given green-and-white blocks with geometric designs and asked to replicate a pictured model by manipulating the blocks in a way that matches the model within a specified time limit. Scaled scores are reported. Reliability coefficients for the 5- to 7-year-old norming sample ranged from .84 to .80 for the Symbolic Memory subtest and from .84 to .81 for the Cube Design subtest. A review of 10% of the responses determined coders were 98% accurate in scoring symbolic memory performance and 88% in scoring cube design performance.

Table 3.

Descriptive statistics for measures used to determine language ability status.

Language Test Domain Reliability TD
M (SD)		 DLD
M (SD)
Nonverbal UNIT Sym memory α = .80–.84 9.82 (2.69) 6.95 (2.66)
Cube design α = .81–.84 10.00 (2.00) 9.33 (2.76)
Both ITALK Language use Sensitivity and specificity of 0.80 with a 4.18 cutoff score 5.90 (0.30) 3.57 (0.51)
English BIOS Language input NA 48.98 (20.59) 54.31 (20.97)
BESA Morphosyntax α = .88–.89 42.17 (25.35) 18.44 (14.38)
Semantics α = .58–.73 48.46 (15.78) 33.43 (14.68)
TOLD-P:3 Morphosyntax α > .96 74.12 (12.21) 63.86 (6.10)
Semantics α > .96 76.14 (13.31) 68.33 (8.60)
TNL Production α = .84–.74 5.27 (2.53) 2.57 (1.47)
Comprehension α = .81–.91 5.68 (2.23) 3.86 (1.40)
Narratives Grammar 98% coder accuracy on errors 46.35 (21.68) 26.83 (19.23)
Spanish BIOS Language input NA 51.02 (20.59) 45.69 (20.97)
BESA Morphosyntax α = .88–.90 58.40 (23.73) 26.25 (17.00)
Semantics α = .66–.70 50.14 (16.22) 31.97 (15.37)
Narratives Grammar 97% coder accuracy on errors 70.67 (20.35) 53.41 (25.87)
Open in a new tab

Note. TD = typical development; DLD = developmental language disorder; UNIT, Universal Nonverbal Intelligence Test; Sym = symbolic; ITALK = Instrument to Assess Language Knowledge; BIOS = Bilingual Input Output Survey; BESA = Bilingual English–Spanish Assessment; TOLD-P:3 = Test of Language Development–Preschool: Third Edition; TNL = Test of Narrative Language.

Data Analyses

Only children with complete data were included in the analyses. The data were analyzed in three stages. First, descriptive statistics for the UNIT subtests were calculated to investigate relative performance for the bilingual children in the study. Next, t tests were calculated to compare the bilingual group's performance to the norming sample for the UNIT. A univariate analysis of variance (ANOVA) was conducted to compare the effect of language ability status on symbolic memory and cube design performance in kindergarten. Finally, effect sizes were calculated using Cohen's d to determine the magnitude of difference between TD and DLD group performance.

Results

Descriptive statistics were calculated to evaluate task performance across language ability and are summarized in Table 3. t tests were used to compare group performance on symbolic memory and cube design for the UNIT 5- to 7-year-old norming sample and the bilingual TD and DLD groups, presented in Table 4. For symbolic memory performance, there was a significant difference between the norming sample (n = 46, M = 10.52, SD = 2.69) and the DLD group, with a large effect size, but there was not a significant difference for the TD group. There was no significant difference on cube design performance between the norming sample and either bilingual group. Univariate analyses for the effect of language ability status on nonverbal cognitive performance showed that ability group differences were significant for symbolic memory performance but not for cube design, as shown in Table 4. There is a 98% probability that a child selected at random from the norming sample and TD bilingual group would score higher than a child in the bilingual DLD group based on the effect sizes on the ANOVA and t tests, reported in Table 4 (McGraw & Wong, 1992).

Table 4.

Comparing nonverbal cognitive performance for Spanish–English bilingual kindergartners with typical development (TD) and developmental language disorder (DLD) performance and the Universal Nonverbal Intelligence Test (UNIT) norming sample.

UNIT TD compared to DLD
 TD and DLD compared to norming sample
Combined
 TD
n = 146
 DLD
n = 23
F p d (ES) 95% CI df t p d (ES) df t p d (ES)
SM 20.96 < .001 1.07 1.65–4.09 190 1.53 .13 0.26 67 5.06 < .001 1.33
CD 1.83 .08 0.32 0.51–1.85 190 1.36 .18 0.26 67 1.80 .08 0.28
Open in a new tab

Note. CI = confidence interval; SM = symbolic memory; CD = cube design.

Conclusion

Performance on Nonverbal Cognitive Tasks Similar for Cultural Groups

The hypothesis that there is a difference in symbolic memory and cube design nonverbal cognitive task performance for bilingual children from Latino households was not supported. The performance of bilingual participants with TD did not statistically differ from the TD norming sample on symbolic memory or cube design. Lack of significant differences in performance between the norming sample and the Latino TD group extends the findings of Bosse and Papafragou (2018). In their study of German–English bilingual university students, they found that language-specific encoding for visuospatial concepts did not significantly impact nonverbal spatial memory performance. Here, we found that language-specific encoding of manner of motion did not significantly impact nonverbal cognitive performance in tasks that required specific manipulation of objects.

DLD Status Interacts With Nonverbal Cognitive Performance

t-test and ANOVA results confirm hypotheses of differential nonverbal cognitive task performance for DLD but not TD groups. The performance of the bilingual DLD group was significantly lower than both the monolingual and bilingual TD groups on a measure of nonverbal cognitive tasks (i.e., specifically symbolic memory), though not on cube design. Our study extends findings of processing deficits in bilingual Spanish-Catalan–speaking preschoolers to bilingual Spanish–English kindergartners (Aguilar-Mediavilla et al., 2014). The observed differential profile of nonverbal IQ task performance between bilingual children with DLD and their bilingual peers with TD and the monolingual norming sample of the UNIT adds to the growing body of literature documenting abilities and deficits in cognitive processing in the DLD population.

Discussion

This study adds nuance to the literature of DLD and bilinguals with DLD and provides insight into how underlying cognitive abilities support complex behavioral data across domains of performance. It expands the body of literature describing performance deficits in children with DLD on nonverbal cognitive skills to include early sequential Spanish–English bilingual children. Moreover, the study demonstrates that deficits in nonverbal cognitive skills are not equal across tasks for bilingual children with DLD.

While nonverbal cognitive testing is considered best practices for children with limited English proficiency, the results presented here point to the need for circumspect interpretation of results by school-based practitioners and researchers using nonverbal measures in the DLD population. Clinicians and researchers should not presume that measures of nonverbal cognitive testing reflect a potential for performance discrete from and in contrast to observable language behavior. Special care should be given to examining the specific skills targeted by a given assessment, nonverbal cognitive or verbal language. In addition to informing diagnostic decisions about the locus and severity of suspected deficits, nonverbal IQ testing can guide clinicians in planning intervention appropriate for a child's relative strengths and weaknesses in cognitive abilities (DeThorne & Schaefer, 2004). For example, the use of visuospatial strategies to facilitate language learning in bilingual children during language intervention is supported by the findings in this study.

Processing speed and short-term memory play a central role in learning in the CHC framework and are targeted in the cube design and symbolic memory tasks, respectively. Further investigation is needed to determine which facets of these cognitive abilities drive the significant difference in symbolic memory performance for bilingual DLD children observed in the current study. Research on nonverbal IQ tasks in the bilingual DLD population that includes independent measures of discrete aspects of memory (e.g., dual task performance, figure learning tasks, sorting span) and processing speed (e.g., reaction time from 5-choice) could inform theoretical models of cognition in bilingual children with DLD. Additional research is also needed to disambiguate the interaction of crystallized and fluid intelligence, verbal mediation, and short-term memory to better translate findings of cognitive deficits in the DLD population to effective assessment and intervention.

A better understanding of the development of gestures in school-aged Spanish–English bilinguals with DLD is necessary to contextualize performance on nonverbal tasks. The lack of significant difference between the bilingual TD group and the norming sample did not support previous findings of divergent gesture use in adult Spanish–English bilinguals (Negueruela et al., 2004). The participants in this study were, on average, exposed to English by the age of 3 years. It could be that the gestures used in the UNIT are recognizable to the children in the study due to acculturation or because the gestured directions are similar enough to previous gestural input from parents who live and work in mainstream communities. Or, it is possible that the Spanish–English bilingual administrators of the UNIT shared culturally influenced gestures with the participants. Overall, Ortiz model of nonbiased assessment of Latino children provides qualitative insight into cultural sensitivity to clinicians but needs more thorough controlled quantitative experimental data to support or reject the model's conclusions (Ortiz, 2001; Ortiz, Flanagan, & Dynda, 2002).

Finally, findings of significant differences in nonverbal cognitive performance also raise questions about the interpretation of studies that match peers with TD and DLD on cognitive measures, such as IQ batteries. Depending on the nonverbal cognitive testing administered and consequent biases, the DLD group comparisons may not be generalizable to clinical populations (Botting, 2005; Earle et al., 2017).

Limitations

This study had a restricted DLD sample size of 23 participants. Consequently, the results of the ANOVAs lack the certainty associated with studies with larger samples. Including specific measures of processing speed and short-term memory across cognitive domains (psychomotor, auditory, etc.) would also allow for greater clarity in the commonalities between cognitive abilities tested and these memory networks. Furthermore, in the current study, the instructions given during the administration of the UNIT (e.g., frequency of specific gestures, use of verbal language) were consistent with the instruction manual to use “other gestures as necessary to encourage the examinee to respond” (p. 55) and driven by the administrator–child interaction, as opposed to a rigidly constricted research protocol. The interaction of mode of instruction and cultural influences are therefore only answerable in the broadest possible sense. A study with more controlled procedures for test administration may find different results.

Future Directions

Using video recordings to analyze gesture use in UNIT test administration would permit a more definite study of the interaction between culture, gesture, and test performance. Additionally, future research should determine how learning mechanisms dependent on fluid intelligence mediate nonverbal cognitive skill performance and crystallized language knowledge in bilingual and monolingual children with TD and DLD. For these groups, tasks that depend on encoding information with more stable presentation in short-term memory, such as cube design and semantics, should be positively related. Likewise, symbolic memory should be related to phonological and syntactic tasks given their shared reliance on encoding information with dynamic representation in short-term memory.

Of additional interest are the differences in the relationships between nonverbal cognitive skills and verbal language performance over time between TD and DLD children. The association between language and cognitive ability appears to be dynamic and changes over time. Bilinguals are hypothesized to experience different cognitive and linguistic developmental trajectories than monolingual children in areas such as inhibition and metalinguistics (Luk, Anderson, Craik, Grady, & Bialystok, 2010). Children with DLD are also thought to experience differential trajectories than children with TD in language and cognition (Gallinat & Spaulding, 2018). The intersectionality of bilingualism, cognition, and language ability status is an open question for future research.

Acknowledgments

This research was supported in part by Grant R01 DC010366 and DoE LEADER Grant H325D140096. We would like to thank the families who participated in this research.

Funding Statement

This research was supported in part by Grant R01 DC010366 and DoE LEADER Grant H325D140096.

References

  1. Aguilar-Mediavilla E., Buil-Legaz L., Pérez-Castelló J. A., Rigo-Carratalà E., & Adrover-Roig D. (2014). Early preschool processing abilities predict subsequent reading outcomes in bilingual Spanish–Catalan children with specific language impairment (SLI). Journal of Communication Disorders, 50, 19–35. [DOI] [PubMed] [Google Scholar]
  2. Archibald L. M., & Gathercole S. E. (2007). Nonword repetition in specific language impairment: More than a phonological short-term memory deficit. Psychonomic Bulletin & Review, 14(5), 919–924. [DOI] [PubMed] [Google Scholar]
  3. Bohman T. M., Bedore L. M., Peña E. D., Mendez-Perez A., & Gillam R. B. (2010). What you hear and what you say: Language performance in Spanish–English bilinguals. International Journal of Bilingual Education and Bilingualism, 13(3), 325–344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bosse S., & Papafragou A. (2018). Does language affect memory for object position? A crosslinguistic comparison. Spatial Cognition & Computation, 18(4), 285–314. [Google Scholar]
  5. Botting N. (2005). Non-verbal cognitive development and language impairment. The Journal of Child Psychology and Psychiatry, 46(3), 317–326. [DOI] [PubMed] [Google Scholar]
  6. Bracken B. A., & McCallum R. S. (1998). Universal Nonverbal Intelligence Test. Austin, TX: Pro-Ed. [Google Scholar]
  7. Bracken B. A., & McCallum R. S. (2016). Universal Nonverbal Intelligence Test (2nd ed.). Austin, TX: Pro-Ed. [Google Scholar]
  8. Brown L., Sherbenou R. J., & Johnsen S. K. (1982). Test of non-verbal intelligence. Austin, TX: Pro-Ed. [Google Scholar]
  9. Brown L., Sherbenou R. J., & Johnsen S. K. (1997). Test of nonverbal intelligence (3rd ed.). Austin, TX: Pro-Ed. [Google Scholar]
  10. Burgemeister B. B., Blum L. H., & Lorge I. (1972). Guide for administering and interpreting the Columbia Mental Maturity Scale (3rd ed.). New York, NY: Harcourt Brace. [Google Scholar]
  11. Cattell R. B. (1963). Theory of fluid and crystallized intelligence: A critical experiment. Journal of Educational Psychology, 54(1), 1–22. [DOI] [PubMed] [Google Scholar]
  12. DeThorne L. S., & Schaefer B. A. (2004). A guide to child nonverbal IQ measures. American Journal of Speech-Language Pathology, 13(4), 275–290. [DOI] [PubMed] [Google Scholar]
  13. Dollaghan C., & Campbell T. F. (1998). Nonword repetition and child language impairment. Journal of Speech, Language, and Hearing Research, 41(5), 1136–1146. [DOI] [PubMed] [Google Scholar]
  14. Earle F. S., Gallinat E. L., Grela B. G., Lehto A., & Spaulding T. J. (2017). Empirical implications of matching children with specific language impairment to children with typical development on nonverbal IQ. Journal of Learning Disabilities, 50(3), 252–260. [DOI] [PubMed] [Google Scholar]
  15. Flanagan D. P., & McGrew K. S. (1998). Interpreting intelligence tests from contemporary Gf-Gc theory: Joint confirmatory factor analysis of the WJ-R and KAIT in a non-white sample. Journal of School Psychology, 36(2), 151–182. [Google Scholar]
  16. Gallinat E., & Spaulding T. J. (2018). Differences in the performance of children with specific language impairment and their typically developing peers on nonverbal cognitive tests: A meta-analysis. Journal of Speech, Language, and Hearing Research, 57(4), 1363–1382. [DOI] [PubMed] [Google Scholar]
  17. Gillam R. B., & Pearson N. A. (2004). TNL: Test of narrative language. Austin, TX: Pro-Ed. [Google Scholar]
  18. Gillam R. B., Peña E. D., Bedore L. M., Bohman T. M., & Mendez-Perez A. (2013). Identification of specific language impairment in bilingual children: I. Assessment in English. Journal of Speech, Language, and Hearing Research, 56(6), 1813–1823. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kaufman A. S., & Kaufman N. L. (2004). Kaufman Brief Intelligence Test–Second Edition (KBIT-2). Circle Pines, MN: AGS. [Google Scholar]
  20. Leonard L. B. (2014). Children with specific language impairment. Cambridge, United Kingdom: MIT Press. [Google Scholar]
  21. Luk G., Anderson J. A., Craik F. I., Grady C., & Bialystok E. (2010). Distinct neural correlates for two types of inhibition in bilinguals: Response inhibition versus interference suppression. Brain and Cognition, 74(3), 347–357. [DOI] [PubMed] [Google Scholar]
  22. McGraw K. O., & Wong S. P. (1992). A common language effect size statistic. Psychological Bulletin, 111(2), 361. [Google Scholar]
  23. Negueruela E., Lantolf J. P., Jordan S. R., & Gelabert J. (2004). The “private function” of gesture in second language speaking activity: A study of motion verbs and gesturing in English and Spanish. International Journal of Applied Linguistics, 14(1), 113–147. [Google Scholar]
  24. Newcomer P. L., & Hammill D. D. (1997). Test of Language Development–Primary: Third Edition (TOLD-P:3). Austin, TX: Pro-Ed. [Google Scholar]
  25. Newton E., Roberts M. J., & Donlan C. (2010). Deductive reasoning in children with specific language impairment. British Journal of Developmental Psychology, 28(1), 71–87. [DOI] [PubMed] [Google Scholar]
  26. Ortiz S. O. (2001). Assessment of cognitive abilities in Hispanic children. Seminars in Speech and Language, 22(1), 17–38. [DOI] [PubMed] [Google Scholar]
  27. Ortiz S. O., Flanagan D. P., & Dynda A. M. (2002). Best practices in working with culturally diverse children and families. In Thomas A. & Grimes J. (Eds.), Best practices in school psychology IV (pp. 337–351). Bethesda, MD: National Association of School Psychologists. [Google Scholar]
  28. Peña E. D., Gillam R. B., Bedore L. M., & Bohman T. M. (2011). Risk for poor performance on a language screening measure for bilingual preschoolers and kindergarteners. American Journal of Speech-Language Pathology, 20(4), 302–314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Peña E. D., Gutiérrez-Clellen V., Iglesias A., Goldstein B., & Bedore L. M. (2018). BESA: Bilingual English–Spanish Assessment Manual. Baltimore, MD: Brookes. [Google Scholar]
  30. Perrone-Bertolotti M., Rapin L., Lachaux J.-P., Baciu M., & Lœvenbruck H. (2014). What is that little voice inside my head? Inner speech phenomenology, its role in cognitive performance, and its relation to self-monitoring. Behavioural Brain Research, 261, 220–239. [DOI] [PubMed] [Google Scholar]
  31. Raven J. C., & Court J. H. (1998). Raven's progressive matrices and vocabulary scales. Oxford, United Kingdom: Oxford Psychologists Press. [Google Scholar]
  32. Robinson P. (2001). Individual differences, cognitive abilities, aptitude complexes and learning conditions in second language acquisition. Second Language Research, 17(4), 368–392. [Google Scholar]
  33. Roid G. H., Miller L. J., Pomplun M., & Koch C. (2013). Leiter International Performance Scale (3rd ed.). Dale Wood, IL: Stoelting. [Google Scholar]
  34. Saar V., Levänen S., & Komulainen E. (2018). Cognitive profiles of Finnish preschool children with expressive and receptive language impairment. Journal of Speech, Language, and Hearing Research, 61(2), 386–397. [DOI] [PubMed] [Google Scholar]
  35. Slobin D. I. (2003). Language and thought online: Cognitive consequences of linguistic relativity. In Gentner D. & Goldin-Meadow S. (Eds.), Language in mind: Advances in the study of language and thought (pp. 157–192). Cambridge, MA: MIT Press. [Google Scholar]
  36. Sokolov A. (2012). Inner speech and thought. New York, NY: Plenum Press. [Google Scholar]
  37. Tallal P., & Stark R. E. (1981). Speech acoustic-cue discrimination abilities of normally developing and language-impaired children. The Journal of the Acoustical Society of America, 69(2), 568–574. [DOI] [PubMed] [Google Scholar]
  38. Tomblin J. B. & Nippold M. A. (Eds.). (2014). Understanding individual differences in language development across the school years. New York, NY: Psychology Press. [Google Scholar]
  39. Tomblin J. B., Records N. L., & Zhang X. (1996). A system for the diagnosis of specific language impairment in kindergarten children. Journal of Speech and Hearing Research, 39(6), 1284–1294. [DOI] [PubMed] [Google Scholar]
  40. Vugs B., Hendriks M., Cuperus J., & Verhoeven L. (2014). Working memory performance and executive function behaviors in young children with SLI. Research in Developmental Disabilities, 35(1), 62–74. [DOI] [PubMed] [Google Scholar]
  41. Wechsler D., Kaplan E., Fein D., Kramer J., Morris R., Delis D., & Maelender A. (2004). Wechsler Intelligence Scale for Children–Third Edition (WISC-IV). San Antonio, TX: The Psychological Corporation. [Google Scholar]

Articles from Journal of Speech, Language, and Hearing Research : JSLHR are provided here courtesy of American Speech-Language-Hearing Association

RESOURCES