Poster Sessions

PS1F01

Recognizing Language Disorder Indicators Across Linguistic Profiles: Gaps in Teacher-Reported Identification Competence

Genesis Arizmendi; University of Arizona
Raúl Rojas; University of Kansas

Disparities in language disorder identification among language minoritized students suggest that early recognition processes may vary across linguistic profiles in schools. This study examined determinants associated with teachers’ self reported competence in recognizing language disorder indicators. A national survey of 1,161 K–8 educators assessed deficit-oriented beliefs about bilingualism, exposure to language disorders training, and frequency of consultation with speech language pathologists (SLPs). Identification competence was endorsed more often for monolingual English-speaking students (55.6%) than for emergent bilingual (30.9%) or variety-speaking students (19.6%); 20.1% reported no competence for any group. Logistic regression estimated adjusted odds of competence for language minoritized profiles. Training was associated with higher odds (OR = 2.41, 95% CI [1.62, 3.57], p < .001), and more frequent SLP consultation showed a positive association (OR = 1.24 per unit, 95% CI [1.11, 1.39], p < .001). Deficit oriented beliefs were not independently associated (p = .57), and teacher role was not predictive (p = .148). Findings highlight modifiable professional supports linked to more equitable K-8 classroom identification.

Funding: ASHA AARC Award & UT Austin Early Career Provost Fellowship

---

PS1F02

Effects of vocabulary and memory in Czech sentence recall: the latent regression approach

Filip Smolík; Czech. Acad. Sci.,, Inst. Pschol.

Sentence recall is a widely used clinical measure, yet its underlying cognitive and linguistic components remain debated. The study investigated how vocabulary, phonological memory, and complex working memory contribute to sentence recall performance in Czech children with typical language development (N=179) and language disorder (N=57). Using norming data from a Czech assessment battery comprising sentence imitation and vocabulary comprehension, and including the WISC digit span tasks, the authors applied structural equation modeling to examine latent relationships among these abilities. Models incorporated factors for vocabulary, forward digit span (phonological memory), backward digit span (central executive), and sentence recall, with age included as a covariate. Model fit indices indicated good overall fit. In the two-group model, both vocabulary and phonological memory significantly predicted sentence recall, while the central executive component showed no significant effect. Vocabulary effects were stronger in typically developing children, whereas phonological memory had a larger effect in the DLD group. Overall, the findings support a strong role of language skills in sentence recall and suggest that the cognitive structure underlying recall may differ between typical and language-impaired children.

---

PS1F03

Feasibility of Two New Treatments for Object Relative Sentences

Samantha Lord; University of Arizona
Mia Parisoff; University of Arizona
Elena Plante; University of Arizona

Treatment research for complex sentences lags behind that for other types of language deficits.  The few available studies use very different methods, harnessing both implicit and explicit approaches to treatment.  This study presents feasibility data for a larger-scale study that directly contrasts implicit and explicit treatment methods for the treatment object relative clause constructions.  Ten children received 20 days of treatment over 10 weeks. The two treatment methods used the same complex sentence input, the same number of exposures to each sentence, and an overlapping set of visual materials but differed in terms of their reliance on implicit vs. explicit learning mechanisms. Generalization to untrained contexts was measure by three measures of complex sentence production (sentence production in a priming task and a sentence repetition task, and words recalled in the sentence repetition task) . The results suggest that both treatments can lead to generalization for some children.  However, in-treatment performance, particularly in the Explicit condition, was not a good indicator of subsequent generalization.

Funding from R01 DC021429.

---

PS1F04

Rate and Type of Code-Switching in Bilingual Children’s Narratives

Nadia Lopez-Padilla; University of Maryland, College Park
Jessica Nolasco; University of Maryland, College Park
José Ortiz; University of Maryland, College Park

This study examines the rate and types of code-switching produced by Spanish–English bilingual children during a narrative retelling task. Since approximately 20% of children in the United States are exposed to more than one language, understanding code-switching behaviors is essential. Although code-switching is common in bilingual expressive language, it is usually not incorporated into standardized language assessments. Narrative retelling provides a structured and natural context to examine these behaviors in detail. Participants included 26 typically developing bilingual children (ages 3;0–6;11) from the Washington, D.C. area. Language samples were collected in Spanish-only, English-only, and code-switching conditions via Zoom. Transcripts were coded for inter-sentential and intra-sentential switches, including insertions and alternations, using established coding procedures. Frequency counts were calculated for total words, utterances, and switch types. Findings from this ongoing study will clarify common code-switching patterns and produce more culturally and linguistically appropriate assessment practices for bilingual children.

---

PS1F05

Convergence Across Multiple Sources of Evidence in the Identification of Developmental Language Disorder in Bilingual Children

Joseph Hin Yan Lam; University of California, Irvine
Ryan Sainsbury; Utah State University
Jiali Wang; Texas A&M University
Lisa Bedore; Temple University
Ron Gillam; Utah State University
Elizabeth Peña; University of California, Irvine

This study examined how convergence across language domains in identifying developmental language disorder (DLD) in bilingual children, and the convergence patterns at the measurement and child levels. We examined the degree of converging evidence in identifying DLD in bilinguals, the latent profiles of bilingual children based on evidence patterns, and the relationship between bilingual exposure and evidence patterns. Participants included 186 Spanish–English bilingual kindergarteners. Parents’ and teachers’ reports,  standardized assessment, narrative language samples, and dynamic assessment were administered. ?-statistics showed stronger agreement between performance-based assessments than with parents’ reports. Latent profile analysis revealed a two-profile model based on performance rather than a distinct DLD subtype, and children with DLD distributed across both profiles. The convergence and agreement between evidence were largely similar across exposure groups and exposure group did not predict profile membership. These findings suggest that measure-level convergence in identifying DLD in bilingual children. However, DLD did not appear as a single form of impairment but varied weaknesses across measurements. Therefore, the diagnosis of DLD in bilinguals requires the interpretation of multiple pieces of evidence collectively.

Funding: R01DC007439-01 (Peña)

---

PS1F06

Examining the dimensionality of children’s own perception of language skills among Spanish-English bilingual children

Youzhi Yao; Texas A&M University
Marc Goodrich; Texas A&M University
Lisa Fitton; University of South Carolina
Jiali Wang; Texas A&M University
Anny Castilla-Earls; UT Dallas

Standardized assessments developed for monolingual populations often inadequately identify language disorders in Spanish–English bilingual children, and validated bilingual measures remain limited. The Houston Questionnaire (HQ), a child self-report measure of bilingual experience and proficiency, has shown associations with standardized language assessments. This study examined the dimensionality of the HQ and attempted to replicate its previously reported three-factor structure in a kindergarten sample.

Participants were 234 kindergarten and first-grade students from Texas and South Carolina. Examiners orally administered 25 HQ items in English or Spanish; children responded verbally or nonverbally using yes/no and Likert-scale formats. Confirmatory factor analysis evaluated model fit using ?², RMSEA, CFI, TLI, and SRMR.

The hypothesized three-factor model showed poor fit. Exploratory analyses indicated misfit related to friend-number items and response format effects. Removing problematic items and separating school-related experiences improved fit substantially.

Findings suggest the original factor structure does not replicate in this sample, limiting interpretation of its dimensional scores. Differences in age range, sample size, and instructional context may explain discrepancies. Alternative theoretically guided models will be evaluated.

R01ActfHD114547
R21ActfHD106072

---

PS1F07

Predicting Phonological Awareness from Toddlerhood: Contributions of Family- and Child-level Factors

Soujin Choi; University of Minnesota
Julia Nikolaeva; Boys Town National Research Hospital
Brittany Manning; Northwestern University
Lauren S. Wakschlag; Northwestern University
Elizabeth Norton; Northwestern University

Phonological awareness (PA) is an early predictor and core deficit of developmental dyslexia; however, it cannot be directly assessed in children until preschool age, limiting the opportunities for early identification. This study takes a novel approach to earlier identification by examining whether family- and child-level corollary factors assessed in toddlerhood can predict later PA. A longitudinal sample of 339 toddlers oversampled for early language and mental health risk was followed across preschool and early school-age. Family-level predictors included family history of reading difficulties, maternal education, and maternal nonword repetition. Child-level correlates included expressive vocabulary and irritability at age two. PA was assessed between ages 4.5-8 using age-appropriate standardized measures. Stepwise linear regression analyses showed that family-level factors accounted for 15.0% of variance in PA, with only maternal education as a significant predictor. Adding child-level predictors increased explained variance to 24.6%. Only vocabulary size at age two significantly predicted later PA beyond family factors. These findings demonstrate the value of integrating toddlers’ vocabulary size and maternal education and suggest feasible early indicators for identifying toddlers at risk of later reading difficulties.

Funding source: NIH R01DC016273, R01MH107652

---

PS1F08

Complex word reading among children with language-based learning disabilities

Kelsey Davison; Purdue University
Erin Dickinson; Purdue University
Carissa Mastrangelo; Massachusetts General Hospital Inst of Health Professionals
Alex Kaminsky; Massachusetts Institute of Technology
Karolina Wade; Massachusetts Institute of Technology
John Gabrieli; Massachusetts Institute of Technology
Joanna Christodoulou; Massachusetts Gen Hosp Inst of Health Professionals
Rebecca Marks; Purdue University

Children with language-based learning disabilities (LBLD) often struggle to learn to read. Yet these difficulties are highly heterogenous, reflecting variation in underlying linguistic strengths and weaknesses. Although research on LBLD has historically emphasized phonological challenges, other language skills and print-related experiences may differentially support reading, especially for complex, polysyllabic words. This study examined word-, child-, and environment-level predictors of complex word reading in 71 language-typical learners and 85 children with LBLD (ages 6-12 years). Item-level responses on a complex word reading task were analyzed using cross-classified generalized random-effects models. Children with LBLD demonstrated lower complex word reading accuracy. Across participants, vocabulary, morphological awareness (MA), and word reading fluency positively associated with performance. MA made a greater contribution to complex word reading in children with LBLD. Additionally, phonological awareness was uniquely associated with performance within this group. Parent-reported home literacy experiences related to performance only among language-typical children. Findings highlight the role of linguistic skills in complex word reading and suggest that supports should be tailored by developmental profile.

This work was supported by the NIH under awards R01HD106122 and F32HD110967.

---

PS1F09

Examining a Cross-Linguistic NWR Task for Farsi-Speaking Children

Tahmineh Maleki; Louisiana State University
Janna Oetting; Louisiana State University

Nonword repetition (NWR) is a useful clinical tool reducing effects of prior linguistic knowledge on children’s performance. However, bilingual children who acquire a second language (L2) alongside their first language (L1) often exhibit lower performance in NWR tasks in their L2 compared to their L2 monolingual peers. These findings suggest that there remain unwanted biases using language-specific NWR tasks. As such, there is  continued concern of overidentification of childhood language impairment (LI) among bilingual learners. To address these biases, Chiat’s (2015) cross-linguistic (CL-NWR) task is based on universal features of several languages. The current study examined the CL-NWR along with Dollaghan and Campbell’s English-specific NWR task ( DC-NWR) with 25 TD monolingual Farsi-speaking children aged 4;0 to 7;0 years who lived in Iran. The NWR tasks were administered via zoom with percentage phoneme correct and percentage item correct scoring. A significant main effect seen for task, and scores were higher using the CL-NWR task compared to the DC-NWR task.

Funding: Louisiana state University paid Tahmineh Maleki as a GA.

---

PS1F10

Associations between parent- and child-report of language-related behaviors and direct assessment in school-aged children

John Gallagher; San Diego State University; University of California, San Diego
Tatiana Ramos-Gallardo; San Diego State University; University of California,San Diego
Erica Gutmann; San Diego State University; University of California, San Diego
Alyson Abel; San Diego State University

Involvement of parents and children in language assessment is considered best practice, yet the accuracy of parent-reported language concerns is mixed, and the associations between children’s self-reported language skills and performance on direct language measures remain underexplored in research. This study investigated the relationships between parent- and child-reported language behaviors and direct assessment. Data was collected from 34 parent-child dyads of children ages 9;0-12;9 with and without Language Impairment (LI). Children received an omnibus language assessment. Parents and children completed a questionnaire probing the child’s frequency of language-related behaviors potentially characteristic of LI. Twelve of 27 parent-reported items and three of five child-reported items were significantly associated with direct assessment scores. Findings suggest that specific questions asked to parents and school-aged children can support the holistic assessment of LI. Recommendations regarding incorporating parent and child report into language assessment practices will be shared.

Funding Source: ED/OSERS-OSEP H325D230037

---

PS1F11

Word Characteristics and Item Performance for Nonword Repetition, Nonword Spelling, and Nonword Reading

Tiffany Wong; University of Arizona
Elena Plante; University of Arizona

Phoneme-level skills are shared across multiple language modalities.  Nonword repetition tasks are commonly used to assess phoneme skill in developmental language disorder (DLD).  However, nonword reading and nonword spelling tasks are less commonly used.  One hundred and three young adults, with and without DLD, completed the Nonword Repetition, Nonword Spelling, and Nonword Reading subtests from the standardization version of the TILLS-2.  Significant group differences were found for all three subtests, indicating phoneme deficits occur in all three modalities into the adult years.  A series of linear regressions were used to discover which word characteristics predicted the passing rates for the three sets of subtest items.  Results indicate modality differences in terms of the word characteristics associated with passing items for the three subtests.

---

PS1F12

Differential Predictors of Word Reading, Fluency, And Comprehension in Adults with and without Developmental Language Disorder

Nazmiye Atila Caglar; University of Delaware
Stephanie Del Tufo; University of Delaware
Frances Sayako Earle; University of Delaware

The aim of this study was to examine the phonological and cognitive-linguistic determinants of reading (word-level, fluency, comprehension) in adults with and without Developmental Language Disorder (DLD). 311 monolingual English-speaking adults aged 18–33 years (64 with DLD, 247 typically developing (TD)) completed standardized measures of phonological awareness (PA), rapid automatized naming (RAN), language comprehension (Token Test), global executive functions (GEC), and nonverbal cognitive abilities, as well as assessments of reading. Results revealed that RAN was a significant predictor of word-level reading and fluency in both groups, showing particularly strong effects for timed word reading. PA contributed to word-level reading outcomes in TD adults but not in adults with DLD. Reading comprehension was predicted by word-level reading only in TD adults.

Funding: This project was supported by grants F31DC014194, R21DC016391, and R01DC019901A1 awarded to FSE. NAC is supported under a YLSY scholarship awarded by the Turkish Ministry of National Education. The content is solely the responsibility of the authors and does not necessarily represent the official views of the sponsoring agencies.

---

PS1F13

Teaching words in two languages: parental input to typically developing and late-talking bilingual toddlers

Emily Bagan; University of Wisconsin – Madison
Caitlyn Slawny; University of Wisconsin – Madison
Margarita Kaushanskaya; University of Wisconsin – Madison

Parents may employ various strategies to teach their children words. There are few prior examinations of parent and child behaviors during word learning activities, particularly for bilinguals and children who are late-to-talk. We examined language choice and labeling used by bilingual English-Spanish toddlers and their parents during a parent-mediated word learning activity. Parent-child dyads completed a word learning activity in which parents were instructed to teach their children the names of two novel objects, each with a novel English-like label and Spanish-like label, using an interactive digital learning platform. Parent and child utterances were coded for language (English vs. Spanish) and novel word use. Findings will reveal how parents of bilingual toddlers, including those who are late-to-talk, may vary in their language choice and input frequency when teaching their children novel words, and factors influencing this variation. This is critical, given that parental language input shapes children’s vocabulary acquisition.

Funding Source: National Institutes of Health Grants RO1 DC020447

---

PS1F14

Language and executive functions among school-age autistic females

Amy Banasik; UW Madison Waisman Center
Marianne Elmquist; UW Madison Waisman Center
Megan Tennessen; UW Madison Waisman Center
Susen Schroeder; UW Madison Waisman Center
Angela John Thurman; UC Davis MIND Institute
Leonard Abbeduto; UC Davis MIND Institute
Audra Sterling; UW Madson Waisman Center

Language and cognitive skills, such as executive functions (EFs), play a key role in academic outcomes and adaptive skills. While language and EFs are understood to be closely related, this relationship is not yet well understood in autism spectrum disorder (ASD), despite language and EFs being known areas of weakness in ASD. Much of what has been reported focuses on autistic males, which leads to a gap in knowledge of language and EFs in autistic females. This poster will begin to close that gap by presenting data on a sample of school-age autistic females. Participants completed nonverbal computer-based EF tasks measuring inhibition, shifting, and working memory. Additionally, they completed two computer-based tasks of grammatical and lexical comprehension, and standardized measures of expressive and receptive grammar. Overall, participants performed well on all tasks, and preliminary results suggest a significant positive relationship between some EFs and the standardized language tasks. This poster will contribute to the growing body of work aimed at understanding the relationship between EFs and language in autistic females.

---

PS1F15

Attention Across Labeling Moments in Children’s Word Learning

Shalini Banerjee; Indiana University Bloomington
Daniel P Kennedy; Indiana University Bloomington
Linda B. Smith; Indiana University Bloomington
Ishanti Gangopadhyay; Indiana University Bloomington

Children’s word learning depends not only on the linguistic input they receive but also on their ability to focus on relevant information in the learning environment. While prior work shows that attention to social cues and referent objects facilitates word learning, less is known about how attention shifts across distinct moments within a labeling episode. Using eye-tracking, we examined 4-7-year-old children’s visual attention before labeling (e.g., “This is a…”) and during labeling (e.g., “… Dax”) while learning novel words. Before labeling, children looked more to the target object than to the speaker’s face. During labeling, children looked more to the speaker’s face than to the target object. Greater looks to the speaker’s face before and during labeling predicted higher word-learning accuracy, whereas greater looks to the target object were not associated with word-learning outcomes in either labeling moment. These findings highlight the moment-to-moment attentional dynamics in word learning and set the stage for future research on attentional mechanisms underlying language learning in children with learning difficulties.

This project was funded by the SLHS Graduate Student Research at Indiana University.

---

PS1F16

Do the Intentions of Parent’s Communication Matter for Their Young Multilingual Children?

Gabriela Barajas; University of California, Irvine
Giselle Ferrer; University of California, Irvine
Mariana Gomez Becerra; University of California, Irvine
Joseph Hin Yan Lam; University of California, Irvine
Stephanie Reich; University of California, Irvine
Natasha Cabrera; University of Maryland

This study explores the communicative intents of parents interacting with their children in Spanish/English households. Using existing video records from the longitudinal Baby Books 2 project, we will analyze mother-child and father-child interactions when children were 9 months of age. Participants included low-to-moderate income families from Southern California and the Washington D.C. area. Using a coding framework that is developed from extant literature and iteratively refined from insights emerging from the data, we identify specific parent communicative intents including prompting, responsive feedback, and nurturance to analyze interaction patterns across participating families for 101 multilingual  families. We expect that parents utilize these communicative intents dynamically to expose their children to a rich communicative context. This research provides a portrait of Spanish/English communicative interactions during infancy that can enhance communicative disorder interventions and home-interventions alignment for children from multilingual backgrounds who are at risk for language disorders.

Funding Source: UCI Undergraduate Research Opportunities Program (UROP), NICHD #R01HD078547-05-PIs: Natasha Cabrera and Stephanie Reich

---

PS1F17

Graduate Student Use of Gestures and Language Expansion Cueing Before and After Scaffolded Training of Shared Book Reading

Sheri Bayley; Nevada State University
Katrina Nicholas; Nevada State University
Aysiah Taylor; Nevada State University
Sarah Tempest; Southern Hills Hospital, HCA Healthcare

The purpose of our study was to test whether The Bundle of Learning (Landa, n.d.) materials – which use scaffolding for graduate student clinicians to target imitation, language, and play during shared book experiences – helped them increase their gesture and language expansion cueing during dialogic reading and self-assessment of such improvements. Twenty Speech-Language Pathology graduate students enrolled in a clinical seminar class participated. Analyses will include pre- and post-training gesture and language expansion cueing and pre- and post-training perceptions of competency in these skills. These results will demonstrate the efficacy of training graduate student clinicians to use shared book reading skills via the Bundle of Learning materials and compare observed behaviors to self-ratings of gesture and language expansion use and clinical confidence. Materials were purchased through a clinical donation from the Scottish Rite Foundation.

---

PS1F18

Unpacking the Response to Articulatory Cues: Individual Differences in Early Literacy Outcomes

Robyn Becker; Montclair State University
Grace Clark; Montclair State University
Iyad Ghanim; Kean University

Although some studies suggest articulatory cues may help enhance early literacy outcomes, research has not yet isolated their unique contribution to observed improvements. This study seeks to fill this gap by providing phonemic segmentation training using articulatory cues to two different comparison groups and analyzing reading/spelling outcomes. Twenty-seven four and five year-olds were seen individually for seven weekly sessions during a typical day in a preschool classroom. Participants were randomly assigned to one of three groups which differed in intervention strategy. Speech/language ability, age, and scores on attention and motivation assessments were included in analyses to explore if any of these factors interact with the between-group condition (intervention strategy) to contribute to phonemic segmentation, nonword reading, and nonword spelling outcomes. Group-level differences, as well as multilevel models were used for data analysis. Results suggested that articulatory cues may help younger, less attentive children as well as those who have weaker receptive language skills. This research contributes to setting the foundation for future work investigating implications for clinical populations, including children with speech/language disorders and dyslexia.

---

PS1F19

Referential Effects on Positional Consonant and Feature Accuracy in Preschoolers with and without DLD

Sara Benham; University of Vermont
Lisa Goffman; Boys Town National Research Hospital

Children with developmental language disorder (DLD) show deficits in the production of novel words, characterized by reductions in accuracy and increases in variability. However, prior work indicates that word forms become more stable when they are associated with meaning via an object referent. Words are organized in the lexicon based on the position of sounds within a word, and we ask whether particular positions show increased consonant substitution errors and production variability in DLD. Forty-two preschoolers with and without DLD imitated novel words multiple times in non-referential and referential conditions. Children with DLD were less accurate than TD children in all consonant positions and both groups were more accurate when words had no object referent. For children with DLD, the referent had no effect on accuracy, whereas for TD children, syllable 1 coda accuracy decreased when words were paired with a referent, revealing that the referent affects production accuracy differently in children with DLD. Findings are interpreted within a developmental account and highlight lexical-phonological relationships across time and within language status.

Funding: 1L40DC021328 (Benham), DC017904 (Benham); R01DC016813 (Goffman), R01DC004826 (Goffman)

---

PS1F20

Morphosyntactic Delays in Mandarin-speaking Children with Developmental Language Disorder: A Narrative Sample Analysis

Yixuan Song; Nanjing Normal University
Ying Hao; Nanjing Normal University

The developmental trajectory of morphosyntactic skills in Mandarin-speaking children with developmental language disorder (DLD) remains unclear. This study compared age-related changes (3–5 vs. 5–7 years) between DLD and typically-developing (TD) groups using narrative sample analysis. We focused on production of four morphemes (classifiers, aspect markers “-zhe”, “-le”, “-guo”) and six syntactic structures (equative sentences, existential sentences, “Ba”-construction, “Bei”-construction, serial-verb sentences, pivotal sentences). Further, we examined correlations between the production of these syntactic structures and macrostructure elements (setting, goal, action, consequence). Both groups showed age-related increase in classifiers, “-le”, existential sentences. Compared to the TD group, the DLD group showed persistently delayed production of classifiers, “-le”, “-zhe”, serial-verb sentences and pivotal sentences. In the TD group, only one macrostructure element was significantly correlated with the production of specific sentence structures, whereas in DLD group, multiple macrostructure elements were correlated significantly with specific sentence structures, in particular existential sentences. Overall, Mandarin-speaking children with DLD manifested a delay in morphosyntactic development and relied on existential sentences, a type of simple sentence, to facilitate the production of macrostructure elements compared to TD age-matches.

---

PS1F21

Feature-Level Analysis of Unprompted Imitation During Lexical Speech Recast Intervention in Down Syndrome

Angela Brinckerhoff; Vanderbilt University Medical Center
Stephen Camarata; Vanderbilt University Medical Center

Children with Down syndrome (DS) often experience reduced speech intelligibility; naturalistic interventions such as Lexical Speech Recast (LSR) may support their speech development. A prior case analysis of a 4-year-old child with DS demonstrated that his unprompted imitations of clinician recasts yielded higher speech accuracy than his original initiations, with a moderate effect size (d = 0.45). However, accuracy scores alone could not fully capture the nature of modifications made. This secondary case analysis examines feature-level modifications between the child’s initiations and his unprompted imitations during LSR sessions. Data were drawn from 11 Zoom-based sessions within a larger study. Utterances were transcribed and phonemes were coded using distinctive feature analysis. Modifications between initiations and unprompted imitations will be summarized descriptively to explore systematic patterns across the data at both the phoneme- and word-level.

This analysis will use materials from a larger study funded by the National Institute on Deafness and Other Communication Disorders (NIDCD R21DC019280). The authors acknowledge the support provided by this grant, but the data and conclusions herein are exclusively the authors’ and not endorsed by NIDCD.

---

PS1F22

Do speaker’s gestures help children remember and understand stories?

Chelsea Brown; University of Colorado Boulder
Eliana Colunga; University of Colorado Boulder

Gestures are a rich form of nonverbal communication. Language accompanied by gestures, either viewed or produced, has been found to be beneficial across many areas. However, the nuances of how and when gestures are beneficial remain unclear. One explanation is that viewing gestures may encourage listeners to construct a more robust mental representation. This account aligns with the Gesture as Simulated Action (GSA) framework, which proposes that gestures reflect motor activity arising from mental simulations of actions or perceptual states. Building on this framework, the present study uses a narrative comprehension task to examine whether observing gestures prompts comparable representations in listeners, thereby supporting inference making and more detailed memory formation.

Children, aged 3.5-6, viewed stories accompanied by gestures or not and answered memory and inference questions. Children who saw gestures recalled significantly more for questions directly tied to gestures, suggesting gestures strengthen memory for specific content. While no significant quantitative group differences emerged for inference questions, qualitative analyses suggest that when gestures align with content, children’s responses converge more closely with the narrator’s mental representation.

---

PS1F23

Do children leverage relationships among senses of polysemous novel words?

Ashley Reece; Univeristy of Wisconsin – Madison
Haley Vlach; University of Wisconsin – Madison

Previous work has revealed that children use prototypes to learn polysemous word senses (Lakoff, 1987). However, little work has documented how children use prototypes to learn the senses of different types of polysemous words, such as perceptual versus radial. The current study examines whether children take advantage of multiple relationships among the senses of a novel word to enhance their learning and retention of its distinct senses. Preschoolers were randomly assigned to a perceptual polysemy or radial polysemy condition. All children were taught four novel labels which were fast mapped on to the three different “senses” (i.e., objects).  In the perceptual condition, two of the three objects shared a distinguishing feature with the prototype but not with each other. In the radial condition, none of the objects shared any features with each other. The results reveal that children readily learn polysemous words with shared perceptual senses but may struggle with polysemous words that only have shared conceptual senses (radial polysemy).

We would like to thank the James S. McDonnell Foundation and the Wisconsin Alumni Research Foundation for funding this work.

---

PS1F24

Mandarin-English Bilingual Speech Sound Development in Preschool Children

Herman Cheah; University of Kansas
Margarethe McDonald; University of Kansas

Mandarin-English bilingual preschoolers are a growing but understudied population with distinct phonological challenges. This study examines speech sound development in 16 bilingual children (3;0 to 6;11) to better distinguish between difference and disorder. We compared production accuracy of shared versus non-shared phonemes and identified predictors of overall intelligibility. Participants completed standardized articulation tasks (GFTA-3, Mandarin Speech Test), while their parents completed questionnaires about their child’s language exposure. Phonemes were analyzed in matched triads (shared, English-unique, Mandarin-unique) controlling for frequency and age of acquisition. Preliminary results indicate higher accuracy for shared phonemes likely due to dual reinforcement. Additionally, errors on shared sounds were the strongest predictors of reduced intelligibility. This is contrary to our hypothesis that non-shared phonemes with complex articulation would drive accuracy deficits, while shared phonemes may serve as a protective factor in development. These findings provide data-driven guidance for culturally responsive assessment by highlighting specific phonemes most vulnerable to error in the Mandarin-English bilingual context.

Funding Source: American Speech-Language-Hearing Foundation New Century Scholars Research Grant to MM

---

PS1F25

Feasibility of Enhanced Milieu Teaching with Phonological Emphasis (EMT+PE) for Korean Children with Cleft Palate ± Without Cleft Lip (CP?±?L)

Michelle Choi; Arizona State University
Nancy Scherer; Arizona State University

Rationale: Children with cleft palate ± lip (CP±L) demonstrate reduced consonant inventories and delayed expressive language scores, increasing risk for later literacy and academic difficulties. This study examined the feasibility of Enhanced Milieu Teaching with Phonological Emphasis (EMT+PE), a naturalistic intervention shown to improve speech and language outcomes, in Korean-speaking families.

Methods: A multiple baseline design across behaviors was used to evaluate Korean parents’ implementation of EMT+PE strategies. Standardized assessments were administered at pre- and post-intervention, and spontaneous language samples were collected at pre, post, and during the intervention. Reliability of assessment measures and treatment fidelity were measured.

Results: Parent coaching resulted in significant increases in parents’ use of EMT+PE strategies. Children made significant gains in their vocabularies and consonant inventories, while their speech accuracy (i.e. Percent Consonants Correct (PCC)) declined as a tradeoff with vocabulary expansion.

Conclusions: EMT+PE via telepractice appears to be feasible for South Korean parents and shows promise for improving speech and language development for Korean children with CP±L.

Funded by the Institute for Social Science Research, Arizona State University

---

PS1F26

Beyond Task Accuracy: Comparing Planning Task Rule Violations between Preschoolers with and without Developmental Language Disorder

Quinn Harter; University of Arizona
Leah Kapa; University of Arizona

Children with developmental language disorder (DLD) frequently demonstrate deficits in executive function skills relative to peers with typical language (Niu et al., 2024), including on complex executive function measures like planning tasks (Abdul Aziz et al., 2017). In this project, we compared planning performance between preschoolers with and without DLD using the Tower of Hanoi (TOH) planning task. In addition to considering participants’ TOH accuracy based on the number of trials they successfully solved, we also compared the groups’ frequency and types of task rule violations, which are not necessarily reflected in their task accuracy scores. Participants with and without DLD did not differ in the number of TOH trials solved nor the percentage of their task moves that violated rules. However, the frequency of specific types of TOH rule violations varied between the groups. These findings suggest that even when children with and without DLD have equivalent task accuracy, their understanding of task rules and/or the strategies they employ to complete tasks may differ.

Research supported by NIH R21DC018624.

---

PS1F27

IEP status and semantic abilities of bilingual children with and without risk for language disorder

Samir Kassem; San Diego State University, University of California San Diego
Amy Pratt; University of Cincinnati
Elizabeth Peña; University of California Irvine
Ashley Sanabria; San Diego State University

A converging evidence approach to identification has shown promise for evaluating bilingual children’s risk of developmental language disorder (DLD). However, more research is needed to clarify associations across recommended evidential measures. This study compared the proportion of bilingual children with an Individualized Education Plan (IEP) who passed a parent-reported language screener to those who failed, and examined differences in Spanish and English semantic abilities across these groups. Parents of bilingual children in PreK–3rd grade (N = 421) completed the Inventory to Assess Language Knowledge (ITALK; Peña et al., 2018) and reported IEP status. Children completed semantic assessments in both languages. Results showed a higher likelihood of having an IEP alongside a failed ITALK among second and third graders, but not younger children. Semantic differences appeared only in English among first graders without IEPs. These findings suggest that ITALK use in early elementary grades should be paired with direct language screeners and that semantic measures should be used cautiously when identifying DLD risk in bilingual children.

This work was funded by the Institute of Educational Sciences IES R305A210136 PIs: Sanabria & Peña, 2021.

---

PS1F28

An Examination of Adolescents’ Procedural and Academic Discourse during Group Interaction

Angela Kim; Texas Christian University
Alice Regalado-Lee; Texas Christian University
Danielle Brimo; Texas Christian University

Collaborative learning (CL), where students work together to achieve an outcome, supports achievement, higher-order thinking, and positive social outcomes. Describing students’ language production in CL is foundational in developing classroom supports for students with communication disorders. Although students produce academic and procedural discourse to engage in CL, lexical and syntactic differences among these different discourses remain underexplored. Academic language supports learning through elaboration and explanation, while procedural language involves planning, coordinating, and adjusting. This study compared lexical and syntactic measures in procedural and academic discourses.

Twenty-two seventh-grade private school students were audio-recorded while completing a group assignment. Transcriptions were verified and analyzed using Systematic Analysis of Language Transcripts. Utterances were coded as academic or procedural. Measures included the number of different words (NDW), mean length of utterance in words (MLUw), and syntactic complexity. Descriptive results indicated 60.52% of utterances were task-relevant. Students produced nearly twice as many procedural as academic utterances (M ratio = 0.47). Planned paired samples t-tests will examine NDW, MLUw, and discourse complexity results.

Texas Christian University supported this work under the Harris College Student Research Grant.

---

PS1F29

Does Word Class Matter? The Impact of Syntactic Category on Children’s Contextual Word Learning

Ju Hee Kim; SUNY Binghamton
Jiayu Yu; SUNY Binghamton
Dawna Duff; SUNY Binghamton

Rationale: This study examined whether word class (noun, verb, adjective) affects the quality of inferences made by school aged children about i) meaning and ii) syntactic class of novel words presented in context.

Methods: As part of a larger study, 4th-6th graders (n=46) with typical language (CELF-5 CoreLanguageScore >= 85) read and listened to short stories (n=45) containing novel nonwords (n=9), balanced across nouns, verbs and adjectives. Children gave definitions of novel words (n=822) immediately after each story and at one-week post.  Definitions were scored separately for semantics and word class. Results are analyzed using mixed-effects models with word class as fixed effect and participant, item and time as random effects.

Results: Word class did not significantly predict semantics scores. Analysis of word class scores is in progress; we predict a significant effect.

Conclusions: This study contributes to understanding of variables that impact contextual word learning. We interpret our findings relative to theoretical accounts of word learning and in light of multiple variables which may impact contextual word learning, including word concreteness.

Funded in part by NIDCD, 1R21DC020018-01A1

---

PS1F30

Social Stories and Neuroimaging Environments: Supporting Preschoolers’ Health Literacy

Leslie Kokotek; University of Louisville, University of Cincinnati
Adhirai Garibaldi; University of Toronto
Karla Washington; University of Toronto
Jennifer Vannest; University of Cincinnati

Neuroimaging in preschoolers supports understanding of the brain-language relationships. However, challenges related to preparation for participating in the neuroimaging environment often limits data collection among young children. This study evaluated the feasibility of a brief, video-based social story in preparing 4- to 5-year-old children for neuroimaging. Typically developing children (TD, n=16) and children with developmental language disorder (DLD, n=8), viewed the video before their scan. Attempt and success rates were tracked across 35 sessions, with a scan considered ‘attempted’ if the child entered the scanner, and ‘successful’ if the neuroimaging protocol was completed. Collectively, 34 scans were attempted (97%) and 26 (74%) scans were completed. Group level attempt and success rates were also high for both groups (TD = 96% attempt, 77% success; DLD = 100% attempt, 67% success), exceeding rates reported in previous pediatric neuroimaging research. Ultimately, the video-based social story was a scalable, child-centered approach to supporting participation and health literacy in preschool neuroimaging.

Partial funding received from the National Institutes of Health (NIDCD R01DC09337) and by the Canadian Institutes of Health Research (Canada Research Chair, CRC-2022-00366).

---

PS1F31

Comparison of complexity-based morphosyntactic treatment outcomes for school-aged children with DLD or Down syndrome

Megan Rowe; Brigham Young University
Markelle Luddington; Brigham Young University
Kirsten Hannig Russell; Brigham Young University

Common difficulties observed in the language of English-speaking children with developmental language disorder (DLD) and children with Down syndrome (DS) include marking tense and agreement on verbs, including specific difficulty with BE verbs. Difficulties with syntactic movement often exacerbate these issues, resulting in increased omission errors for children with DLD and with DS. A complexity-based approach to treatment target selection is one suggested way to address these needs. Using single-case AB experimental design, we explored effects of complexity on a treated, more complex structure (i.e., auxiliary BE questions) and possible generalization to simpler but related structures (i.e., auxiliary BE sentences, copula BE questions, copula BE sentences) for two participants. Mixed effects were observed for treated items, yet generalization occurred for untreated items at varying levels for both participants. Outcomes provide additional support for a complexity-based approach when selecting targets for treatment of morphosyntax.

---

PS1F32

Different information supports verb learning and sentence processing in children with and without developmental language disorder

Justin Kueser; Boys Town National Research Hospital
Emma Kate Thome; Boys Town National Research Hospital
Jacqueline Zhang; San Diego State University University of California San Diego
Katherine Gordon; Boys Town National Research Hospital
Cowdrey Kaylee; Boys Town National Research Hospital

Two challenges in developmental language disorder (DLD) are sentence comprehension and verb vocabulary. These challenges are connected: verbs are the nexus of the sentence, linking agents and patients. Prior work showed that knowing patient-verb associations impacts sentence comprehension in DLD. However, despite this prior work, two gaps remain: First, work on verb learning in DLD has not targeted the patients associated with verbs. Second, work has not examined how verb learning relates to sentence processing. In Experiment 1, we taught 29 5-8-year-old children with and without DLD about patients associated with advanced verbs. Within participants, patients were taught in either a specific cooccurrence condition (e.g., you can swat moths/ladybugs/bees) or a semantic feature condition (e.g., you can swat bugs_that_fly). Learning was measured during learning phase sessions and one week later. In Experiment 2, we measured sentence processing with the same verbs with new nouns to test generalization. Teaching semantic features (vs. specific nouns) improved generalization and sentence comprehension for children with DLD. Semantic features make an ideal target to promote verb knowledge and sentence comprehension.

Funding from NIH/NIGMS P20GM109023.

---

PS1F33

How is oral language currently being supported in the classroom?

Theresa Pham; University of Western Ontario
Lisa Archibald; University of Western Ontario

Oral language is a critical skill essential for access to the school curriculum and long-term academic achievement. In this study, we asked how teachers and other educational professionals (e.g., speech-language pathologist (SLPs)) (n = 102) support children’s oral language development in the classroom via qualitative online survey. We found that oral language is currently being supported 1) through strategies specific to oral language strands, 2) indirectly through emotional literacy or general teaching strategies (e.g., cold calling), and 3) non-verbal communication (visuals, body language). Results illustrate how oral language is currently being supported through various strategies, and paves the way for improved education and training on best practices for supporting oral language.

The study was funded by the Ontario Ministry of Education.

---

PS1F34

How and when early interventionists recommend English-only to Latiné families with autistic children

Tatiana Ramos-Gallardo; San Diego State University/University of California San Diego
Ana Dueñas; University of Washington
Giang Pham; San Diego State University
Fernanda Castellón; University of California, Santa Barbara
Carmen Orendain Soto; The University of Texas Rio Grande Valley
Sarah Rieth; San Diego State University

Caregivers participating in early intervention (EI) programs report a lack of support for home languages and clinical recommendations to adopt English-only language policies. English language recommendations (ELRs) impact the quantity and quality of linguistic input children receive and exacerbate existing assimilation pressures that threaten children’s opportunities to engage with their home language, family, and community. To our knowledge, studies have not yet examined ELRs during intervention sessions. Using a sequential explanatory mixed-methods design, we examined 52 video recordings of caregiver-mediated EI sessions on the frequency of ELRs and their association with provider demographics. Using thematic analysis, we examined the nature of ELRs. Results showed that 11 of 16 providers made ELRs (38.5% of videos). ELRs were not associated with provider demographics. This presentation will discuss nine themes describing the content and context of ELRs. Findings suggest that systemic devaluation of Spanish and simultaneous prioritization of English underlie inequitable, harmful intervention practices (e.g., ELRs). Cultural/linguistic match did not eliminate ELRs in this study, underscoring the need for equitable pre-service training.

Funding is from IES R324A190076.

---

PS1F35

Tense Morphology in Clinicians’ Input during Complex Syntax Intervention

Danyang Wang; Towson University
Janna Oetting; Louisiana State University
Amanda Owen Van Horne; University of Delaware

Children with developmental language disorder (DLD) demonstrate weaknesses in both tense morphology and complex syntax across dialects of English, yet intervention targets often address these domains separately. This study examined whether clinician input during a complex syntax intervention provides rich and varied models of tense morphology. Using a secondary analysis of data from one arm of a randomized controlled teletherapy trial, we analyzed clinician utterances during an intervention targeting complement clauses embedded in science instruction for 10 children with DLD (4-7 years old). A total of 1,753 clinician utterances containing complement clauses were coded. Clinicians produced 2,491 tense morphemes, with regular past tense occurring most frequently, followed by verbal -s and irregular past tense. Past tense morphemes occurred primarily in independent clauses, whereas present tense forms occurred more often in dependent clauses; copula and auxiliary BE forms were largely confined to dependent clauses. Clinicians produced 47 unique matrix verbs, most with overt tense marking. Results suggest that complement clause interventions can simultaneously support tense morphology through dense, structurally informative input, supporting integrated approaches to grammar intervention.

---

PS1F36

Grammatical Morpheme Accuracy in Deaf and Hard of Hearing Children: Finiteness Marking and Allomorph Voicing Effects

Meganne Muir; Vanderbilt University
Angie Aikens; Vanderbilt University
Angie Walker; Vanderbilt University Medical Center
Mary Dietrich; Vanderbilt University, Vanderbilt University Medical Center
Erin Schuweiler; Kansas State School for the Deaf
Jena McDaniel; Vanderbilt University School of Medicine

Deaf and hard of hearing (DHH) children frequently demonstrate weaknesses in grammatical morphology, yet the source of these errors remains unclear. Two accounts from the developmental language disorder literature offer distinct explanations: vulnerability in finiteness marking versus reduced phonetic salience of unstressed morphemes. This study evaluates whether morphological weakness in DHH children aligns more closely with finiteness or surface phonetic factors.

Language samples from 33 DHH children (3;0–5;11) were analyzed at the token level using generalized linear mixed-effects models. Across Brown’s morphemes, children showed lower accuracy for finite than non-finite morphemes (ß = -1.36, p < .001). Third-person singular –s showed lower accuracy than plural –s, despite shared surface phonetic forms (ß = -1.69, p < .001). Associations between voiced and voiceless allomorphs were inconsistent, with an association observed only for past tense –ed.

Findings suggest that morphological weakness in DHH children is more closely related to finiteness than to surface phonetic factors, which showed only morpheme-specific associations.

Funding: Project REACH (H325D220072; U.S. Department of Education) and the National Institute on Deafness and Other Communication Disorders (R01DC021188).

---

PS1F37

From Data to Goals: Sources Guiding IEP Goals and Target Selection in School-Based SLP Practice

Rebecca Summy; The University of Alabama
Kelly Farquharson; Florida State University

School-based SLPs, who routinely serve students with developmental language disorder (DLD), must determine which language abilities warrant annual IEP goals while ensuring that intervention targets are academically relevant. Although standardized assessments are often required for eligibility determination, these measures frequently demonstrate psychometric limitations and rarely translate directly into functional, instructionally meaningful targets. Consequently, school-based SLPs must integrate multiple sources of data when developing IEP goals and selecting therapy targets; however, little is known about how these data sources are prioritized in everyday practice.

This study uses the Experience Sampling Method (ESM) to examine real-time clinical decision-making among elementary school SLPs. Participants received three brief surveys per day over five consecutive days. When SLPs reported recently targeting a language-related goal, they specified the language targets addressed and the data sources that informed their clinical decision-making. A separate one-time survey collected data regarding IEP goal development, collaboration patterns, and caseload characteristics.

Descriptive analyses will identify patterns in data source prioritization and explore differences based on demographic variables. Findings will inform understanding of clinical decision-making in school-based language intervention.

Funded by the American Speech-Language-Hearing Association’s Advancing Academic Research Award (AARC)

---

PS1F38 (WITHDRAWN)

Untold Stories of Children with Developmental Language Disorder: Personal Narrative Skills and Theory of Mind

Claire Neufeld; University of Alberta
Monique Charest; University of Alberta
Marleen Westerveld; Griffith University
Pamela Filiatrault-Veilleux; University of Alberta

Personal narratives are central to daily life but very challenging for children with developmental language disorder (DLD). Theory of Mind (ToM) has been associated with narrative abilities in other populations but has not yet been investigated in the context of personal narration in DLD. This study explores the personal narrative skills of 5- to 7-year-old children with DLD, both in terms of microstructure and internal state term (IST) use, and how they relate to their ToM skills. Twenty English-speaking children with DLD (mean age = 83.25 months; 9 boys) were assessed using the Global TALES protocol and two measures of ToM (i.e., one direct and one indirect measure). Preliminary results indicate that ToM ability is more strongly associated with verbal productivity and linguistic complexity than IST use in the personal narratives of young children with DLD. By examining discourse-level and social-cognitive skills, this study will help inform more meaningful assessment and intervention for children with DLD that better reflects daily communication.

This study was funded by the Social Sciences and Humanities Research Council (SSHRC) of Canada.

---

PS1F39

What predicts predictive processing?

Mariel Schroeder; Purdue University
Amanda Yuile; MGH Institute of Health Professions
Virginia (Claney) Outzen; Purdue University
Barbara Brown; Purdue University
Sharon Christ; Purdue University
MaryCarson Adams; Purdue University
Campbell Patterson; Purdue University
Risa Stiegler; Purdue University
Arielle Borovsky; Purdue University

Linguistic prediction facilitates real-time language processing and language acquisition. Converging evidence suggests that vocabulary supports prediction, but how morphosyntax, phonology, and executive function impact prediction is less clear. Using a large sample of 42-month-olds with heterogeneous linguistic and cognitive skills, we found that morphosyntax, but not vocabulary or phonology, supported semantic prediction of a sentence-final noun using combinatory information from the agent and action. These findings suggest that grammatical knowledge may play a larger role than expected in predictive language processing. Future analyses expected to be completed by the conference include examining how additional language and executive function measures as well as late talking status contribute to predictive language processing.

Funding: R01DC018593; F31DC023491

---

PS1F40

Exploring effortful production in a sentence recall task to distinguish between children with DLD and English language learners

Carol Miller; Penn State University
Sean Redmond; University of Utah
Tracy Preza; University of Illinois Urbana-Champaign
Pamela Hadley; University of Illinois Urbana-Champaign
Dancheng Liu; University at Buffalo
Abdlrahman Alabdallah; Penn State University
Jinjun Xiong; University at Buffalo

Indicators of effortful sentence production may help screening tasks distinguish children with developmental language disorder (DLD) from children who are English language learners (ELLs). Using the Redmond Sentence Recall (RSR) screener, we asked: Is sentence production more effortful for ELLs than matched children who are not ELLs? Is sentence production more effortful for children with DLD than ELL peers? We analyzed transcripts of RSR productions from 76 children who received ELL services, matched for grade and ethnicity to 76 children who did not. Another group of 58 children were identified as DLD. Dependent variables included stalls (fillers, pauses, and repetitions), revisions (alterations of words, morphemes, or syntactic structure), and words per minute (WPM). The ELL group had a significantly higher stall rate than the Match group but did not differ from them on revision rate or WPM. The DLD group did not differ from the ELL group on stall rate or revision rate but had significantly lower WPM. ROC analysis suggested that WPM on RSR is a promising variable for further research on discriminating children with DLD from ELLs.

NSF, IES, NIDCD

---

PS1F41

Implicit-Explicit Phonological Awareness Instruction for AAC Users: Evidence of Instructional Variability in School-Based Practice

Kelly Stewart Nace; University of Tennessee Health Science Center
Jillian McCarthy; University of Tennessee Health Science Center
Erinn Finke; University of Tennessee Health Science Center

Students who use augmentative and alternative communication (AAC) are at elevated risk for literacy difficulties, making explicit phonological awareness (PA) instruction critical for early word-reading development. Research consistently demonstrates that PA develops most effectively through systematic, explicit instruction rather than incidental literacy exposure. Speech-language pathologists (SLPs) frequently report teaching PA to AAC users, yet the alignment between instructional intent and explicit PA practice remains unclear. This study examined paired survey responses comparing SLPs’ self-reported PA instruction with coded descriptions of explicit PA activities. Open-ended instructional responses were coded to identify PA instruction (e.g., blending, segmentation, letter-sound mapping). A McNemar test evaluated within-participant alignment. Results indicated inconsistent instructional reporting. The asymmetry was not statistically significant, suggesting variability in conceptualization and implementation of PA instruction. Findings highlight inconsistent instructional coherence rather than absence of PA engagement, indicating the need for structured frameworks that translate evidence-based literacy instruction into explicit intervention for AAC users.

This project is funded through an OSEP leadership training grant (Project PAL; HD325D230072).

---

PS1F42

Does Gesture Use During Naturalistic Caregiver-Child Interaction Predict Later Vocabulary in Spanish-English Bilingual Toddlers?

Murielle Standley; Northwestern University
Alexa Nuñez Magaña; Northwestern University
Katie Meyer; Northwestern University
Sefela Yalala; Northwestern University
Adriana Weisleder; Northwestern University

Early gesture robustly predicts spoken language in monolinguals, yet less is known about predictive gesture–language relations in bilingual toddlers, who distribute communicative information across modalities and languages. Examining gesture in caregiver–child interactions may refine prediction of later language outcomes. We asked whether gesture use at 26 months predicts vocabulary at 33 months beyond concurrent spoken language and whether prediction differs by gesture type. Forty Spanish–English bilingual children completed a 20-minute naturalistic caregiver–child free-play interaction at 26 months and a vocabulary assessment at 33 months. Child language measures were derived from transcripts (MLUw, NDW), and gestures (deictic, representational, conventional) were coded, calculating gesture rate per minute. Gesture rate ranged from 0.58–5.70 (M = 2.06, SD = 1.22). Controlling for concurrent spoken language and covariates showed that greater gesture use predicted lower later vocabulary. Deictic and conventional gestures showed similar negative associations; representational gestures were not predictive. Greater gesture use at 26 months may reflect modality allocation rather than lower language ability in bilingual toddlers.

Funded by NIDCD R21DC018357 and a Graduate Research Grant from Northwestern University.

---

PS1F43

Sex Differences in Vocabulary and Receptive–Expressive Language Discrepancies in Autism: Evidence from Parent Report and Standardized Language Assessment

McKenzie Scoppa; University of Texas at Dallas
Pumpki Lei Su; University of Texas at Dallas

In typical development, girls show a modest vocabulary advantage over boys in toddlerhood, but whether autistic toddlers show similar patterns is unclear. Autistic children also show a receptive–expressive discrepancy that differs from typical development and may vary by sex and autism severity. This study examines sex differences in receptive and expressive vocabulary and discrepancy, autism severity-related effects based on ADOS Total Score, and whether discrepancy patterns converge across parent report (MCDI) and standardized assessment (PLS-5). Data include an ongoing cohort of 14 autistic children (7F; 7M). Overall, girls had larger, but non-significant receptive (Mgirls=435.6; Mboys=353.4) and expressive vocabulary (Mgirls=280.1; Mboys=265.0) and receptive-expressive vocabulary discrepancy (Mgirls=155.4; Mboys=88.4). A significant autism severity × sex interaction revealed that autism severity was associated with a larger receptive–expressive discrepancy in girls, but not boys, demonstrating a moderation effect of sex. Discrepancy direction differed by measure: all children showed receptive>expressive profiles on the MCDI, whereas PLS-5 estimates were mixed. Findings underscore interpreting receptive–expressive profiles in the context of sex, autism severity, and assessment, rather than discrepancy patterns alone.

Funded by NIH R21DC021803.

---

PS1F44

If I Could Do It Again: How School SLPs Would Change Their Most Recent Assessment of Dual Language Learners

Adriana Valtierra; Vanderbilt University
Fiona Higgins; Vanderbilt University
Emily Byrd; Vanderbilt University Medical Center
Jena McDaniel; Vanderbilt University School of Medicine

Dual language learners (DLLs), or children from homes in which a language other than English is spoken, comprise 21% of school-age children. When DLLs are referred for a school speech-language evaluation, speech-language pathologists (SLPs) must differentiate communication disorders from differences that result from children learning multiple languages. Although the knowledge and resources surrounding evaluation of DLLs have improved in recent years, a discrepancy remains between how school SLPs conduct a DLL evaluation and their preferred practices given ideal evaluation conditions.

This study improves our understanding of the relation between ideal evaluation conditions, barriers, and actual evaluation practices for school SLPs working with DLLs. We used t-tests to compare the number of norm-referenced and alternative measures (a) used in a DLL evaluation and (b) endorsed given ideal evaluation conditions. Results suggest no significant difference for either case, but moderate effect sizes suggest we may be underpowered to detect a difference. Through qualitative analysis, we identified three major themes. A subset of SLPs reported barriers, with lack of time being the most common.

Funding: U.S. Department of Education (H325D220072, H325D230072)

---

PS1F45

Understanding How Early Childhood Environments Shape Children’s Language Development: A Descriptive Multi-Site Study

Stephanie Pedraza Marin; University of Iowa
Ethan Kutlu; University of Iowa
Katherine Pribyl; University of Iowa
Adrian Bradley; University of Iowa
Philip Combiths; University of Iowa

A common method for assessing children’s language development is the analysis of spontaneous language samples, which provide insight into how children use language in naturalistic contexts. Because language development is shaped by the environments in which children are raised, examining variation in home language experiences is critical for understanding differences in language development. This study examines how early childhood home environments shape children’s language development by comparing home language experiences across families recruited from different Neighborhood Centers in the Midwest. Spanish-English caregiver-child dyads were recorded during naturalistic home play sessions. Linguistic diversity in the naturalistic input will be examined through measures of sentence length and complexity, intelligibility, lexical diversity (number of different words NDW), verbosity (words per minute WPM), and code-switching frequency and flexibility. In addition, a social network analysis will be conducted on a multilingual questionnaire, with each child’s home network size and home language diversity serving as predictors of language outcomes. We hypothesize that children who are raised in larger networks with more linguistic diversity are expected to have increased lexical diversity, verbosity, and sentence intelligibility.

Funding: NIDCD-R21DC021249

---

PS1F46

Parent-Report and Standardized Measures as Indexes of Late-Talking in Bilingual Toddlers

Clara Merkel; UW-Madison, Waisman Center
Chenelle Walker; UW Madison, Waisman Center
Margarita Kaushanskaya; UW Madison, Waisman Center

Late-talking children, producing fewer than 50 words at 24 months, are at increased risk for long-term language difficulties. While language profiles and developmental trajectories for monolingual late talkers are well-delineated, less is known about bilingual children who are late to talk. This study explores language profiles of Spanish-English bilingual toddlers aged 2;0-2;11. The sample includes both typically developing and late-talking toddlers, identified via parent-report on the MB-CDI. Receptive and expressive vocabularies were measured via the Spanish-Bilingual ROWPVT-4 and EOWPVT-4, which measure bilingual children’s conceptual vocabulary (knowledge of concepts across both languages). Preliminary analyses indicate that bilingual late-talkers scored significantly lower than typically developing peers on vocabulary measures, with clinically low scores on the expressive vocabulary measure for some, but not all of the bilingual children. These preliminary findings suggest that standardized measures of bilingual vocabulary vary in sensitivity to late emergence of expressive language in bilingual late-talkers, pointing to the importance of considering the diversity of language experience in designating bilingual children as late to talk.

Funding Source: National Institutes of Health Grants RO1 DC020447.

---

PS1F47

Early Language and Behavior Predictors of Later Academic Performance

Hanyan Sun; Peabody College, Vanderbilt University
Andrew Chang; Peabody College, Vanderbilt University
Jason Chow; Peabody College, Vanderbilt University

Early language and social-emotional development form the foundation for children’s success in educational and social settings. Although language and behavior are bidirectionally related, school-based screening practices often rely on time-and cost-efficient tools with limited validation, potentially missing children with co-occurring or emerging delays. The longitudinal study examined whether kindergarten language and behavior screeners predicted Grade 3 reading and math achievement beyond baseline academic skills. A sample of 182 kindergarten students was followed through third grade. Standardized and teacher-developed language and behavior screeners were administered in kindergarten, and third-grade achievement was assessed using Virginia Standards of Learning (SOL) reading and mathematics scores. Hierarchical regression models controlled for demographics and baseline academic performance (KTEA-3 Brief). The standardized language screener significantly predicted both reading and math outcomes, explaining unique variance beyond initial skills. In contrast, teacher-developed language and behavior screeners were not significant predictors. Findings highlight the importance of psychometrically validated early screening within MTSS frameworks and suggest that informal tools may lack sensitivity for identifying children at risk for later academic difficulties.

Funding source: Institute of Education Sciences Grant #R324B200039

---

PS1F48

Multilingual Assessment Practices of School-Based Speech-Language Pathologists

Megan Buckley; Brigham Young University
Carolyn Ferguson Wood; Brigham Young University
Connie Summers; Brigham Young University

Previous research has demonstrated that school-based speech-language pathologists have increased their use of best practices in assessing multilingual children over time, while some barriers remain. The current student aimed to update this previous research and examine the role of speech-language pathologists’ language status (i.e., monolingual, bilingual, and multilingual) in make assessment decisions. Over 300 speech-language pathologists completed a survey about multilingual assessment practices. Almost half of the participants reported assessing multilingual children often or sometimes. Participants reported the lack of bilingual clinicians, the lack tools in languages other than English, the lack of their own knowledge of a child’s culture/language, and the lack of sufficient time as continued challenges in multilingual assessment. Participants are prepared and comfortable collaborating during the assessment process. The implications of currently reported assessment practices will be discussed as well as recommendations for future preparation.

PS2F01

Swedish-speaking children with DLD in comparison with TD and L2 learners: A new look at grammatical challenges

Christina Reuterskiöld; Linköping University
Marcus Nyström; Lund University
Joost van de Weijer; Lund University
Anna Eva Hallin; Karolinska Institute
Kristina Hansson; Lund University

Comprehension entails rapid analysis of different aspects of linguistic information. International studies show that children use the gender cues provided in the article of the noun phrase (NP) to retrieve nouns (Spanish, Dutch, German). Little is known about how Swedish-speaking children with DLD or L2 learners use the gender cue in the Swedish NP. We investigate children’s processing of the Swedish NP which has two genders (uter and neuter), in sentences with varying grammatical complexity and using eye tracking. We also study how processing is linked to vocabulary skills and production of the NP. Participants are children with DLD, children with TD and L2 learners. Two measures are generated from eye tracking: Anticipatory use of gender cues (time to target) and facilitative use of gender cues (proportion looks to target). Preliminary results show different patterns of processing, but vocabulary skills appear important for the ability to produce the Swedish NP. Results are discussed considering international research and in terms of strategies strengthening NP processing and production.

Funding: Riksbankens Jubileumsfond for the Advancement of the Humanities and Social Sciences.

---

PS2F02

Temporal Trends in Standardized Language Scores in Developmental Language Disorder: A Meta-Analysis

Arinjoy Bhattacharjee; University of Connecticut
Tammie Spaulding; University of Connecticut

The Flynn Effect, documented in norm-referenced tests of verbal and non-verbal intelligence, refers to the rise in test scores over time. The need for updated norms to counteract the Flynn Effect is one reason new versions of tests are published. Similar to intelligence tests, publishing new versions of standardized, norm-referenced language tests is also common practice. Frequently, clinicians are instructed to use the newer versions of language tests in their assessment with the assumption that the normative data in the prior versions are outdated. If the Flynn Effect is present on language tests, this would support the assumption of outdated norms, as failing to update norms would result in systematic score inflation, ultimately impacting score validity. However, whether the Flynn Effect is apparent on such tests of language is unknown. The current study investigates if there is a Flynn Effect in vocabulary assessments, with a focus on children with Developmental Language Disorder (DLD). Because such language tests are commonly administered to this population in research as well as clinical practice, test score validity is an important concern.

---

PS2F03

Using wearable fNIRS and new processing algorithms to examine within participant learning trajectories over time: A tutorial

Sarah Crow; The University of Texas at Dallas
Susmi Sharma; The University of Texas at Dallas
Janie Lewicki; The University of Texas at Dallas
Lauren Tigner; The University of Texas at Dallas
Paulina Skolasinska; McGill University
Ari Segel; Washington University
Emma Speh; Washington University
Adam Eggebrecht; Washington University
Julia Evans; The University of Texas at Dallas

Documenting change is fundamental to understanding the process of intervention among individuals with language disorders. We demonstrate the clinical applicability of wearable fNIRS systems and the recently developed NeuroDOT processing pipelines for examining within-person cortical dynamics of learning. Using a microgenetic research design with dense data sampling, we examined changes in the prefrontal cortical hemodynamic response in an adult female participant who completed identical spoken sentence repetition and auditory fixation tasks across eight days. In addition to behavioral accuracy, hemodynamic data were collected with a fNIRS system (NIRSport2) using a prefrontal optode montage and analyzed using NeuroDOT. Channel-wise analyses revealed significant differences in ?HbO2 oxyhemoglobin concentration across right and left prefrontal channels over days for the Sentence Repetition task, only. Brain maps further revealed qualitative differences across the days. Behavioral assessments do not fully capture what occurs during speech repetition tasks, and leveraging neuroimaging can help identify and discriminate between disordered and neurotypical populations.

Funding was provided by the NIDCD K18 DC021149 (PI Evans) and the University of Texas Dallas Friends of Brain Health Grant (PI Skolasinska).

---

PS2F04

Coaching Teachers to Increase Linguistic Accessibility in Content Assessments

Melanie Randall; The University of Arizona
Mary Alt; The University of Arizona
Melissa White; The University of Arizona
Sarah Cretcher; The University of Arizona

Teachers typically do not understand the characteristics of students with developmental language disorders (DLD). This limited understanding results in challenges in creating assessments that are linguistically accessible to students with DLD or multilingual students. We investigated whether we could coach secondary teachers to write linguistically accessible assessments that are more reflective of content mastery than language skill.

Five secondary teachers participated in four training sessions to learn about linguistic accessibility. Prior to training, each teacher submitted sample tests that were given a linguistic accessibility quotient based on 4 morphosyntactic/semantic elements: active voice, negation, question length, and vocabulary. After training, teachers modified test questions and submitted revised assessments.

Using paired-sample Bayesian t-tests, we compared linguistic accessibility quotients of tests pre- v. post-training. We describe which elements were most amenable to training.

A replicable program for coaching teachers to write linguistically accessible test questions has the potential to allow disabled or multilingual students to better show their content knowledge.

The first author is supported through OSEP Grant H325D230067 and a research grant from the University of Arizona Graduate and Professional Student Council

---

PS2F05

Calibrating a Developmental Language Disorder (DLD) Screening Task for Adults

Gerard Poll; Miami University
William Boone; Miami University
Carol Miller; The Pennsylvania State University

Developmental language disorder (DLD) affects individuals into adulthood, but screening tasks for DLD in adults are lacking. Sentence repetition tasks (SRTs) are a promising method of screening for DLD in children, but existing studies lack a systematic way of selecting sentences that appropriately challenge adults. Rasch sample-item targeting is a method to identify test items that maximize precision and discrimination of ability. Fifty-one adults, 26 at risk for DLD, repeated 67 sentences with varied lengths and structures. One set of 16 well-targeted sentences was compared to 16 sentences with a wide range of item difficulties. The targeted set resulted in a more precise (smaller standard error) measure of adults’ language ability than the wide-range set. The targeted set resulted in a higher AUC, a measure of classification accuracy. The analysis also suggested that additional items are needed for the item bank for an even more precise measure of ability. Rasch sample-item targeting is a promising approach for systematically identifying items for an SRT that is well calibrated to adults.

---

PS2F06

A Pilot Study Examining Morphosyntax Learning in Children who are Deaf/Hard-of-Hearing Using an Artificial Grammar Learning Task

Heidi Mettler; University of Illinois Chicago
Luis Villalpando; University of Illinois Chicago
Tina Grieco-Calub; University of Illinois Chicago

Children who are deaf or hard-of-hearing (DHH) who use spoken language are at risk of morphosyntactic delays, even with early intervention and optimized hearing technology. These delays may relate to cognitive weaknesses often observed in DHH children, such as in working memory. Scaffolding linguistic input to accommodate working memory weaknesses may facilitate morphosyntactic learning in this population. We are piloting an auditory artificial grammar learning task modified to reflect a within-subjects design. We aim to run eight DHH children as pilot participants. Children will hear morphosyntactic-like structures in short (3 syllables) and long (5 syllables) conditions and complete grammaticality judgments. The dependent variable is percent accuracy. Data collection has not yet begun, but we predict that children will show above-chance level performance, with a pattern of higher accuracy in the short condition. This pilot study will provide proof-of-concept of our task, which will enable us to use it with a larger sample to assess how stimulus length affects morphosyntax learning at a mechanistic level in DHH children.

This project is not currently funded.

---

PS2F07

Why Initial Conditions and Heritage Language Exposure Matter: Morphosyntax Development in Spanish–English Bilingual Children

Joseph Hin Yan Lam; University of California, Irvine
Jissel Anaya; Ohio State University
Danyang Wang; Towson University
Jiali Wang; Texas A&M University
Lisa Bedore; Temple University
Elizabeth Peña; University of California, Irvine

This study examined how early bilingual morphosyntax performance and language exposure predict the development of Spanish-English bilingual morphosyntax during middle childhood. We examined (1) the longitudinal structure and invariance of Spanish and English morphosyntax clusters, (2) cross-linguistic effects of initial conditions on growth, and (3) the role of English exposure over time. Participants included 416 unique Spanish–English school-age bilinguals from seven- to eleven-year-olds who completed a standardized bilingual morphosyntax cloze task in both languages. A subset of 135 participants repeated the task annually. Measurement and growth models showed acceptable fit and metric invariance for Spanish and English morphosyntax clusters. Parallel-process growth models revealed that higher initial English morphosyntax performance significantly predicted lower Spanish growth, but not vice versa. Time-varying exposure effects were weak and mostly nonsignificant, with a small negative effect of English exposure on Spanish morphosyntax at age nine. Findings suggest the importance of initial conditions in bilingual language development during middle childhood.

Funding: R01DC010366 and 1R21DC011126-01 (Peña)

---

PS2F08

A Framework Guided Matrix Analysis of System Level Supports for Comprehensive Evaluations

Emma Barnett; University of Wisconsin-Milwaukee
John Heilmann; University of Wisconsin-Milwaukee
Alyssa Wojtyna; University of Wisconsin-Milwaukee
Dawn Merth-Johnson; Wisconsin Department of Public Instruction

Comprehensive language evaluations are a best practice in speech-language pathology, yet many clinicians report persistent implementation barriers. Implementation research offers methods to understand system-level determinants that influence how clinicians adopt evidence-based practices. We used a Framework-Guided Matrix Analysis of implementation artifacts from 25 district leaders who supported clinicians across six districts.  These continuous improvement teams completed four meetings to identify barriers and develop strategies to support comprehensive evaluations in their districts. Our qualitative analysis categorized strategies according to the Active Implementation Framework’s competency, organization, and leadership drivers. Early strategies focused on structural supports, such as ensuring access to appropriate assessment materials. Over time, districts shifted toward organizational supports that refined workflows and integrated processes. Districts varied their prioritized strategies and pace, reflecting the influence of local context. Findings highlight that successful translation of research to practice requires robust, context-responsive system-level supports. We will discuss how implementation frameworks can help discover system-level insights needed to reduce the research to practice gap for children with language disorders.

Funding: Wisconsin Department of Public Instruction (with funds from the US Department of Education).

---

PS2F09

The importance of Task Context in Multilingual Language Samples

Mariana Gomez Becerra; University of California, Irvine
Joseph Hin Yan Lam; University of California, Irvine
Stephanie Reich; University of California, Irvine
Natasha Cabrera; University of Maryland

Parent-child interaction samples provide ecological and valid insights into parental language input and children’s expressive language, capturing variability that may not emerge in standardized assessments. Language performance is highly context-dependent, yet little is known about how activity type shapes linguistic productivity and complexity in multilingual families. This study examined task-related differences in multilingual parent-child interactions across four activities: book reading, play with no toys, play with toys and cleanup. Data were drawn from 40 parent-child dyads participating in a longitudinal intervention study, with interactions recorded when children were 30 months old. Transcripts were completed using SALT and verified by trained research assistants. Measures of linguistic productivity and complexity (TNU, TNW, NDW, MLUw, and syntactic index) will be reported. Exploratory factor analyses and within-participant ANOVAs will be conducted, controlling for interaction duration. We anticipate that shared book reading will elicit greater linguistic productivity and morphosyntactic complexity than other tasks. Findings will inform clinical interpretation of language samples and highlight the importance of task selection in multilingual assessment.

Funding source(s): DOED #H325D220045- PI: Elizabeth Peña; NICHD #R01HD078547 PIs: Natasha Cabrera and Stephanie Reich; NICHD #1R03HD110672 PI: Stephanie Reich

---

PS2F10

Validity and Reliability of a Novel Morpheme Learning Task for Use with Multilingual Children

Lizbeth Finestack; University of Minnesota – Twin Cities
Yue Zhao; University of Minnesota – Twin Cities
Nidhi Kohli; University of Minnesota – Twin Cities
Kerry Ebert; University of Minnesota – Twin Cities

One recommended approach for assessing multilingual children is dynamic assessment. While many variations on dynamic assessment exist, with no single validated approach. We examined the feasibility of a novel morpheme learning task when assessing the language abilities of 4- to 6-year old multilingual children. Children completed a researcher-designed novel morpheme task twice within a 1- to 2-month period, along with two narrative dynamic assessments. Results indicated that the English-based assessment was difficult for most children. Evaluations of both test-retest reliability and concurrent validity revealed relatively weak psychometric properties. We will continue to explore analyses that consider specific child characteristics, including level of English proficiency. Our findings suggest that further development and evaluation of alternative assessment approaches for multilingual children are critical to aid in the identification of language impairment in this population.

Funding: NIDCDR01DC019895.

---

PS2F11

Clinician-client relationships in a grammatical language intervention study for DLD

Kerry Ebert; University of Minnesota – Twin Cities
Olivia Matthys; University of Minnesota – Twin Cities
Thomas Murray; University of Minnesota – Twin Cities
Lizbeth Finestack; University of Minnesota – Twin Cities

Rationale: Clinician-client relationships play an important role in children’ s speech-language intervention, but have rarely been measured in controlled intervention studies. We examine ratings of the clinician-client relationship by 5- to 9-year old children with developmental language disorder (DLD) within a randomized controlled trial of a grammatical intervention. More specifically, we consider: whether children with DLD can validly assess the clinician-client relationship, child characteristics associated with stronger relationship ratings, and whether relationship ratings can predict language intervention outcomes.

Method: Children with DLD completed four intervention sessions and then rated their working relationship with their clinician. The rating measure included subscales for the emotional bond and for collaborative work (i.e., on goals and tasks in intervention). Clinicians completed a parallel rating.

Results: Children’s ratings on the work subscale correlated with clinician ratings of both work and emotional bond. Older children provided higher relationship ratings, while clinicians rated the work subscale higher for children with stronger language skills.

Conclusions: We demonstrate that children with DLD can validly rate their relationship with a clinician, and consider impact on intervention outcomes.

Funding: NIDCDR01DC019374.

---

PS2F12

Priming for Tense and Aspect Markers in Vietnamese Monolinguals with and without DLD

Ngoc Do; San Diego State University, University of California San Diego
Giang Pham; San Diego State University

Tense and aspect markers are a hallmark weakness for children with developmental language disorder (DLD). To our knowledge, the present study is the first to examine children’s tense/aspect production in Vietnamese, the most spoken language in the Austroasiatic family and yet understudied in DLD research. We created a structural priming task that paired four key Vietnamese tense/aspect markers (dang–progressive, dã–past, s?–future, xong–telic) with four verb types (activity, state, accomplishment, achievement) to test production of 196 monolingual children aged 4–6, of whom 28 were classified as having DLD. We compared the tense/aspect production of children with DLD to that of typically developing peers and tested the impact of verb type on tense/aspect production to evaluate the Aspect Hypothesis. Logistic mixed-effects models revealed main effects for group and tense/aspect, and a group*tense/aspect interaction. The DLD group performed lower than the typically developing group across tense/aspect markers, with group differences varying by specific markers. Verb type impacted production of the dã–past marker only. Findings contribute to efforts to identify and characterize DLD in Vietnamese-speaking children.

Funding is from NIH R01DC019335.

---

PS2F13

Early Expressive Language Milestones as Predictors of Later Language Outcomes in Spanish–English Bilingual Children

Hana Dussan; University of Wisconsin-Madison
Chenelle Walker; University of Wisconsin-Madison
Margarita Kaushanskaya; University of Wisconsin-Madison

Bilingual children with DLD are at increased risk of misdiagnosis or of not receiving early intervention services (Norbury et al., 2016), and it is unknown whether early language milestones are predictive of DLD in bilingual children. The present study examines whether early expressive language milestones predict later low language status in Spanish-English bilingual children. Thirty bilingual children aged 4 to 6 years (15 DLD, 15 TD) were selected from an existing database and individually matched on chronological age and proportion of Spanish language exposure (Peña et al., 2014). Retrospective parent-reported ages of production of first words, two-word combinations, and sentences were used as predictors of language outcomes, with non-verbal IQ and maternal education also considered. Language outcomes were determined using the Bilingual English Spanish Assessment (BESA). Preliminary between-group comparisons indicated significant differences between groups on nonverbal IQ and maternal education, but comparable parent-reported expressive milestones, in both English and Spanish. Ultimately, this study will inform early risk identification practices and support clinicians in monitoring and referral decisions for bilingual children.

---

PS2F14

Dual-Language Input Structure and Language Ability Jointly Shape Bilingual Children’s Word and Semantic-Feature Learning

Anna Duncan; Boston University
Kimberly Crespo; Boston University

Children’s bilingual books integrate two languages and are commonly used to support word learning across languages. Prior research suggests that bilingual input can facilitate word learning, with some evidence that mixed-language formats may confer advantages over single-language formats. However, this work has focused almost exclusively on acquisition of labels, leaving open questions about how bilingual formats supports learning of semantic features and listening comprehension. Therefore, the present study examined the effect of bilingual storybook format on the learning of novel words, their associated semantic features, and narrative comprehension in children with a range of language abilities. Spanish-English bilingual children aged 4-6 years listened to one story where the languages were separated, and a different story where the languages were intermixed. Preliminary findings suggest language mixing may support word learning and listening comprehension, while language ability may play a larger role in how children build an understanding of what words mean beyond their labels.

This research was supported by NIDCD K23 DC022006 awarded to Kimberly Crespo and T32 DC013017 predoctoral fellowship awarded to Anna Duncan.

---

PS2F15

Determining language deficits in school-aged children with and without speech sound disorder: Exploring narrative analysis using elements of microstructure and macrostructure

Abigail McBournie; Auburn University
Allison Plumb; Auburn University
Anna Ehrhorn; Auburn University

Speech sound disorder (SSD) often co-occurs with developmental language disorder (DLD). However, language assessment methods may not detect this co-occurrence. Emerging evidence suggests that some microstructure and macrostructure elements may detect DLD while minimizing impact of speech errors; however, no known research has determined if a narrative composite of these elements may better identify DLD in SSD. The present study aims to explore what narrative analysis methods are sensitive and specific in identifying DLD in school-aged children while minimizing the effects of SSD. Sixty previously collected narrative samples will be transcribed and analyzed for four microstructure and four macrostructure elements, and scored using the novel Analysis of Narrative Development for Speech Sound Disorder (ANDS) Tool (i.e., a narrative composite score). Planned analyses will determine what elements of microstructure and macrostructure will differ between SSD and TS when controlling DLD, and whether ANDS may be more sensitive and specific for this population. Findings could improve language diagnostic practices for school-aged children with SSD to ensure intervention targets all deficits prior to service dismissal.

No financial support was received for this work.

---

PS2F16

Open-Science Practices Among Child Language Researchers: An Empirical Analysis

Amanda Eiser Hess; University of Virginia
Anne Reed; Texas State University
Andrea Ford; University of Cincinnati

This presentation will present the results of a secondary analysis of child language researchers’ engagement in open-science practices. Although overall engagement in open science remains fairly low in communication sciences and disorders (CSD; Eiser Hess et al., 2026; El Amin et al., 2023; Pfeiffer et al., 2025), differences across subdisciplines have yet to be explored. Drawing on data from a previous review, this secondary analysis examines three open-science practices–preregistration, open data, open materials–among empirical research articles published in the area of child language across six years (2019-2024) in Language, Speech, and Hearing Services in Schools (LSHSS) and Communication Disorders Quarterly (CDQ). We will compare the prevalence of these practices in child language research to trends across the broader CSD field. In addition to quantitative prevalence data, we will provide qualitative descriptions of how child language researchers implement these practices (e.g., where studies are preregistered, what types of materials are shared).

The work was supported by the Institute of Education Sciences (R305B200020) and the Office of Special Education Programs, Office of Special Education and Rehabilitative Services (H325 D240052).

---

PS2F17

Consonant blends: How does oral segmentation relate to written spelling proficiency?

Olivia Driscoll; Vanderbilt University
Jane Eppstein; Vanderbilt University
Hannah Krimm; University of Georgia
Ashleigh Hayward; Vanderbilt University
C. Melanie Schuele; Vanderbilt University

Children with deficits in phonological awareness (PA) struggle with phoneme segmentation. Words with consonant blends (CCVC/ CVCC) are more challenging to segment than CVC words, and children with PA deficits often require focused intervention on segmenting CCVC, CVCC words (e.g., skit, desk). For children with letter sound knowledge, written spelling is a way to capture segmentation skills. This study explores the relation of oral segmentation and written spelling of words with consonant blends in children who are not yet proficient in segmentation.

Kindergarten and first graders (n = 48) completed two tasks with a set of 10 CVC and 10 CCVC/ CVCC words: oral segmentation and written spelling. Ongoing analysis includes comparing proficiency of oral segmentation and written spelling, describing oral segmentation error types, and calculating spelling accuracy for error types.

Findings inform instructors of which oral segmentation errors may require focused intervention. Findings guide the use of written spelling measures to efficiently monitor progress in PA skill.

Funding: Project REACH (H325D220072; U.S. Department of Education) and the National Institute on Deafness and Other Communication Disorders (R01DC021188).

---

PS2F18

Spanish-English bilingual children’s grammatical productivity in verbs, articles, and direct object clitics

Alicia Escobedo; CSU San Bernardino
John Gallagher; SDSU/UCSD
Irina Potapova; SDSU
Sonja Pruitt-Lord; SDSU

Grammatical productivity has been shown to differentiate between typically developing children and their peers with DLD in monolingual English-speaking populations (Gladfelter & Leonard, 2013). The current study investigates the grammatical productivity of Spanish-English bilingual children in Spanish by examining children’s productions of articles, direct object clitics, and verbs. These grammatical features have been previously identified as potential clinical markers in Spanish-speaking children (Castilla-Earls et al., 2022). Participants in the current study were preschool-aged bilingual children who completed Spanish language samples. Preliminary results of this study suggest that typically developing children demonstrated significantly higher productivity in Spanish when compared to their peers with DLD. Results of this study will provide more information on how language sample measures like grammatical productivity may contribute to bilingual language assessment.

Funding: NIH NICHD F31HD111303.

---

PS2F19

Measuring the Individualized Teacher Language Input to Preschoolers with Emotional and Behavioral Challenges

Eric Eubank; Vanderbilt University
Jason Chow; Vanderbilt University
Alexandra Davidson; Vanderbilt University
Mackenzie Ringo; Vanderbilt University

Children with emotional and behavioral challenges face heightened risk for language delays, yet individualized classroom language experiences remain largely unexamined. Most research relies on global measures, averaging teacher language input across children, potentially masking variability. This study pilots a novel observational measure of differential attention to quantify how teachers distribute language-supportive talk to individual preschoolers. Participants included 20 teachers and 29 focal children with elevated externalizing behaviors. We coded twenty classroom videos, utterance-by-utterance, to identify the target of teacher-talk and whether utterances were language-supportive (e.g., contingent responses, expansions, open-ended questions; inter-rater agreement was high [93.6%]). Across classrooms, 19% of teacher utterances were coded as language supportive. Focal children received fewer language-supportive utterances (M = 8 individually directed; 10.3 including group-directed) than non-focal peers (M = 22.5). Generally, instances of language-supportive input for children with challenging behavior were inflated through classroom-level averages. Findings (1) demonstrate the feasibility of measuring differential attention in classroom contexts, and (2) highlight the importance of child-level analyses in understanding individual experiences of language-learning opportunities in inclusive preschool settings.

Funding: Institute of Education Sciences (R305A140487)

---

PS2F20

Characteristics Influencing Bilingual Educators’ Decisions and Speech-Language Pathologists ‘ Judgments during the Pre-Referral Evaluation Process for Bilingual Learners

Janelle Flores; University of Houston
Anny Castilla-Earls; University of Texas at Dallas

Bilingual learners’ disproportionate representation in speech-language services persists, with limited research exploring the underlying reasons. Given the pivotal roles of Bilingual educators (BEs) and speech-language pathologists (SLPs) in the pre-referral and evaluation of bilingual learners, this study explored the influence of personal characteristics (Professional Cultural Efficacy [PCE] and experience) and child language characteristics (vignette profiles). Participants (31 BEs and 31 SLPs) identified the vignette’s language category (atypical, borderline, typical), indicated their prereferral decision or evaluation plan, and completed a questionnaire assessing PCE and experience. Linear Mixed Models revealed that vignette profiles significantly influenced BEs’ decisions and SLPs’ judgments. BEs and SLPs systematically assigned a higher likelihood of referral or evaluation for atypical profiles and a lower likelihood for typical profiles, reflecting case-sensitivity. Interclass Correlation Coefficient (ICC) analysis indicated moderate agreement among BEs (ICC = 0.64) and fair agreement among SLPs (ICC = 0.47). SLPs had higher accuracy on language categories, except for one atypical vignette profile. Findings underscore the importance of child language characteristics over personal characteristics in decision-making and clinical judgments.

Partially funded by the Texas Speech-Language-Hearing Foundation

---

PS2F21

How word characteristics relate to learning and long-term retention for kindergarten children with DLD

Katherine Gordon; Boys Town
Stephanie Lowry; Boys Town
Justin Kueser; Boys Town

Many word-learning interventions include a consistent dosage across words. However, children with DLD vary across words in the total dosage needed to learn word forms and meanings. In the current study, we conducted a secondary data analysis of an interactive book reading intervention for kindergarten children with DLD. We assessed how the characteristics of concreteness, frequency, and emotional valence related to children’s ability to learn the words and retain the words over a 12-week post-training delay. We found that concreteness and frequency positively related to the learning and retention of word forms. None of the factors significantly related to the learning and retention of word meanings, as assessed via a definition task. Currently, we are coding children’s learning and retention of each semantic feature of each definition to determine the features of definitions that were easier or more difficult to learn and retain. Overall, understanding the characteristics of words, and semantic features of word meanings, that are more difficult for children with DLD to learn will inform clinical word-learning interventions.

NIH-NIDCD, R01DC01282.

---

PS2F22

Benchmarking Developmental Change in Lexical Overlap During Caregiver-Child Conversations

Emily Harrington; University of Illinois Urbana-Champaign
Bohyang Jin; University of Illinois Urbana-Champaign

Lexical overlap is a common linguistic phenomenon that may play a special role in early caregiver–child conversations. This study examined lexical overlap in children with typical language development, asking whether specific caregiver utterances increased the likelihood of child lexical overlap across development. Data were obtained from the Champaign corpus in CHILDES (44 dyads sampled at 21, 24, 27, 30, 33, and 36 months). Child lexical overlap and caregiver imitation and expansion were coded using automated CHIP and CHIPUTIL programs in CLAN. A mixed-effects logistic regression model estimated the likelihood of child lexical overlap as a function of caregiver utterance type (imitation/expansion vs. “other”), timepoint, and their interaction. Across development, children were more likely to respond with lexical overlap following caregiver imitation or expansion. However, lexical overlap following “other” caregiver utterances increased with age. This pattern suggests a developmental shift in which children initially rely on caregiver-supported contexts but, with increasing language and conversational ability, respond with lexical overlap to a broader range of caregiver utterances. Findings provide a benchmark for future work with children with language disorders.

Funding: N/A

---

PS2F23

Interlocutor Effects on Code-Switching Patterns in Bilingual Children With and Without Developmental Language Disorder

Michelle Hernandez; University of Houston
Anny Castilla-Earls; University of Texas at Dallas

Bilingual children frequently code-switch (CS), alternating between two languages within a conversation. While typically developing bilingual children adjust their language use based on their conversational partner, less is known about whether children with developmental language disorder (DLD) show similar accommodation patterns or how language proficiency shapes CS  behavior. CS in children with DLD may often be misinterpreted as language confusion or language disorder rather than recognized as a strategic communicative strategy. This study examined whether interlocutor language context influences code-switching frequency and type selection, and whether relative proficiency predicts these patterns in children with and without DLD. Thirty-six Spanish–English bilingual children (16 DLD, 20 TD), ages 5;0–6;11, completed a collaborative task across three interlocutor conditions: English-only, Spanish-only, and bilingual code-switching. Generalized linear mixed-effects models showed that children accommodated to interlocutor language use by increasing code-switching in the bilingual condition. Relative proficiency significantly predicted switching frequency, with stronger effects in single-language contexts. Code-switching patterns were better explained by proficiency than diagnostic status, challenging deficit-based interpretations and highlighting the need for proficiency-informed assessment practices.

This project was supported by an F31 fellowship from the NIDCD (F31DC021619) awarded to Michelle Hernandez.

---

PS2F24

Multi-Modal Communication During Reading and Play Between Mothers and Young Children Learning to Use High-Tech Augmentative and Alternative Communication

Fiona Higgins; Vanderbilt University
Elizabeth Biggs; Vanderbilt University

Children with complex communication needs and their parents use many modalities to communicate. These modalities include aided augmentative and alternative communication (AAC) as well as gestures, signs, vocalizations, and speech. Parents frequently are recommended to model AAC to support their child’s AAC language learning; however, they often report that modeling is difficult due to their limited skill using AAC. Examining language samples from two common activities—reading and play—can illustrate what parents do naturally at home as certain activities may better lend themselves to AAC modeling than others. Here, we compare reading and play activities and examine (a) child AAC use frequency, (b) mother AAC use frequency, and (c) patterns in AAC use including spontaneity and AAC type (e.g., speech generating device, sign, picture symbol, with or without speech). Eleven children and their mothers provided language samples. Children were young (M = 4;6, range 3-6 years) with varied developmental disabilities (autism, Down syndrome, developmental delay, apraxia, genetic syndrome), and all had access to a speech generating device at home.

Funding: ED/OSEP H325D230072

---

PS2F25

Prosodic features of child-directed reading in bilingual caregivers of children with and without risk for speech-language difficulties

Kelsey Davison; Purdue University, Boston University
Juliana Ronderos; Teachers College Columbia University, Boston University
Megan Gross; University of Massachusetts Amherst
Mara Breen; Mt. Holyoke College
Jennifer Zuk; Boston University

Caregiver-child shared reading is known to support monolingual and bilingual children’s language development. In monolingual families, caregivers’ intonation and pausing while reading (i.e., oral reading prosody) is linked to children’s storybook listening comprehension. However, little is known about oral reading prosody in bilingual shared reading contexts, where caregivers’ language dominance and reading practices vary widely. This study examined factors influencing Spanish-English bilingual caregivers’ prosody during shared reading interactions. Twenty-six caregivers of preschoolers, with and without speech-language risk, presented one English and one Spanish book to their child. Utterances were categorized as direct reading, translation, or image-based storytelling. Utterance-level measures of intonation range and pauses (grammatical, ungrammatical) were extracted from audio recordings. Linear mixed-effects models revealed that caregiver language dominance and storybook presentation type associated with caregiver prosody, with intonation range differing between direct reading and translation. Associations varied based on the language of the book text. Further, caregivers of at-risk children had fewer grammatical pauses. Findings highlight the importance of considering caregiver language profiles when supporting bilingual shared reading.

This work was funded by an ASHFoundation New Investigators Research Grant.

---

PS2F26

How We Score Matters: Relations of Oral Narrative Retell Scoring Approaches to Cognitive and Language Skills in Grade 2

Molly Leachman; University of Michigan
Young-Suk Grace Kim; University of California, Irvine

This study examined relations among three oral narrative retell scoring approaches (quality rubric, total number of words, keyword scoring) and their associations with language and cognitive skills in second grade students (N = 529). Using the Test of Narrative Language, we compared scoring methods and investigated their relations to vocabulary, grammatical knowledge, inference-making, perspective-taking, comprehension monitoring, working memory, and attentional control. Structural equation modeling demonstrated excellent model fit across all approaches, with most showing similar patterns of relations to language and cognitive skills. Notably, the keyword approach uniquely showed significant direct relations between vocabulary and grammatical knowledge to retell performance, likely reflecting its requirement for precise lexical retrieval and syntactic accuracy. Complementary patterns emerged for students receiving speech and language services (n = 80), but with notable decreases in relations with higher-order skills. Findings suggest that oral narrative retell provides robust measurement of children’s narrative skills across scoring approaches, while also highlighting that keyword scoring may be particularly sensitive to foundational language abilities relevant for identifying language disorders.

Funding sources: R305A130131, R305A180055, P50HD052120, H325D220045

---

PS2F27

Beyond Mean Length of Utterance: Employing Range and Variability Measures of Utterance Length to Characterize Linguistic Alignment between Autistic Children and their Caregivers

Janine Mathee-Scott; Marquette University
Grace Corrigan; Michigan State University
Courtney Venker; Michigan State University

Research suggests that caregivers of children with typical language and those with language delays fine-tune the complexity of their language input to their children’s language level. The alignment of linguistic complexity between autistic children and their caregivers has also been explored, but several open questions remain. This study examines the utility of newly introduced measures of utterance length variability: range (RLU) and standard deviation (SDLU), for examining how autistic children and their caregivers align the complexity of their language to one another. These measures, derived from language sample analysis, appear to be useful for characterizing linguistic complexity alignment between autistic children (n=40, Mage=41.78 months) and their caregivers. Findings suggest that autistic children’s linguistic complexity measures (MLU, RLU, and SDLU) significantly predict caregivers’ corresponding measures (p’s< .001) and that autistic children’s broader receptive and expressive language (as measured by the Preschool Language Scales-Fifth Edition) also significantly predict caregivers’ linguistic complexity (p’s< .01). This suggests that caregivers and their autistic children align the complexity of their input to one another both in real time, and according to the child’s broader language abilities.

Funding sources: NIH R01DC020165; R21DC016102

---

PS2F28

Investigating Narrative Skills using responses from questions: An insight into Macro and Micro structures of Language and Working Memory

Diya Nair; University of Western Ontario
Lisa Archibald; University of Western Ontario

Narrative abilities refer to the telling of a story or an event. Despite the seeming simplicity of the task, narrative skills require complex language skills. Corroborating evidence for the role of complex language in narratives comes from findings that better narratives are produced by children performing higher on language tests (Paris and Paris, 2003; Silva et al., 2014). In narratives, story elements must be organized in a coherent way (known as macrostructure) in order to make sense of the story told and using sophisticated linguistics devices to improve the clarity of the story (known as microstructure) such as conjunctions, mental verbs and clearly referenced pronouns. Questions focused on macrostructure asked prior to narrative production have been reported to result in increases in macro- but not microstructure elements in 4-6 year olds (Silva and Cain, 2019). In contrast, Nair et al. (2024) reported that questions, irrespective of focus (macro-or microstructure), led to an increase in microstructure elements. The current study examined the relationship between macro- or microstructure question response accuracy, language skills, working memory abilities and children’s quality of narratives.

---

PS2F29

Evaluating the Psychometric Properties of the DELV-Screening Test for School-Age Language Minorities

Julia Benoit; University of Houston
Monique Mills; University of Houston

Rationale

The purpose of the current study was to identify the factor structure of the Diagnostic Evaluation of Language Variation-Screening Test (DELV-ST, Seymour et al., 2022) Part II, using exploratory factor analysis and confirmatory factor analysis.

Methods

A total of 142 children, ages 6-12 years, were included in this study from Ohio (n = 103) and Texas (n = 41). Children completed a battery of tests of hearing, cognition, language, vocabulary, narration. The current study focuses on performance of the DELV-ST Part II.

Results

Exploratory factor analysis indicated four factors with eigenvalues greater than 1.0. Confirmatory factor analysis indicated a multidimensional structure (more than one underlying factor) of the DELV-ST Part II, demonstrated by the improved fit when adding a second factor. However, when adding a further factors, there is minimal gain when balanced by fit, parsimony and interpretability.

Conclusions

Our study of children in Midwest and Southwest regions of the United States replicates that of two other studies which examined children in the Southeast region of the United States. It appears as if the construct validity of DELV-ST, Part II is stable.

NIH funded this work.

---

PS2F30

Preserved Category Representations in Children with Developmental Language Disorder Despite Semantic Deficits

Minh Bui; Speech, Language, and Hearing Sciences, Purdue University
Pat Deevy; Speech, Language, and Hearing Sciences, Purdue University
Laurence Leonard; Speech, Language, and Hearing Sciences, Purdue University
Arielle Borovsky; Speech, Language, and Hearing Sciences, Purdue University

Semantic depth is crucial for differentiating concepts into categories. Despite well-documented semantic depth deficits in children with developmental language disorder (DLD), there is conflicting evidence on whether this group exhibits differences in category representations. Here, we predicted that children with DLD could differentiate between subgroups (e.g., mammals vs birds) within a single domain (e.g., animals), but they would show this distinction less strongly than age-matched typically developing peers. Compared to across-domain distinctions (e.g., animals vs plants), within-domain distinctions (mammals vs birds) appear later in development and have not been directly assessed in DLD. Thirteen children with DLD and 14 age-matched typically developing children gave graded relatedness judgments of items within and across subgroups in a domain. Following our hypothesis, children with DLD successfully captured within-domain subgroups. Contrary to our predictions, both groups of children show equally distinct representations of within-domain subcategories. Despite presumed semantic deficits, the DLD group successfully leveraged relatively few featural differences among concepts (that have highly covarying features) to differentiate them into meaningful categories.

---

PS2F31

What’s going on in there? Literacy intervention with AAC users.

Isabella Maranghi; University of Tennessee Health Science Center
Jillian McCarthy; University of Tennessee Health Science Center
Erinn Finke; Calvin University

Speech-language pathologists (SLPs) play a critical role in supporting literacy development, yet the implementation of evidence-based practices (EBP) remains inconsistent, particularly for children who use augmentative and alternative communication (AAC). Results of this analysis examined how SLPs describe their literacy intervention practices for AAC users to better understand current approaches and alignment with EBP.

One hundred eleven ASHA-certified SLPs or clinical fellows serving children ages 3–21 in school or outpatient settings participated in a survey related to literacy practices. Thematic analysis of open-ended responses revealed that many SLPs rely on indirect or non-explicit literacy activities, such as sight-word exposure. While some reported components of EBP, phonological awareness instruction was often limited to early skills, with few describing advanced decoding or encoding strategies.

Findings highlight a persistent research-to-practice gap and underscore the need for enhanced training, clearer guidelines, and targeted support to improve literacy outcomes for AAC users.

This project is funded through an OSEP leadership training grant (Project PAL; HD325D230072).

---

PS2F32

Assessing primed productivity in young preschoolers at risk for DLD

Tracy Preza; University of Illinois Urbana-Champaign
Windi Krok; Purdue University
Pamela Hadley; University of Illinois Urbana-Champaign

Identification of developmental language disorder before 4 years of age is challenging; therefore, there is a need to develop more sensitive measures for younger preschoolers. This study explored whether measures of subject diversity and functional morpheme productivity (e.g., articles, progressive -ing, auxiliary BE) when applied to a priming task could distinguish young preschoolers with ongoing risk for DLD from those with improved language abilities. Participants included 56, 3;6-year-old children who were identified with risk for DLD at 2;6. At 3;6, they were assigned to severe, moderate, mild, and improved grammatical ability groups based on standardized testing. Subject diversity and primed productivity scores computed from the Sentence Diversity Priming Task resulted in good classification accuracy. We will discuss the value of using a structured elicitation task and primed productivity measures for identifying 3-year-old children at risk for DLD. We will also discuss the importance of exploratory studies for distinguishing children with mild manifestations of DLD from typical peers.

Funding: NIDCD U01DC017135

---

PS2F33

“Tell Me What This Means in Real Words”: Caregiver and SLP Perspectives on Supporting Families of Children with DLD

Maura O’Fallon; University of Delaware
Victoria Coons; University of Delaware
Michelle Erskine; Towson University
Sarah Curtiss; University of Delaware
Amanda Owen Van Horne; University of Delaware

Parents of children with Developmental Language Disorder (DLD) report stress and uncertainty related to their child’s language needs. Simultaneously, speech-language pathologists (SLPs) face barriers to parent collaboration. We used an implementation science framework to understand how SLPs can support parents of children with DLD. We conducted focus groups with parents of children with DLD (n=25) to understand their lived experiences and support needs, and focus groups with SLPs (n=11) to understand perspectives on supporting parents and children with DLD. Using rapid qualitative analysis, we identified preliminary themes across groups. Caregivers and SLPs recognized the emotional toll of parenting a child with DLD, and the need for plain-language information. Although, caregivers prioritized emotional support while SLPs focused on sharing information. These mismatched foci likely hinder collaboration. Participants in both groups also cited low familiarity with “DLD.” These data suggest that SLPs must strengthen their capacity in supporting caregivers’ emotional needs, as well as their own knowledge of DLD as a diagnostic label. These data also underscore the need for parent-friendly communication around DLD.

Funding source: NIH/NIDCD, 1R21DC022878-01, PI: Owen Van Horne

---

PS2F34

Investigating Diagnostic Stability from Pre-K to Kindergarten Among Multilingual Students in an Urban School District

Kristen Victorino; Rutgers University; School of Health Professions
Celeste Domsch; Rutgers University; School of Health Professions

This retrospective chart review study examines stability in speech-language evaluation practices and results for culturally and linguistically diverse (CLD) preschool students in an urban school district. We compare results from T1 (~age 3) and T2 (~age 5) evaluations, including whether students continue to be eligible for speech-language services at T2 and if so, under what IEP classification; how multilingual (ML) status is considered in assessments at T2 vs T1; and whether variables such as race, ethnicity, ML status, language of instruction or intensity of intervention in preschool were predictive of change in test scores and/or continued eligibility at T2. Results will be discussed with a focus on closing the research-to-practice gap in schools and improving equitable speech-language evaluations and access to educational opportunities for CLD students.

This study was funded by a Rutgers Equity Alliance for Community Health Community Impact grant.

---

PS2F35

“I Don’t Know Whose Job It Is to Train Anyone in Behavior”: A Phenomenological Account of School-Based SLPs’ Experiences with Behavior Support

Igor Penedo Silva; Vanderbilt University
Michelle Lifson; Vanderbilt University
Zoe Hussey; Vanderbilt University
Phoebe Ahn; Vanderbilt University
Colleen Walsh; Vanderbilt University
Jason Chow; Vanderbilt University
Rachel Kasul; Vanderbilt University

Students with language disorders who receive school-based intervention often present with challenging behaviors, requiring speech-language pathologists (SLPs) to address both behavioral and communication needs to ensure effective service delivery. Students’ challenging behavior is highly prevalent in SLPs’ school practice; however, research suggests that most SLPs receive minimal formal training in behavior management in graduate school. This creates a gap between graduate SLP preparation and the realities of clinical practice. Addressing this training-to-practice gap is essential for preparing SLPs to meet both the behavioral and communication needs of their students. The current study explored 20 practicing school SLPs’ experiences with challenging child behavior. Through individual, semi-structured interviews on Zoom, participants described how these experiences influence service delivery, revealed limitations in their graduate training, and underscored the need for stronger professional support structures. Key findings highlight the real-world factors that prevent and facilitate effective behavior support within language services. Future research should evaluate what aspects of behavior management should be explicitly taught in SLP graduate programs.

This work was supported, in part, by the Office of Special Education Programs (H325D2300371; H325D220055) and the National Institute of Mental Health (T32-MH18921).

---

PS2F36

Comparing Book and Video Retell Tasks in Spanish Narrative Language Samples of Bilingual Preschoolers

Alice Regalado-Lee; Texas Christian University
Jean Rivera-Perez; Texas Christian University

Purpose: Language sample analysis (LSA) informs bilingual assessment, including identification of developmental language disorder (DLD), yet elicitation context may influence narrative productivity and microstructure. This preliminary study compared Spanish outcomes from a wordless picture-book retell and a nonverbal video retell in bilingual preschoolers.

Method: Ten Spanish-dominant bilingual preschoolers produced Spanish narratives in both contexts. Samples were transcribed and coded in SALT. Measures included productivity (total utterances, NTW) and microstructure (MLUw, NDW, MATTR). Paired within-child analyses were conducted; NDW was additionally modeled with NTW as an offset to account for sample length.

Results: Video retells elicited longer samples than book retells (+10.3 utterances; +40.9 NTW; p=.05). MLUw did not differ (-0.20; p=.714). NDW was higher in video retells (+21.6; p=.004), but NDW per word did not differ after adjusting for NTW (rate ratio=1.01; p=.92). MATTR differences were not reliable.

Conclusions: Video retells increased Spanish sample length without meaningful changes in microstructure. Video stimuli may enhance feasibility for Spanish LSA; consistent context is recommended for monitoring.

---

PS2F37

Item Response Theory Analysis of English Sentence Repetition Items for Spanish-English and Vietnamese-English Bilinguals

Alexander Choi-Tucci; University of California, Irvine
Giang Pham; San Diego State University
Lisa Bedore; Temple University
Elizabeth Peña; University of California, Irvine

The Bilingual English Assessment of Morphosyntax (BEAM) is being developed as a tool to aid in evidence-based assessment for Developmental Language Disorder (DLD) in bilingual children by monolingual clinicians. Items were developed to be sensitive to differences in children’s levels of exposure to English and their home language and have been tested with bilingual speakers of Spanish-English and Vietnamese-English. Using item response theory analyses, we examined the item-level parameters of the 24 BEAM sentence repetition items across 1) home language groups, 2) language exposure groups, 3) age groups, and 4) DLD risk groups. Results suggest that one sentence repetition item should be removed from the BEAM due to parameterization issues across multiple grouping variables. Overall, the BEAM sentence repetition items appear to have appropriate difficulty and discrimination for differentiating levels of English morphosyntax ability across the groups tested. There may be differential patterns of item difficulty across home language (Spanish or Vietnamese) and exposure groups. Specific implications for bilingual test design and assessment will be discussed.

Funded by NIH NIDCD 5R01DC018329-05.

---

PS2F38

Revision Strategies During Word Learning in School-Aged Children who are Deaf or Hard of Hearing and Typically Hearing Peers

Jacqueline Zhang; San Diego State University / University of California San Diego
Emily Lund; Texas Christian University
Alyson Abel; San Diego State University

Even when word learning outcomes are comparable to typically-hearing (TH) peers, the process of word learning may be more effortful for children who are deaf or hard-of-hearing (DHH) with aided hearing and learning spoken language. Because robust word learning requires revising initial word meanings as additional information becomes available, differences in revision strategies may contribute to this increased effort. This study examined revision behavior in response to additional information during word learning in DHH and TH children, focusing on patterns of switching and maintaining prior responses. Ten children (6-8-year olds; 5 DHH, 5 TH) with typical language and nonverbal cognition completed a novel word learning task in which participants first chose a referent, and then were incrementally given either confirmatory or contradictory information and opportunities to maintain or revise their previous response. We predict no group differences in accuracy but distinct patterns of revision behavior: following additional information, DHH children will revise responses more frequently than TH children. This increased flexibility during revision may reflect a more effortful learning strategy for DHH children.

Funding: NIH/NIDCD T32-DC007361-18

---

PS2F39

Self-collected parent-child interactions to evaluate the expressive language of children with Down Syndrome: A Cross-context comparison

Tiffany Edgar; University of Wisconsin-Madison
Marianne Elmquist; University of Wisconsin-Madison
Miriam Kornelis; University of Minnesota Twin Cities
Claudia Schabes; University of Wisconsin-Madison
Lizbeth Finestack; University of Minnesota Twin Cities
Audra Sterling; University of Wisconsin-Madison

Children with Down syndrome (DS) show persistent expressive language delays, requiring early intervention. Progress in intervention research is limited by the lack of feasible, sensitive outcome measures. We evaluated the validity of expressive language measures derived from caregiver-recorded parent–child interactions in two contexts: free play with a standardized toy set and a family-selected activity. Twenty-seven children with DS (M age = 49 months) completed two 15-minute home recordings. Transcripts were analyzed using Systematic Analysis of Language Transcripts to derive number of different words (NDW), mean length of utterance in morphemes (MLUm), and percentage of intelligible utterances. Pearson correlations examined associations with MacArthur–Bates Communicative Development Inventory (MCDI), Preschool Language Scale–Fifth Edition Expressive Communication raw score, and Intelligibility in Context Scale (ICS). NDW was strongly associated with MCDI scores across contexts. MLUm was significantly associated with PLS-5 scores, with stronger relations in free play. Results support caregiver-collected samples as scalable outcome measures.

Funding: National Institute on Deafness and Other Communication Disorders (R21HD111807), Eunice Kennedy Shriver National Institute of Child Health and Human Development (P50HD105353), Institute for Clinical and Translational Research (TL1TR002375).

---

PS2F40

The Efficacy of Let’s Know!2 for Improving Oral Language Outcomes in First-Grade Students with Typical Development or At-Risk for Developmental Language Disorder

R.J. Risueño; Old Dominion University
Shelley Gray; Arizona State University
Savannah Romeo; Arizona State University

Small-group multicomponent language interventions have shown promise for increasing language and reading comprehension skills in students with developmental language disorder (DLD). This study evaluated the efficacy of Let’s Know!2 for improving oral language in first-graders with typical development (TD) and those at risk for DLD. A modified randomized controlled trial compared the efficacy of Let’s Know!2 for an at-risk treatment group (n=10), a treatment group with TD (n=4) and an at-risk control group (control; n=7). Participants in the treatment conditions completed 42 20-minute small-group lessons scheduled four days per week. Participants were assessed using curriculum-aligned measures of vocabulary, comprehension monitoring, inferencing, and text structure knowledge at pre-test, mid-intervention after completing a narrative unit, and post-intervention after completing an expository unit. Significant within-group increases in vocabulary and inferencing scores were observed for the at-risk intervention group from pre- to mid- and post-test. At mid-test, both intervention groups outperformed the at-risk control group in narrative vocabulary. For narrative comprehension monitoring, the TD intervention group outperformed the at-risk intervention group at mid-test and both at-risk groups at post-test. Clinical implications discussed.

Funding: H325D170061

---

PS2F41

Mental State Language Use Across Context In Adolescents with Fragile X Syndrome and Down Syndrome

Latifatu Mohammed; University of Wisconsin-Madison
Jill Hoover; University of Massachusetts- Amherst
Audra Sterling; University of Wisconsin-Madison

Mental state language (MSL) use is associated with children’s socioemotional and socio-cognitive learning. Children’s ability to understand and acknowledge others’ mental states during interactions is associated with better social outcomes. In Down syndrome (DS) and fragile X syndrome (FXS), children show delays in using MSL during interactions. Given the importance of language sampling context(narrative and conversation) in language outcomes, it is necessary to assess MSL across contexts, as results may have clinical implications for assessment and intervention. This study evaluates MSL use in adolescents with FXS and DS across narrative and conversational contexts and examines the frequency, type, and diversity of MSL. 20 participants, 10 in each group, aged 9-16 years, completed a 10-minute conversation with an examiner and a narrative task using a wordless picture book. Wilcoxon signed-rank tests and Wilcoxon rank-sum tests will be used to analyze within-group and between-group differences, respectively. We hypothesize that children in both groups will produce more MSL terms in narrative than in conversations, and will differ significantly in the frequency, type, and diversity of MSL use in conversation.

Funding: R01CC019092(MPI Sterling and Hoover)

---

PS2F42

Identifying the Presence of Shared Characteristics in Participants with Developmental Language Disorders (DLD) among Varying Language Deficit Severity Profiles

Susmi Sharma; University of Texas at Dallas
Richard M. Golden Golden; University of Texas at Dallas
Jim Montgomery; Ohio University
Ron Gillam; Utah State University
Julia Evans; University of Texas at Dallas

Consistent with decades-long dilemma among clinicians and researchers in the Developmental Language Disorders (DLD) literature, this work was motivated by ongoing challenges in identifying measures that reliably and consistently classify children with DLD. Using repeated elastic net logistic regression on the Montgomery database (223 DLD and typically developing (TD) participants), our earlier analysis identified 9 cognitive and linguistic features as part of the DLD-deficit profile. However, that analysis utilized -1 SD cutoff in composite z-language scores for DLD classification, a criterion considered too liberal by some researchers. In this study, we tested whether the nine features would remain a DLD-deficit profile when modeling datasets using conservative language-cutoff criteria. We reclassified the Montgomery database at severe cutoff criteria (-1.25 and -1.5) and ran repeated elastic net logistic regression analyses on these datasets to identify the most effective features for DLD classification. Our analysis yielded that 4 of the previous 9 features are part of the DLD-deficit profile, suggesting that all children, whether mild or severe DLD, share unique common cognitive and linguistic constructs.

This project was funded by NIDCD K18-DC021149 (Evans).

---

PS2F43

The impact of context on mother and father verbal responsiveness to their children with Down syndrome

Claudia Schabes; University of Wisconsin-Madison
Audra Sterling; University of Wisconsin-Madison

Individuals with Down syndrome (DS), the most common genetic cause of intellectual disability (ID), struggle to learn and use language above and beyond what is expected for their mental age. Understanding the early language environment of young children with DS is crucial to inform assessment and intervention. Parental responsivity is one of the most powerful predictors of later language outcomes for children with ID; however, research on responsivity has focused primarily on dyadic, mother–child interactions. We sought to investigate the influence of caregiver gender and interaction context (dyadic versus triadic) on the responsivity of caregiver communication. Caregiver communication acts were coded in mother–child, father–child, and triadic interactions across 15 families of 2–5-year-olds with DS. Our logistic mixed model supported a significant three-way interaction between caregiver gender, context, and child expressive language. Limitations, future directions, and clinical implications will be discussed.

Funding: Waisman core grant NIDCHD 50HD105353 (Chang); Vilas life cycle award (Sterling); University of Wisconsin–Madison startup funds (Sterling).

---

PS2F44

Cue-based retrieval and working memory in bilingual children with and without DLD: Contributions of language ability and bilingual experience

Yasmine Ouchikh; The Graduate School and University Center of the City University of New York
Katherine Paulino-Gomez; The Graduate School and University Center of the City University of New York
Klara Marton; Brooklyn College of the City University of New York; Bárczi Gusztáv College of Special Education of Eötvös Loránd University

Children with developmental language disorder (DLD) exhibit weaknesses in working memory (WM). External cues support retrieval in adults, yet cue use in WM has not been systematically studied in children with and without DLD. The present study examined whether cueing supports WM performance in bilingual children and whether language ability and bilingual experience differentially relate to performance. English-Spanish school-aged children with and without DLD (BiDLD; BiTLD) completed a recognition and a cue-based retrieval task (data collection is ongoing). Preliminary results indicate higher performance in the cue-based task: the effect is primarily driven by BiTLD children’s performance but a similar, though weaker pattern is seen in children with BiDLD. For BiTLD children, language ability is related to accuracy in the recognition task, while only Spanish exposure is related to accuracy in the cue-based task. For children with BiDLD, both Spanish proficiency and exposure are related to accuracy across tasks. These findings suggest that cue-supported WM performance varies by task demands and individual differences, with different factors contributing to performance in bilingual children with and without DLD.

---

PS2F45

Features of Book Language and its Relation to Vocabulary Growth in Preschool Age Children

Rebecca Vasile; Florida State University
Sonia Cabell; Florida State University
Matthew Cooper Borkenhagen; Florida State University
Beth Phillips; Florida State University

Shared book reading research is integral to better supporting young children’s language. Shared book reading consists of two types of language: 1) the written language of the text, (book language), and 2) the language that the caregiver adds to the shared book reading experience, (extratextual talk; Read et al., 2025). Most research on shared book reading has either studied it as a holistic activity (e.g., Dowdall et al., 2020) or studied the relation between extratextual talk and children’s outcomes (e.g., Read et al., 2025) while neglecting the potential of book language. The current study extends both the previous studies of shared book reading and of book language in two ways: 1) examining text and word-level features of book language from shared book reading encounters in the early childhood education classroom, holistically and by genre (i.e., narrative and informational books), and 2) completing exploratory analyses to examine whether the book language children were exposed to in the classroom relates to children’s vocabulary growth across the preschool academic year.

Funding sources: R305A2204420, Institute of Education Sciences and H325D210062, U.S. Department of Education

---

PS2F46

Are lexical morphemes more than sound + meaning? A cross-modal fNIRS priming study with English monolinguals, Spanish-English and Chinese-English bilinguals

Yuan Wang; University of Michigan
Rachel Eggleston; University of Michigan
James Baybas; University of Michigan
Nuo Chen; University of Michigan
Wei-Hung Lin; University of Michigan
Ziqi Di; University of Michigan
Aleksandra Witkowska; University of Michigan
James Booth; Vanderbilt University
Hélène Deacon; Dalhousie University
Ioulia Kovelman; University of Michigan

Word reading relies on building strong connections among the sound, meaning, and orthography of words within children’s mental representations. Morphological structure has received substantial attention as a candidate mechanism linking form and meaning during word recognition. Yet the underlying mechanisms by which children leverage morphemic structure, particularly across different language backgrounds, remain unclear. The current project draws on a larger study of literacy in children of varied language (dis)abilities. Young English monolingual and bilingual speakers of structurally-distinct languages (Chinese-English and Spanish-English), N=530, completed a cross-modal audio-visual priming during fNIRS neuroimaging. We hypothesized that the brain bases of morphological processing are distinct from sound-only and meaning-only processing, likely to engage left frontal regions previously found active during morphological processing in monolingual children and adults. We further hypothesized cross-linguistic transfer in bilinguals, with Chinese bilinguals showing stronger meaning-based processing along the ventral neural path and Spanish bilinguals showing stronger sound-based processing along the dorsal neural path. Findings will inform theories of language and literacy development in linguistically diverse learners.

Funding Source: National Institutes of Health (R01HD111637; R01HD109224)

---

PS2F47

What makes you use that test, and is it working for you and your caseload? Factors that influence speech-language pathologists’ perception of standardized norm-referenced assessment quality, policies, and related self-efficacy.

Genesis Romero; The University of Arizona
Mary Alt; The University of Arizona

Speech-language pathologists (SLPs) use standardized, norm-referenced assessments (SNAs) to diagnose communication disorders and determine service eligibility.  However, SNAs are sometimes used inappropriately, and concerns remain regarding SLPs’ understanding of test quality and psychometrics. There is limited research on the demographic and workplace factors that influence how SLPs perceive, select, and use these diagnostic tools. Prior research has examined factors affecting SLPs’ test selection and knowledge, including practice setting and experience working with culturally and linguistically diverse (CLD) populations; however, much of the literature is outdated or not comprehensive.  We surveyed 176 U.S.-based SLPs using a web-based questionnaire that included items assessing clinician, client, and workplace factors influencing their perceptions of SNA quality, attitudes toward policy requirements, and self-efficacy. Data were analyzed descriptively and using Mann–Whitney U tests to examine differences between SLPs grouped by clinical setting, U.S. Census Bureau region, and rate of working with CLD populations.  Effect sizes were calculated for significant findings.  Results may inform training, policy, and advocacy efforts to promote appropriate and psychometrically-sound assessment practices.

The first author is supported through the OSEP Grant H325D230067.

---

PS2F48

Parental Interactional Features and the Language Development of Children with Developmental Disabilities Using Augmentative and Alternative Communication (AAC)

Syrina Merilan; Georgia State University
Rose Sevcik; Georgia State University
Maryann Romski; Georgia State University

This study analyzes the association between parental interactional features and the vocabulary size of children with developmental disabilities (DD) using AAC. Preliminary analysis of 23 parent-child dyads who participated in a 12-week AAC language intervention, found that there was no association between parental interactional features and vocabulary size. Instead, mean length of turn had a strong relationship with vocabulary size. Additionally, strategies parents were coached to employ during these interventions did not alter the turn length or interactional features used by parents throughout the session. These findings bring to light a different association between parental interactional features and vocabulary size for children with DD using AAC compared to their typically developing peers and those with DD who do not use AAC. When coaching parents to use communication supports with their children with DD using AAC, it is recommended that practitioners stress the importance of shortening their turns to allow their children ample opportunity to communicate.

The data were sourced from an archival dataset funded by The U.S. Department of Education, Institute of Education Sciences Grant R324A070122.

PS3S01

Written Language in School-Aged Children

Alana Maffesoli; Grand Valley State University
Hannah Simon; Grand Valley State University
Rylee Davis; Grand Valley State University
Isabella Przybylska; Grand Valley State University
Madison Hartley; Grand Valley State University
Lindsay Edwards; Grand Valley State University
Courtney Karasinski; Grand Valley State University

Many children in the United States do not exhibit proficient writing skills. Few studies investigate strengths and areas that would benefit from support in multilingual writers or writers with language disorder. The current investigation compared narrative, expository, and persuasive writing of kindergartners through twelfth graders who are typically-developing monolingual English-speakers, monolingual English-speakers with language disorder, and typically-developing multilinguals. Multilinguals demonstrated strengths in clausal density in narration, spelling in exposition, and capitalization in persuasion. They may benefit from support for lexical diversity and productivity, spelling, and punctuation in narration. Aspects of macrostructure in narrative writing and overall effectiveness in persuasive writing may benefit from support; however, cultural differences in storytelling and persuasive discourse should be identified and respected. Learners with language disorder may need support for macrostructure in narrative and expository writing and expanding persuasive writing. These findings can help clinicians identify children with language disorder and design appropriate supports for improving the writing of school-aged children with a variety of linguistic profiles.

Funding: Grand Valley State University Graduate School; Grand Valley State University Center for Scholarly and Creative Excellence

---

PS3S02

Complex Syntax in Parents’ Extratextual Book Talk

Ivy Lin; Vanderbilt University
C. Melanie Schuele; Vanderbilt University School of Medicine

Mothers and fathers (16 mother-child dyads, 16 father-child dyads) read one of three books (Bear’s Bargain, Mooncake, Bear’s Shadow, by Frank Asch) to their typically-developing preschooler (Briet-Smith et al., 2017). We coded extant samples for complex syntax (i.e., dependent clauses) to understand the amount and types of complex syntax parents used. There was wide individual variation across parents. Parents reading one of the three books produced less complex syntax. Analyses compare mothers and fathers on complex syntax variables, including proportion, frequency, density and frequency of complex types.

---

PS3S03

What Prevalence Figures did Authors Report from 2017-2025 for Developmental Language Disorder?

Ethan Paik; Vanderbilt University
Brian Weiler; Vanderbilt University School of Medicine
C. Melanie Schuele; Vanderbilt University School of Medicine

Developmental language disorder was suggested for adoption in 2017 and adopted inclusion criteria subsumes the full range of nonverbal IQ within the normal range (i.e., 70 and above).  Thus, DLD is a broader phenotype than specific language impairment. We explored the reported prevalence for DLD in articles published in LSHSS, JSLHR, and AJSLP between 2017 and 2025. We found that overwhelmingly the prevalence of SLI is used when reporting DLD prevalence. We discuss the challenges of underreporting prevalence of DLD.

---

PS3S04

Experience Counts: How Bilingual Exposure and Language Ability Differentially Shape Word Learning

Kaityn Contino; Boston University
Anna Duncan; Boston University
Haley Weaver; Boston University
Malvika Khandelwal; Boston University
Kimberly Crespo; Boston University

Word learning requires children to learn the meaning of a word, and to correctly produce its form. Prior work suggests that bilingual experience and language ability separately contribute to children’s acquisition of word-referent mappings. However, the extent to which these factors jointly support the learning of word-referent mappings and correct word-form productions remains unclear. We examined these questions in a cross-situational word learning (CSWL) task with 44 Spanish–English bilingual children ages 5–9. Participants were exposed to English-like novel word–object mappings. A two-alternative forced choice task measured receptive word-learning and a picture naming task measured expressive word-learning. Receptive and expressive performance significantly improved as length of bilingual experience increased. Language ability uniquely predicted receptive but not expressive performance. Taken together, findings suggest that receptive word-learning draws on both bilingual experience and language ability, whereas the encoding and production of newly learned word forms may place greater demands on language-specific experience than on language ability.

This research was supported by NIDCD K23DC022006 funds awarded to Kimberly Crespo and T32DC013017 fellowship awarded to Kaityn Contino, Anna V. Duncan, Malvika Khandewal.

---

PS3S05

Translanguaging Profiles in Majority-Minority Contexts: Spanish-English Narratives in the Early Elementary Years

Abbygail Crandall; University of Arizona

Monolingual norms in clinical assessment can pathologize translanguaging, or the fluid use of a complete bilingual repertoire, as linguistic deficit. This deficit view creates diagnostic inequity in majority-minority communities, where rigid English-only schooling contrasts with bilingual community practices. This study adopts a critical lens to characterize translanguaging in narrative retells, establishing a normative baseline for typical cross-linguistic flexibility in bilingual communities. Participants were 90 Spanish–English bilingual children (Grades 1–3, 30 per grade) attending English-only schools in Southern Arizona. Using Systematic Analysis of Language Transcripts (SALT) administration protocols, children retold Frog, Where Are You? in English and Spanish. Transcripts were coded for frequency, grammatical integration, and discourse function. We report developmental profiles regarding the frequency and structural integration of translanguaging. Analyses examine whether translanguaging function evolves from lexical retrieval mechanisms to communicative cohesion strategies and how these trajectories interact with home language exposure. By establishing base rates for typical development, this work supports speech-language pathologists in adopting culturally responsive, asset-based diagnostic practices.

Funding: Council of Academic Programs in Communication Sciences and Disorders PhD Scholarship

---

PS3S06

The Flynn Effect in Nonverbal Intelligence Tests for Children with Developmental Language Disorder: A Meta-Analytic Study

Arinjoy Bhattacharjee; University of Connecticut
Tammie Spaulding; University of Connecticut

The Flynn Effect, defined as systematic increases in scores on norm-referenced intelligence tests over time, has been widely documented in the general population. However, it remains unknown whether this effect is present in children with developmental language disorder (DLD), for whom nonverbal intelligence (NVIQ) tests are routinely administered. The present study examined whether a Flynn Effect exists in DLD and whether its expression is comparable to that observed in typically developing (TD) peers. A meta-analysis was conducted of studies published between 1994 and 2024 that included children with DLD and age-matched TD controls. Across 231 paired samples, mean NVIQ scores were modeled as a function of “datedness,” defined as the number of years between norming and data collection. Tests were, on average, 16.2 years old at administration and varied across studies. TD children demonstrated a clear Flynn Effect across measures, whereas children with DLD did not show a consistent overall effect. Test-specific analyses revealed variability, indicating that the presence, direction, and magnitude of the Flynn Effect depend largely on the specific test used, with implications for score validity.

---

PS3S07

Teacher-Parent Discrepancies in Bilingual Language Ratings and Their Associations with Language Use Profiles

Soujin Choi; Speech-Language-Hearing Sciences, University of Minnesota
Eugene Wong; Speech-Language-Hearing Sciences, University of Minnesota
Lizbeth Finestack; Speech-Language-Hearing Sciences, University of Minnesota
Kerry Ebert; Speech-Language-Hearing Sciences, University of Minnesota

Bilingual children’s language use varies across home and school contexts, which may lead to parents and teachers to observe different aspects of their abilities. Discrepancies between parent and teacher ratings of language could be treated as inconsistencies; however, they may reflect variation shaped by bilingual experience. We examined teacher-parent ratings discrepancies across language domains and their association to children’s language environments and objective language measures. We collected data from 57 triads including a kindergartener, parent, and teacher, with the children coming from homes where they use languages other than English. Parents and teachers rated children’s English and home language abilities, parents reported home language environment, and children completed English narrative retell and standardized measures of speech production and comprehension. Analyses examined teacher-parent discrepancy patterns across domains, their association with language environment, and relationships between each reporter’s ratings and objective English measure outcomes. The teacher-parent discrepancies were significantly different for vocabulary and sentence domains in English, and vocabulary, speech, grammar, and comprehension in the home language. Findings will inform the interpretation of multi-informant screening data in bilingual assessment.

Funding Source: NIDCD R01DC019895

---

PS3S08

Expressive Language Sampling as a Source for Outcome Measures for Autism Treatment Studies: Feasibility, Practice Effects, and Test-Retest Reliability

Lizbeth Finestack; University of Minnesota – Twin Cities
Sara Kover; University of Washington
Angela John Thurman; MIND Institute and Department of Psychiatry and Behavioral Sciences, University of California Davis Health
Leonard Abbeduto; MIND Institute and Department of Psychiatry and Behavioral Sciences, University of California Davis Health

In response to increasing rates of autism diagnoses among children in the United States, researchers have increasingly focused on developing and evaluating both behavioral and pharmaceutical interventions to support the language development of autistic individuals. However, the rigor of studies on treatment efficacy is limited due to a lack of psychometrically sound outcome measures appropriate for autistic individuals. Traditional standardized, norm-referenced assessments are inadequate due to behavioral-level heterogeneity within the population and the tendency of such assessments to overestimate language ability relative to language use in real world contexts. One potential solution is the use of expressive language samples. In the current study, we examined the psychometric properties of measures derived from two language sampling contexts: narration and conversation with 79, 6- to 23-year-olds on the autism spectrum. Results indicated a high level of feasibility and strong test-retest reliability. Further examination of validity and sensitivity of the measures to change over time is warranted before recommendations for research and clinical use can be made.

Funding: NICHDR01HD074346

---

PS3S09

Production of Spanish determiners over time in bilingual preschoolers with and without DLD

John Gallagher; San Diego State University; University of California, San Diego
Alicia Escobedo; California State University, San Bernardino
Gregory D. Keating; San Diego State University
Sonja Pruitt-Lord; San Diego State University

Research on Spanish-speaking children shows TD versus DLD group differences in accuracy of determiners (e.g., articles), but at least two gaps remain. First, existing evidence comes from monolinguals and school-aged bilinguals. Second, overlap in accuracy between groups in bilinguals is common, making diagnosis using this quantitative measure difficult in “borderline” cases. Qualitative analysis of grammatical error patterns may further elucidate group differences. We analyzed (1) determiner accuracy and (2) error patterns in language samples from 32 Spanish/English bilingual preschoolers with TD and DLD at the beginning and end of a preschool year. Preliminary results revealed that children with TD: (1) had significantly higher accuracy and (2) made significantly more substitution (cf., omission) errors than their peers with DLD. The clinical utility of determiner accuracy and error analysis in bilingual preschoolers will be discussed.

Funding: Local philanthropy, OSEP H325D230037

---

PS3S10

Validating a Shortened BIOS: Congruence of Exposure Classifications

Mariana Gomez Becerra; University of California, Irvine
Joseph Hin Yan Lam; University of California, Irvine
Nahar Albudoor; Ohio State University
Alexander Choi-Tucci; University of California, Irvine
Lisa Bedore; Temple University
Elizabeth Peña; University of California, Irvine

The assessment of bilingual populations is challenging and current best practice is to assess both languages. Although best recommended practice by ASHA is to assess both languages, SLPs face high caseloads and limited time. Bilinguals also demonstrate variability in language exposure and proficiency, and some may not have sufficient exposure in one language to justify assessment. Accurately identifying language exposure profiles is critical to make appropriate individualized assessment decisions. This study examined whether a shortened version of the Bilingual Input Output Survey-Home (BIOS-Home) classifies bilingual children similarly to the original hour-by-hour format. Cut-offs for determining language(s) of assessment were also explored using data from the Bilingual English-Spanish Oral Screener (BESOS). Participants were 1,337 Spanish-English bilingual preschool and kindergarten children from two large-scale studies in Texas and Utah. Preliminary discriminant analyses revealed strong alignment between shortened and original BIOS-Home profiles (84%-85%) across exposure categories. Findings support the use of an efficient exposure measure to guide language of assessment decisions for bilingual children in clinical practice.

Funding sources: H325D220045, R01DC007439 and R01DC010366 (PI: Elizabeth Peña)

---

PS3S11

Exploring Lexical Structure in Autistic Children who Exist in Different Regions of the Language Endowment Spectrum: A Network Modeling Approach

Eileen Haebig; Louisiana State University
Stanley West; Louisiana State University
Arielle Borovsky; Purdue University
Letitia Naigles; University of Connecticut
Lee Evans; Boys Town National Research Hospital
Justin Kueser; Boys Town National Research Hospital
Christopher Cox; Louisiana State University

The wide variation in spoken language abilities in autistic children is not well understood. The current study used network modeling techniques to examine the structure of early lexicons in autistic children who are minimally speaking (n = 41) and in autistic children who have mild to moderate vocabulary delays (n = 82). Semantic networks were built based on two semantic databases: child-oriented word associations and semantic feature norms. Though both groups demonstrate learning that reflects both the associative and semantic feature structures in their learning environments, the different sources of information seem to more heavily influence learning for one group over the other. Autistic children with mild-to-moderate spoken vocabulary delays have networks that are more attuned to association structure, while minimally-speaking autistic children are more attuned to semantic feature structure. This pattern may reflect greater learning from more naturalistic (associative word co-occurrences) learning experiences in autistic children with mild-to-moderate spoken language delays and greater learning from structured learning experiences in minimally-speaking children.

Funding sources: R01DC007428, R03DC013638, R01DC018593, and the Research Competitiveness Subprogram – Louisiana Board of Regents

---

PS3S12

Predicting Late Talkers’ Word-Learning Treatment Success: Unlocking the VAULT with CART Analysis

Heidi Mettler; University of Illinois Chicago
Nora Evans-Reitz; University of Arizona
Mary Alt; University of Arizona

Rationale: We wanted to be able to predict which late-talking toddlers would and would not respond to a particular intervention treatment early enough to adjust the treatment for potential non-responders.

Methods: We assembled both continuous and categorical data from 74 children who completed the Vocabulary Acquisition and Usage for Late Talkers (VAULT) word-learning protocol. Children were classified as responders or nonresponders. We ran Classification and Regression Tree (CART) analyses to sort children into ‘responders’ and ‘nonresponders’ and compared how well the model matched the treatment data.

Results: Our winning model correctly identified 45/51 responders as responders (88% accuracy) and 20/23 nonresponders as nonresponders (86%) accuracy. The model had 5 decision points, including both continuous and categorical variable, and allowed for classification after no more than 3 treatment sessions.

Conclusion: This decision tree has clinical potential as a data-based way to adjust treatment for potential non-responders before they engage in a full course of what is likely to be ineffective treatment.

This work was supported by the National Institute on Deafness and Other Communication Disorders Grant R01DC015642.

---

PS3S13

Exploring the Feasibility of Automated Language Sample Analysis for Bilingual Children

José Ortiz; University of Maryland, College Park
Anna Soares; University of South Florida
Jessica Nolasco; University of Maryland, College Park
Risueño R.J.; Old Dominion University
Restrepo M. Adelaida; University of South Florida
Shelley Gray; Arizona State University

This study examines the efficacy of an automated language sample analysis (LSA) system that integrates automatic speech recognition (ASR) and natural language processing (NLP), when used with bilingual Spanish-English speaking children. Participants came from an existing dataset that included language sample transcripts and audio from 484 Spanish-English bilinguals aged 5 to 7 years. We examined the transcription accuracy of ASR, the concurrent validity, and diagnostic accuracy of automated LSA metrics. Transcription accuracy varied with background noise and participant intelligibility. The fully automated LSA system exhibited comparable performance to metrics derived from manual transcription.

---

PS3S14

Processing of mixed-language input in bilingual autistic toddlers

Caitlyn Slawny; University of Wisconsin – Madison; Waisman Center
Margarita Kaushanskaya; University of Wisconsin – Madison; Waisman Center

Research suggests that language switching carries costs for bilinguals (e.g., Byers-Heinlein et al., 2017). Given core deficits in social-pragmatic communication and executive functioning for autistic individuals (e.g., Eigsti et al., 2011), bilingual autistic children may find mixed-language input particularly difficult to parse. So far, 13 two- to three-year-old autistic toddlers have been tested. The BOSA with the ADI-R were used to diagnose ASD, and the Vineland-3 Spanish was used to assess communication, socialization, daily living skills, and motor skills. The CARS-2 indexed autism severity. Language-specific vocabulary was measured with the MB-CDIs and language ability with the PLS-5 Spanish. We assessed processing of familiar word-object pairs in real time with a field-of-two looking-while-listening task across three conditions: English, Spanish, and mixed. Preliminary results show that autistic toddlers looked at the target above chance for Spanish and English conditions but were at chance for the mixed language condition. Findings will help improve language exposure strategies for bilingual autistic children.

Funding sources: National Institutes of Health Grants F31 DC023102, R01 DC021150, U54 HD090256

---

PS3S15

Metapragmatics in Adolescence: Exploring the Validity of the Transition Pragmatics Interview

Gerard Poll; Miami University
Brooke Harmon; Miami University
Abbey Parks; Miami University
Anne Bartz; Miami University
William Boone; Miami University
Janis Petru; Elmhurst University

The ability to reflect on how to adapt language for a specific social situation is metapragmatic ability. Metapragmatic ability emerges in mid-childhood and can be limited in children with developmental communication disorders, such as autism or developmental language disorder. Metapragmatic ability is correlated with conversational skills in childhood, but less is known about the role of metapragmatics in adolescent and young adult social communication. To further validate a new measure of adolescent social communication, the Transition Pragmatics Interview (TPI), we administered both the TPI and an established measure of metapragmatics to forty-seven individuals (14-21 years) with and without developmental communication disorders. We found that TPI scores correlated with metapragmatics scores (r = .66). Age predicated metapragmatic scores after accounting for disorder status. Metapragmatic ability appears to play an important role in social communication skills in late adolescence and young adulthood.

Funded by NIDCD under award R15DC020521.

---

PS3S16

Evidence of Inequity on the Arizona English Language Learning Assessment

Melanie Randall; The University of Arizona
Mary Alt; The University of Arizona

In Arizona, English Language Development (ELD) programming is mandated for all public school students who have not passed the Arizona English Language Learning Assessment (AZELLA). Students with Individualized Education Plans are held to the same passing requirements as typically-developing peers. Student performance is scored as: emerging, basic, intermediate, or proficient. To exit from mandated ELD programming, students must score “proficient” in all 4 areas of the assessment: listening, speaking, reading, and writing, without individualized accommodations. Researchers sought to determine whether passing the AZELLA is statistically independent of disability status.

We accessed state-published 2024 AZELLA results for children in grades 1-6 that identified disability status. Using a Chi-Square Test for Independence, we determined whether passing rates were independent of disability status at each grade level. Students with disabilities passed the AZELLA at significantly lower rates than non-disabled peers at each grade level.

These data suggest that multilingual students with disabilities are disproportionately impacted by Arizona ELD policy. Requirements for participation in the ELD program must be adjusted to account for disability status. The first author is supported through OSEP Grant H325D230067.

---

PS3S17

Semantic networks in the narratives of children with DLD and typical language

Tomas Savcenko; Czech. Acad. Sci.; Charles University
Filip Smolík; Czech. Acad. Sci.,, Inst. Pschol.

We examined semantic networks in narratives produced by children with Developmental Language Disorder (DLD/SLI) compared to typically developing (TD) peers. Prior research shows that children with language impairments often have weaker vocabulary structures, including fewer semantic associations and less organized lexical networks. Using narratives from 668 children in Gillam’s corpus, we constructed semantic networks for each narrative using word embeddings and computing cosine distances between words to estimate similarity. Words with high semantic similarity were connected, and several network measures—clustering coefficient, transitivity, mean degree, and path length—were analyzed through regression models controlling for MLU and token count. Results showed that TD children had lower clustering, lower degree centrality, and higher transitivity than children with DLD. These findings contrast with previous work on general vocabulary structure, where DLD is associated with less interconnected networks. We suggest that narrative production imposes syntactic and discourse constraints that lead children with DLD to rely on a smaller, more tightly interconnected set of words. The study demonstrates the value of network approaches and vector-based representations for analyzing language production.

---

PS3S18

Preclinical Evidence for Early Intervention in Rett Syndrome

Isabella Myers; The University of Texas at Dallas
Soumya Joshi; The University of Texas at Dallas
Crystal Engineer; The University of Texas at Dallas

Rett syndrome (RTT) is a neurodevelopmental disorder characterized by a severe regression of learned skills, including language, around 6-18 months old. Individuals with RTT are typically diagnosed after the onset of these symptoms and experience lifelong communication impairments. Here we assess if the preserved period prior to regression may be leveraged to support long-term learning. In this study a rodent model of RTT, the Mecp2+/- rat, and typically developing littermates were taught a complex speech sound discrimination task at either a pre- or post-symptomatic timepoint representing an early intervention paradigm and current intervention practices, respectively. Time to learn, task engagement and accuracy was assessed for all groups at a post-regression timepoint. The late-trained Mecp2+/- rats performed significantly worse than age-matched controls on all metrics, demonstrating profound speech sound discrimination impairments. However, early-trained Mecp2+/- rats did not demonstrate these deficits, even when assessed at a post-symptomatic timepoint. This work demonstrates a maintained effect of early training in Mecp2+/- rats, providing support for early intervention in RTT.

This work is funded by the Eagles Autism Foundation and the International Rett Syndrome Foundation.

---

PS3S19

Integrating Social Emotional Learning With Language and Literacy in Early Childhood: A Systematic Review and Multilevel Meta-Analysis

Phoebe Ahn; Vanderbilt University
Andrew Chang; Vanderbilt University
Eon-Joo Jang; Vanderbilt University
Jason Chow; Vanderbilt University

Language, literacy, and social emotional development are closely interconnected in early childhood and jointly support school readiness (Chow et al., 2020; Qi et al., 2006; Duncan et al., 2007). Although integrated intervention models are increasingly discussed, quantitative evidence on their effectiveness remains limited. We conducted a systematic review and preliminary multilevel meta-analysis of early childhood interventions that explicitly embedded social emotional learning within language or literacy instruction for children from birth through kindergarten. Our systematic screening began with 5,002 abstracts and resulted with 61 studies for inclusion. We estimated meta-analytic models in R using metafor with robust variance estimation. Results indicate that integrated language and literacy plus SEL interventions was positive and moderate (g = 0.275, SE = 0.141), suggesting that interventions focused on improving child language and literacy outcomes that integrate SEL components improved outcomes across a range of domains.  Overall, our findings suggest (1) the field recognizes the importance of both developmental domains, (2) intentional and precise measurement of the integration is key, and (3) there is promise for integrated early childhood approaches.

---

PS3S20

The Effect of Treatment Words’ Semantic Properties on Increasing Expressive Vocabulary in Late-Talking Toddlers

Mary Alt; University of Arizona
Heidi Mettler; University of Illinois – Chicago
Nora Evans-Reitz; University of Arizona
Kimberly Leon; University of Arizona
Camila Curutchague; University of Arizona
Sarah Cretcher; University of Arizona

Rationale: We were interested in determining if late-talking toddlers could capitalize on their existing semantic knowledge to assist in increasing their expressive vocabulary.

Methods: We quasi-randomly assigned late talkers enrolled in the Vocabulary Acquisition and Usage for Late Talkers (VAULT) expressive vocabulary intervention protocol to one of two conditions: learning words from dense semantic neighborhoods or learning words from sparse semantic neighborhoods.  Neighborhood density was calculated individually for each child.

Results: We compared treatment effect sizes across conditions using Bayesian t-tests. We also measured children’s ‘learning to learn’ through the rate of words learned per week on the MCDI, comparing their rate pre-treatment to during treatment.

Conclusions: In addition to gaining evidence about the overall efficacy of the VAULT treatment protocol, we will be able to determine the role that semantic density plays in treatment, allowing for more informed decisions about treatment targets.

This work was supported by the National Institute on Deafness and Other Communication Disorders Grant R01DC015642.

---

PS3S21

The difference in strategies speech-language pathologists consider using with elementary vs high-school aged clients who exhibit challenging behavior

Nicole Hoopes; Texas Christian University
Emily Lund; Texas Christian University

Thirty-eight percent of children experience maltreatment by adolescence (Finkelhor et al., 2015), increasing the likelihood of challenging behaviors (Hughes et al., 2017).  Hoopes and Lund (2025) found SLPs often use non-trauma-sensitive techniques when addressing behavior. No study, however, has examined the effect of client age on trauma-informed behavior. Nevertheless, exclusionary discipline occurs at higher rates for secondary versus elementary students (Leung-Gagné et al., 2022). The aim of this study was to determine (a) whether there is a difference in SLPs’ use of non-trauma-sensitive strategies between elementary students and teens, (b) the types of non-trauma-sensitive strategies used with teens, and (c) whether SLP characteristics predict use of non–trauma-sensitive practices with teens. Participants viewed 10 videos of students exhibiting challenging behavior in therapy and provided written responses describing how they would respond. A codebook was developed to classify responses as regulation, connection, or non–trauma-sensitive. Within-subject analyses provide insight into age-related differences in SLPs’ behavior management approaches.

This project is funded by Project INTERSECT, a leadership training grant funded by the Office of Special Education Programs.

---

PS3S22

Linking Domain General Cognition to Language: Executive Function Predictors of Morphosyntax in Young Children

Jill Hoover; University of Massachusetts Amherst
Emelia Ferguson; University of Massachusetts Amherst
Brittany Ciullo; University of Massachusetts Amherst
Amy Banasik; University of Wisconsin Madison
Claudia Schabes; University of Wisconsin Madison
Audra Sterling; University of Wisconsin Madison

A substantial literature links children’s executive functions (EFs)—notably working memory and inhibition—to language development, but many studies use broad composites derived from standardized tests which may overlook the nuances in specific EF–language domain relationships. This study focuses on morphosyntax during a key developmental window and uses narrow, theory-driven measures to clarify that link. Participants include 67 monolingual English-speaking children between the ages of 3;0–6;8. EFs were measured with three experimental tasks: Corsi Block (working memory), Flanker (inhibition), and Local-Global (shifting); morphosyntax was indexed by sentence imitation, grammaticality judgment, and tense/agreement productivity from spontaneous samples. Using latent EF and morphosyntax constructs we test a model in which EF statistically predicts morphosyntactic ability, controlling for age. By including multiple, focused measures of morphosyntax, this work aims to produce a clearer understanding of how domain-general cognition supports grammar, specifically. Insights from the proposed study have the potential to refine theory and guide more targeted assessment and intervention for children with language difficulties.

This research is funded by NIH R01 DC019092 awarded to MPIs Jill Hoover and Audra Sterling.

---

PS3S23

Differences in Lexical Structure Underlie Language Delay in Children

Abigail Hultquist; University of Colorado Boulder
Eliana Colunga; University of Colorado Boulder

Early language delays are associated with academic and social difficulties, yet persistent language delay is not typically identified until ages 4–5, after a critical window for early intervention has passed. Between ages 2–3, many children with limited vocabularies spontaneously catch up to peers (late bloomers), while others show persistent delays (persistent late talkers). This study examines whether early lexical structure can distinguish these trajectories. Using longitudinal CDI data from 122 children assessed monthly between 18 and 30 months, we constructed semantic networks for each child’s productive vocabulary at each month. Children were classified as persistent late talkers if they remained below the 20th percentile across sessions and as late bloomers if they began below but later exceeded it. We analyzed network connectivity metrics of each child’s lexicon at session 1, when all children showed delays. Critically, early vocabulary size does not distinguish groups, but persistent late talkers exhibited significantly weaker small-world structure than late bloomers. These findings suggest that early vocabulary structure may serve as an early indicator of persistent language delay.

Funding: University of Colorado Boulder

---

PS3S24

School Speech-Language Pathologists’ Perspectives on Collaboration and Interprofessional Practice

Zoe Hussey; Vanderbilt University
Colleen Walsh; Vanderbilt University
Phoebe Ahn; Vanderbilt University
Michelle Lifson; Vanderbilt University
Jason Chow; Vanderbilt University

School speech-language pathologists (SLPs) encounter child challenging behaviors that impact intervention. However, behavior management training is limited in SLP graduate programs. Minimal training leaves school SLPs underprepared to support challenging behaviors which can influence their ability to provide intervention as designed. Research highlights collaboration and interprofessional practice (IPP) among school SLPs and other school professionals as one potential solution to increasing preparedness. This study explored practicing school SLPs’ perspectives on collaboration and IPP. We conducted virtual semi-structured interviews with 19 practicing school SLPs and used an inductive analytic approach to identify and refine themes based on emerging data. Findings revealed themes encompassing practicing school SLPs’ perceptions of, training on, and barriers and facilitators to collaboration and IPP. The study highlights suggestions at multiple levels for increased collaboration and IPP, including continued training and school-wide support for educational professionals. Future research should examine the quantity and quality of collaboration and IPP to expand on these findings.

This work was supported, in part, by the Office of Special Education Programs (H325D2300371; H325D220055) and the National Institute of Mental Health (T32-MH18921).

---

PS3S25

Nonlinguistic cognitive processing profiles and language ability in bilingual and monolingual children

Kerry Ebert; University of Minnesota – Twin Cities
Giang Pham; San Diego State University
Michelle Ramos; San Diego State University

Rationale: Developmental language disorder (DLD) has been associated with cognitive processing deficits and is increasingly viewed through multidimensional accounts rather than single-core deficits. This study examined multidimensional nonlinguistic cognitive processing profiles and whether they relate to language ability across diverse linguistic backgrounds.

Method: Participants included 257 children (ages 5–7; 161 bilingual, 96 monolingual) with varying language abilities. Children completed six computerized nonlinguistic tasks crossing three cognitive domains (Processing speed, sustained selective attention, working memory) and two modalities (auditory, visual). Latent profile analysis was used to identify cognitive profiles, with bilingual status and sentence repetition (SR) scores as covariates.

Results: A two-profile solution emerged, characterized by “low” and “high” performance across all tasks. Membership in higher-performing profiles was significantly associated with stronger SR scores but bilingual status did not predict profile membership.

Conclusions: Nonlinguistic cognitive processing ability may be a continuous variable rather than manifesting in discrete processing profiles. This ability is associated with language ability but not bilingual status. Future studies will examine whether this variable can assist with identifying DLD in diverse populations.

Funded by NIDCD R01DC019613.

---

PS3S26

Working Memory Predictors of Receptive and Expressive Vocabulary in Children Who Are Deaf or Hard of Hearing

Daniel Ibarra; Texas Christian University
Krystal Werfel; Boystown National Research Hospital
Emily Lund; Texas Christian University

Children who are deaf and hard of hearing (DHH) demonstrate substantial variability in spoken vocabulary outcomes. Evidence suggests that cognitive–linguistic abilities may contribute to this variability. This study examined the extent to which phonological working memory and lexical retrieval predicted receptive/ expressive vocabulary outcomes in second grade among DHH and typically hearing (TH) children. Participants included 57 DHH children and 41 TH peers who completed measures of working memory, lexical retrieval, receptive vocabulary, and expressive vocabulary. Multiple linear regression analyses indicated that nonword repetition significantly predicted both receptive and expressive vocabulary among DHH children. In combined-group analyses, digit span predicted expressive vocabulary, whereas nonword repetition predicted receptive vocabulary. Group-specific analyses revealed distinct patterns of association by hearing status. Findings highlight the central role of phonological working memory in spoken vocabulary development for DHH children and underscore the importance of considering cognitive–linguistic factors as influences on language development.

---

PS3S27

Measuring Growth in Teacher Language Support: Comparing Quality Ratings and Frequency Counts

Eon-Joo Jang; Vanderbilt University
Caroline Hiskey; Vanderbilt University
Taydi Ray; Vanderbilt University
Rachel Frampton; Vanderbilt University
Jennifer Baumingham; University of Washington
Ann Kaiser; Vanderbilt University

Early language support in toddler classrooms is critical for preventing and addressing language delays, yet studies examining observational measures of specific teacher language support strategies remain limited. This exploratory study compared the sensitivity of rating-based versus count-based observational metrics for detecting change in teachers’ language support strategies. Data were collected pre- and post-intervention from four toddler classroom teachers participating in a professional development trial. Teacher practices were assessed using the Language and Behavior Observation Record (LABOR), which captures both quality ratings (0-2 scale) and frequency counts (proportion of utterances) for language support strategies. Results revealed marked divergence between metrics. One intervention teacher showed zero change on all rating measures yet demonstrated substantial frequency increases. Both intervention teachers showed consistent gains in language modeling and semantic contingency despite minimal rating changes. Findings demonstrate that rating scales exhibit ceiling effects that systematically underestimate professional development effects among high-performing teachers. Results have critical implications for evaluating interventions aimed at improving classroom language environments for children at risk for language disorders.

This project was supported in part by the Institute of Education Sciences (R324A200193).

PS3S28

Opening the Door to Literacy: SLP Confidence and Phonological Awareness Instruction for Children Who Use AAC

Jordyn Julian; UTHSC
Jillian McCarthy; UTHSC
Erinn Finke; UTHSC

Phonological awareness (PA) is foundational to literacy development, yet children who use augmentative and alternative communication (AAC) have inconsistent access to PA instruction. This study examined whether graduate training and clinician confidence predict speech-language pathologists’ (SLPs’) provision of PA instruction with AAC users. A total of 162 SLP responses were analyzed from a national survey. Chi-square analyses and binary logistic regressions were conducted. Confidence in AAC treatment knowledge and completion of graduate AAC coursework significantly predicted whether clinicians taught PA skills. Clinicians who self-reported AAC confidence were four times as likely to provide PA instruction (OR = 4.14, 95% CI [1.85, 10.10]), and those who completed AAC coursework were more than three times as likely to do so (OR = 3.23, 95% CI [1.18, 10.46]). AAC clinical experience and larger AAC caseloads significantly predicted clinician confidence. Findings suggest that while coursework provides foundational knowledge, confidence plays a critical role in translating training into practice. Strengthening AAC preparation and confidence may improve children’s access to foundational literacy instruction.

Funding: U.S. DOE, Office of Special Education Programs (OSEP), grant H325D230072.

---

PS3S29

English Morphosyntactic Accuracy and Difficulty Structure in Discourse-Matched Spanish–English and Vietnamese–English Bilingual Children

Joseph Hin Yan Lam; University of California, Irvine
Karen Zyskind; University of California, Irvine
Giang Pham; San Diego State University
Lisa Bedore; Temple University
Elizabeth Peña; University of California, Irvine

English morphosyntax is a core marker used in the assessment of developmental language disorder, yet bilingual children’s performance varies as a function of cross-linguistic influence and typological differences. Prior cloze-based studies of Spanish–English and Vietnamese–English bilinguals demonstrate shared developmental difficulty gradients alongside language-specific vulnerabilities, but group comparisons have not controlled for discourse-level language ability. In this study, we examined English morphosyntactic accuracy and difficulty structure in Spanish–English and Vietnamese–English bilingual children matched on discourse-level narrative ability, language exposure, and demographics. Participants were selected using 1:1 propensity-score matching and completed an English morphosyntax cloze task targeting multiple grammatical forms. Group differences in overall accuracy were evaluated, and performance was examined across derived difficulty clusters. By holding discourse-level language ability constant, this study tests whether shared and language-specific difficulty patterns persist across typologically distinct bilingual groups. Findings showed that a four-cluster structure in terms of difficulty could be applied to both groups. Convergence was noted for the difficulty of various English morphosyntax structures, but not in terms of cluster boundaries. This suggests a shared developmental order of English morphosyntax across bilingual populations with home-language-specific modulation on difficulty thresholds.

Funding: R01DC018329 (Peña)

---

PS3S30

Exploring the Potential for the CCC-2 Pragmatic Subscales to Differentiate ADHD from DLD in Boys and Girls.

Chris Bradford; University of Utah
Sean Redmond; University of Utah

Pragmatic deficits have been implicated in multiple neurodevelopmental disorders, including developmental language disorder (DLD) and attention-deficit/hyperactivity disorder (ADHD) (Geurts & Embrechts, 2008; Carruthers et al., 2022). Direct comparisons of pragmatic profiles between these groups remain scarce, limiting the utility of pragmatic measures for differential  diagnosis. This study leveraged archival data (Ash et al. 2017; Redmond et al., 2024) to examine potential group and sex differences in parent-reported abilities across the pragmatic subscales on the Children’s Communication Checklist-2 (CCC-2: Bishop, 2006) in 265 children with either typical development (TD), DLD, or ADHD (6- to 9-years). Results indicated consistently lower ADHD group means relative to the TD and DLD groups. Areas under the ROC curve ranged from .662 to .856, with higher levels generally associated with TD vs. ADHD discrimination and with boys relative to girls. The Initiation subscale demonstrated the highest values for ADHD vs. DLD discrimination for both sexes (Sensitivity: .75 .77; Specificity: .83 .89). Findings suggest that pragmatic subscale patterns may support differential diagnosis within a broader assessment framework.

Funding: NIDCD: R01DC017153, R01DC011023

---

PS3S31

Overcoming Barriers to Academic Achievement – The Perspectives of Students with DLD

Julia Wallmann; Division of Sensory Organs & Communication, Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden
Christina Samuelsson; Speech Language Pathology, Department of Clinical science, Intervention and Technology, Karolinska Institutet, Stockholm, Sweden
Birgitta Sahlén; Logopedics, Phoniatrics, & Audiology, Department of Clinical Sciences, Lund University, Lund, Sweden
Christina Reuterskiöld; Division of Sensory Organs & Communication, Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden
Olof Sandgren; Logopedics, Phoniatrics, & Audiology, Department of Clinical Sciences, Lund University, Lund, Sweden
Anna Ekström; Division of Sensory Organs & Communication, Department of Biomedical and Clinical Sciences, Linköping University, Linköping, Sweden

Students with DLD are less likely to perform well academically than their typically developing peers. The aim of this study was to increase understanding of problems students with DLD can face, as well as solutions they themselves propose. The goal was to increase student-informed awareness of strategies to provide support in school. 13 adolescents with DLD in Swedish compulsory mainstream education (13–16 years) participated in semi-structured interviews. Their experiences of barriers as well as activities and strategies supporting academic achievements were analysed using reflexive thematic analysis. Results show that students found support in school to be sporadically delivered, which affected their learning negatively. Support was, however, described as an important facilitator when provided. Moreover, clearly communicated instructions were reported as a facilitator, reducing inactivity resulting from students’ difficulties to understand educational expectations. Finally, challenges related to language-based activities were recurrently reported, but students also shared that their language function varied between activities. Performing in school was perceived to be easier when students could make use of their relative individual linguistic strengths in language-related activities.

Funding: The Swedish Research Council

---

PS3S32

Facilitating Language Skills of Children with Autism:  A Novel Contingent Responses Intervention

Camille Nuttall; Vanderbilt University
Jena McDaniel; Vanderbilt University Medical Center

Targeting vocalizations for children with autism spectrum disorder (ASD) who are not yet talking, offers an opportunity to intervene at an earlier development level than spoken words. Such intervention is expected to support spoken language development due to the strong connection between vocalization development and spoken word development. This study evaluates a novel intervention for vocalizations that utilizes contingent responses to child vocalizations with and without vocal enhancement strategies using a single case research design. There are two active conditions including (1) contingent responses and (2) contingent response plus vocal enhancement. We determine whether the contingent nature of the responses influences the quantity and quality of child vocalizations and whether vocal enhancement strategies offer additional value. This intervention supports future investigation into contingent responses using vocalizations for children with ASD.

Project PAL (H325D230037, U.S. Department of Education)

---

PS3S33

A Pilot Treatment Program Designed to Facilitate Inclusion in Healthcare Decision-Making by Adults with Intellectual Disability

Meredith Saletta Fitzgibbons; Midwestern University
Christina del Toro; Midwestern University
Omar M. Khan; The Douglas Center
Cheney Hurley; University of Iowa
Jacob Oleson; University of Iowa

Rationale: Healthcare providers must learn to elicit responses from adults with intellectual and/or developmental disability (IDD) effectively. Many adults with IDD have difficulty responding to Likert-type scales accurately and with nuance (selecting responses in the middle of the scale rather than on either extreme). These adults also often demonstrate acquiescent responding: the tendency to respond in the affirmative to please the questioner.

Methods: In Study 1, six adults with IDD learned the specific steps involved in completing Likert-type scales. Three of these participants demonstrated acquiescent responding and continued to Study 2, where they answered questions worded in a neutral way or a way that could “lead the witness.” Both studies used a single-subject with multiple baselines design.

Results: In Study 1, no participant responded more accurately or with greater nuance following the intervention. In Study 2, post-intervention performance exceeded baseline performance six times for one participant and one time for another.

Conclusions: We discuss the importance of ensuring that all individuals can express decisions regarding their own healthcare.

Funding source: Midwestern University Speech-Language Pathology Program (Downers Grove, IL) departmental funds

---

PS3S34

The role of cognitive control in disentangling bilingualism and developmental language disorder

Klara Marton; Brooklyn College, CUNY
Thorfun Gehebe; CUNY Graduate Center
Yasmine Ouchikh; CUNY Graduate Center

This presentation will focus on the relationship between cognitive control and language to reveal how cognitive control assessment may complement standardized language testing to disentangle developmental language disorder (DLD) and emerging bilingualism (e.g., English language learners; ELL).

Cognitive control is a complex psychological construct encompassing the core processes that are essential to achieve goal-directed behavior, particularly when facing conflict, habitual practices, or contextual changes (Botvinick et al., 2001; Cohen, 2017). We examined working memory updating and task switching in monolingual and bilingual children with and without DLD (N=142). We hypothesized that children with better language abilities and higher language proficiencies exhibit better cognitive control skills (Marton, 2024).

The outcomes from 4 different cognitive control tasks showed that bilingual typically developing children perform either similarly or better than their monolingual peers, whereas children with DLD produce more perseverative errors, violate more rules, and perform more slowly than their peers. Thus, cognitive control tasks may efficiently complement standardized language tests during the assessment of children with different language skills (i.e. ELL; DLD), and my help us to reduce the number of misdiagnoses.

Funding: Hungarian Academy of Sciences; PSC-CUNY award

---

PS3S35

Exploring la Comprensión Bilingüe in Autistic Children: Designing an Eye-Tracking Paradigm

Lavanda-Sofia Probasco; UMass Amherst
Maria Alejandra Meneses; UMass Amherst
Megan Gross; UMass Amherst

Little is known about bilingual language development in autistic children, and even less about those with minimal spoken language, who are often excluded from research. The current project seeks to address this gap by using eye-tracking to examine how Spanish-English bilingual autistic children process input in English, Spanish, and with code-switching.  Developing such an eye-tracking paradigm requires integrating considerations from previous eye-tracking research conducted with neurotypical bilingual children and monolingual autistic children. Visual stimuli were carefully selected and paired considering lexical properties of their labels in both languages. Auditory sentence stimuli were recorded by a native bilingual adult. Use of both eyetracking software and manual coding of video-recorded data, as well as frequent engaging breaks, are some strategies used to promote engagement of autistic participants. By refining experimental methods that accommodate the unique needs of participants and their families, we hope to make participation in bilingual autism research more accessible overall. Preliminary data will provide an initial look into bilingual language processing in autistic children, including those who are minimally speaking.

Funding: NIH K23DC020224, Spaulding Smith Fellowship, departmental PhD support

---

PS3S36

The MacArthur-Bates Communicative Development Inventory III Revised (CDI-IIIR): A new parent-report measure of language development for English-speaking 2.5- to 4-year-olds

Philip Dale; University of New Mexico
Adriana Weisleder; Northwestern University
Alexandra Carstensen; Arizona State University
Sefela Yalala; Northwestern University
Matthew Betashour; Arizona State University
Natalie Rossman; Arizona State University
George Kachergis; Stanford University
Virginia Marchman; Stanford University

This study revised the MacArthur-Bates Communicative Development Inventory-III (CDI-III) for US English. Originally published as an assessment designed for children aged 30-37 months, the previous US English CDI-III is limited in the age range and the aspects of language it assesses (with few items assessing morphological or metalinguistic awareness). Recent adaptations, most notably the Swedish CDI-III, have been shown to have strong psychometric properties for children 2.5 – 4 years old.  We sought to learn from successful work in other languages and revise the US English CDI-III in order to: 1) expand the target age range and 2) incorporate more aspects of language (morphology, metalinguistic awareness). A committee developed a revised instrument, which was tested with 227 children aged 29-49 months. Analyses revealed very good to excellent reliability of the scales, strong correlations with age, and expected intercorrelations between scales.  The resulting CDI-IIIR is a promising instrument for language assessment of English-speaking children 30-47 months. The ongoing project aims to develop norms and to study the instrument’s utility as a screener in clinical contexts.

Funded by the CDI advisory board.

---

PS3S37

Relation of Fine Motor Skills and Expressive Verb Vocabulary of 30-Month-Olds

Katrina Nicholas; Nevada State University
Lynn Perry; University of Miami
Danielle Hu; University of Wisconsin – Madison
Joseph Lamberti; Nevada State University
Russell Beckley; Independent Consultant

We examined the relation of verbs to fine motor skills in late-talking and typically developing children. Actions that require a child’s use of their hands compose the majority of their early-learned verbs (Maouene et al., 2008). Late-talking children have fewer expressive verbs (Hadley et al., 2006; Horvath et al., 2019) and are statistically more likely to demonstrate poor fine motor skills than their typically developing peers (DeVeney et al., 2024), but it is not known if these deficits are associated. Using a Poisson distribution, we analyzed the fine motor skill scores from the Mullen Early Scales of Learning (Mullen, 1995) and expressive vocabulary inventories (including action words) from the MacArthur Bates Communicative Development Inventory: Word and Sentences (MCDI; Fenson et al. 1993) for twenty-three children (M = 29.77 months; range = 26 – 34 months). We found a statistically significant link between children’s action words and fine motor skills, suggesting that fine motor skills should be considered in evaluation of late-talking children when considering their expressive vocabulary growth trajectory for verbs.

Funding: California State University, East Bay Faculty Support Grant

---

PS3S38

??? where are you?: Code-switching practices in Ukrainian- and Russian-English speaking children

Yana Krutnyk; University of Wisconsin–Madison
Emma Libersky; University of Wisconsin–Madison
Margarita Kaushanskaya; University of Wisconsin–Madison

Code-switching is a common practice amongst bilingual populations, yet little to no work has been conducted on Russian-English and Ukrainian-English speaking populations. We examined the use of code-switching in bilingual children using language samples. Children listened to and retold three stories: once in English, once in Russian or Ukrainian, and once using both of their languages. These language samples were transcribed and coded for code-switches. By documenting typical code-switching practices in these populations, we provide a baseline to help identify atypical code-switching patterns that may characterize language impairment. We will also be able to examine how code-switching varies with children’s language profiles in the full sample. Broadly, this work provides insight into language development within understudied Russian- and Ukrainian-English bilingual populations.

This work was supported by NIH R01DC020447, NIH R01 DC016015, and NIH U54 HD090256.

---

PS3S39

When to worry about late talking: Are parental concerns about language related to late talker status in Spanish-English bilingual toddlers?

Adriana Weisleder; Northwestern University
Milton Guendica; Northwestern University
Sefela Yalala; Northwestern University
Murielle Standley; Northwestern University
Krystal Alvarez-Hernandez; Northwestern University
Anny Castilla-Earls; The University of Texas at Dallas

Parental concerns can be an important source of information about developmental delays, but there is little known on whether they provide accurate information about language delays in bilingual toddlers. The current study adapted the Parental Report of Speech and Language Problems (PRSLP) – a validated instrument for capturing parental concerns about language in older children – for use with toddlers, and investigated whether parental concerns are related to late-talker (LT) status in Spanish-English bilingual 2 year-olds. Sixty-seven parents completed the adapted PRSLP (11 questions) as well as an adapted, bilingual version of the MacArthur-Bates CDI II in English and Spanish (used to classify LT status). Analyses revealed that the odds of having a child who was a LT was significantly higher in parents with concerns than in parents without concerns. Sensitivity and specificity analysis revealed that a cut point of =3 parental concerns correctly classified 79% of participants as LTs. These results suggest that parental concerns about language can be a promising measure when screening for language delays in Spanish-English bilingual toddlers.

Funded by NIDCD grant R21DC018357.

---

PS3S40

Regular Past Tense Use in Southern White English Nonfiction Retells

Sarah Lynn Neiling; University of Cincinnati
Allison Breit; University of Cincinnati
Ying Guo; University of Cincinnati

Identification of Developmental Language Disorder (DLD) for speakers of Southern White English (SWE) is challenged by the difficulty of interpreting modified and unmodified scoring for non-Mainstream American Englishes. Strategic scoring of overt form use may be a promising scoring method, which analyzes dialect-contrastive forms (e.g., for simple past-tense -ed nonmainstream overt: “had bump/ed” and zero-marked: “bumpØ”) and dialect-universal forms (e.g., mainstream overt: “bump/ed”). The current study replicates and extends Oetting et al. (2021) using an informational text retell language sample to examine strategic scoring for the most clinically differentiating feature—regular past tense –ed—for typically developing (TD) and DLD preschoolers who speak SWE. Retells from two intervention studies were transcribed and coded for mainstream overt, nonmainstream overt, and zero-marked forms of past-tense -ed and compared across groups and by form. Results will further the evidence base on strategic scoring use.

Funding: Institute of Education Sciences Grant R324A130205; Institute of Education Sciences Grant R305A200271

---

PS3S41

Comparable Accuracy, Distinct Error Patterns: English Morphosyntax in Spanish-English and Vietnamese-English Bilinguals

Karen Zyskind; University of California, Irvine
Joseph Hin Yan Lam; University of California, Irvine
Alexander Choi-Tucci; University of California, Irvine
Aya M Kahala; University of California, Irvine
Manqi Wang; University of California, Irvine
Lisa Bedore; Temple University
Giang Pham; San Diego State University
Elizabeth Peña; University of California, Irvine

Accurate identification of developmental language disorder (DLD) in bilingual children is complicated by cross-linguistic influence and variability in language exposure, which may produce error patterns that resemble impairment. This study examined the error patterns in an English morphosyntax cloze task between Spanish–English and Vietnamese–English bilingual children, matched on discourse-level narrative ability, exposure, and demographics. Nearest-neighbor Mahalanobis matching yielded 47 pairs of matched participants. Error coding focused on omission, substitution, and no-response patterns across five morphosyntactic structures. Groups did not differ in overall accuracy for either structure. However, preliminary analysis showed Spanish–English children produced significantly more substitution errors, while Vietnamese–English children produced more no-response errors across grammatical forms. Findings suggest that bilingual groups may show comparable accuracy yet differ in qualitative error profiles. Preliminary findings suggest Spanish-English bilinguals produced significantly more substitution errors while Vietnamese-English bilinguals produced significantly more no-response errors. Together, results underscore the importance of typology-sensitive interpretation when evaluating bilingual children’s English morphosyntax and may help prevent overidentification of developmental language disorder.

Funding: R01DC018329 (Peña)

---

PS3S42

Should We Tell Them? Explicit Explanation of Word Meanings on Grammar Treatment

Melissa White; University of Arizona
Sarah Cretcher; University of Arizona
Elena Plante; University of Arizona

We tested whether being provided semantic information about words used in Enhanced Conversational Recast therapy targeting grammatical morphemes facilitates or interferes with learning. In our study, 23 preschool-age children with DLD received recast therapy targeting a variety of grammatical morphemes. Some children were given explicit definitions of target vocabulary to be used with those morphemes. Other children were not provided definitions. At the end of the treatment, children who heard definitions demonstrated slightly greater use of their target forms compared to children who did not hear definitions. However, at follow-up testing weeks after treatment ended, children who did not hear definitions showed higher use of target forms compared to children who heard word definitions. This indicates higher retention in children who did not hear definitions compared to children who did. Results suggest that providing semantic information may interfere with consolidation in this therapy.

This project was funded through R01 DC015642.

---

PS3S43

Identifying Clinically Relevant Language Measures for Developmental Language Disorder Diagnosis in Spanish-English Bilingual Children: A Predictive Modeling Approach

Jiali Wang; Texas A&M University
Lisa Fitton; University of South Carolina
Marc Goodrich; Texas A&M University

Diagnosing developmental language disorder (DLD) in Spanish-English bilingual children remains challenging. While the converging evidence approach offers a comprehensive framework, it is time-consuming and resource-intensive. This study employed predictive modeling to identify which language measures best align with expert diagnostic decisions of DLD. Participants included Spanish-English bilingual kindergarteners assessed across multiple assessments: standardized assessments (vocabulary, sentence repetition), SALT-derived LSA microstructure and macrostructure measures, and teacher-report of children’s language proficiency. Using elastic net logistic regression with 80/20 train-test split and 5-fold cross-validation, we identified features that aligned more closely with the diagnostic decision while controlling for multicollinearity. The model achieved strong classification performance (AUC, sensitivity, specificity >0.85). Based on feature importance, top predictors include children’s standardized assessments (i.e., vocabulary and sentence repetition), and a number of micro- and macrostructure langauge sample measures, such as surbordination index, MLU, conflict and resolution, cohesion, narrative scoring scheme total score, and NDW. Findings provide empirical guidance for streamlining bilingual DLD assessment by prioritizing measures that contribute more diagnostic information, potentially reducing assessment burden while maintaining accuracy.

R01HD114547; R21HD106072

---

PS3S44

The Effect of Language Sample Duration on the Reliability of Mean Length of Utterance in Mandarin-Speaking Children with Autism

Yuqing Li; Beijing Normal University
Ying Hao; Nanjing Normal University

While Language Sample Analysis (LSA) offers high ecological validity for assessing language skills in children with autism, its clinical utility is hindered by time-consuming transcription. This study investigated the minimum language sample length required for reliable MLU assessment in Mandarin-speaking children with autism. Participants were 17 Chinese children with autism receiving language intervention. A dataset of 4 to 6 independent 10-minute language samples was collected per child during unstructured free-play interactions with a parent. Each of these samples was subsequently segmented into 1-, 3-, 5-, 7-, and 10-minute subsamples to calculate MLU using the Child Language Analysis (CLAN) system. Participants were divided into high and low language proficiency groups. Analysis showed that children with different proficiency levels required different minimum sample durations. The low-proficiency group reached clinical-level consistency with the 10-minute sample at 5 minutes (r=0.90), whereas the high-proficiency group required a minimum of 7 minutes to meet the recommended reliability threshold. These findings provide empirical support for optimizing language sampling analysis procedures and offer stratified reference standards for assessing language performance in children with autism.

The authors declare no competing financial interests.

---

PS3S45

From Words to Sentences: Cognate Effects in Bilingual Children

Sophie Levi; San Diego State University & University of California, San Diego Joint Doctoral Program in Language and Communicative Disorders
Sonja Pruitt-Lord; San Diego State University & University of California-San Diego – LCD Joint Doctoral Program

Sentence repetition tasks have clinical utility for assessing expressive language in bilingual children, yet the role of cross-linguistic interactions in shaping performance is not well understood. One such interaction may involve cognates, which are translation equivalents that overlap phonologically and semantically across languages (e.g., English–Spanish telephone–teléfono; English–Hebrew telephone–?????). Cognates reliably facilitate single-word performance, particularly in the language of less bilingual experience. The present study extends prior work on cognate effects to sentence repetition tasks in 4–6-year-old Spanish–English and Hebrew–English bilingual children, and age-matched English monolinguals. Bilingual children completed researcher-derived sentence repetition tasks in English and their additional language, and monolingual children completed versions in English only. Sentences differed only in cognate status, with cognates matched to noncognate items on psycholinguistic variables. Analyses assessed effects of cognate status and task language, with cognate facilitation expected in bilingual (but not monolingual) children, and varying as a function of bilingual language experience. Findings clarify how cross-linguistic lexical overlap may shape sentence-level performance and inform interpretation of sentence repetition in bilingual assessment.

This work was supported by UCSD SEED and SDSU CORE Fellowships.

---

PS3S46

Item Response Theory Analysis of English Cloze-Response Items for Spanish-English and Vietnamese-English Bilinguals

Alexander Choi-Tucci; University of California, Irvine
Joseph Hin Yan Lam; University of California, Irvine
Karen Zyskind; University of California, Irvine
Giang Pham; San Diego State University
Lisa Bedore; Temple University
Elizabeth Peña; University of California, Irvine

To address the lack of bilingual clinicians, the Bilingual English Assessment of Morphosyntax (BEAM) is being developed to identify developmental language disorder in bilingual children. The BEAM is designed to be administered by monolingual English-speaking clinicians. Items for the BEAM were designed to capture variability in English exposure and different home languages. This study uses item response theory to examine the item-level parameters of 83 BEAM cloze-response items across 1) home language groups (Spanish-English and Vietnamese-English), 2) age groups, and 3) DLD risk groups. Results suggest that several items may be considered for removal from subsequent iterations of the BEAM. Items generally performed in expected patterns for age, exposure and risk groups, and had similar levels of difficulty and item discrimination across home language groups. There were specific morphosyntactic targets that demonstrated cross-group differences. Implications for the development of the BEAM and bilingual test design and assessment at large will be discussed.

Funded by NIH NIDCD 5R01DC018329-05.

---

PS3S47

Evaluating Aspects of Social Communication as Predictors of Reading Comprehension in School-Age Autistic Children

Meghan Davidson; The University of Kansas
Kandace Fleming; The University of Kansas
Mohammad Karbakhsh; The University of Kansas
Thomas Gottstein; The University of Kansas

Reading comprehension is poor in many autistic individuals. Oral language abilities are well-established to contribute to poor reading comprehension, but less is known about the additional role of social communication. The goal of this study was to examine the role of reading accuracy, language comprehension, and aspects of social communication (e.g., perspective taking, pragmatics, emotion recognition, social attention) in reading comprehension in school-age autistic children. In a preliminary sample of school-age (9- to 12-year-old) autistic children with phrase-level speech and no intellectual disability, emotion recognition in voices and parent-reported social communication significantly correlated with reading comprehension, but only emotion recognition in voices was a significant predictor after accounting for language abilities. Together, our model with language comprehension, emotion recognition in voices, parent-reported social communication abilities, and reading accuracy accounted for 70% of variability in reading comprehension. This suggests that social communication may additionally contribute to reading comprehension challenges in autistic children.

This study was funded by a grant from the NIDCD (R21-DC020786).

---

PS3S48

A Pilot Study of Production and Comprehension of Mental State Verbs in Narratives by School-Aged Autistic Children

Mohammad Karbakhsh; The University of Kansas
Lauren Tigner; University of Texas at Dallas
Meghan Davidson; The University of Kansas

Mental state verbs (MSVs; e.g., think, remember, decide) are critical for narrative production and comprehension and are closely linked to theory of mind. Autistic children have documented difficulties with narrative skills and internal state language; however, few studies have directly compared MSV use and understanding across narrative generation, retell, and comprehension within the same sample. This pilot study examined MSV production and comprehension across these three narrative modalities in four school-aged autistic children (ages 9–12 years). Participants completed narrative generation, retell, and comprehension tasks using wordless picture books from the Mercer Mayer Frog series. Outcomes included expected MSVs, total MSVs, and comprehension accuracy. Nonparametric within-group analyses revealed significant differences across modalities. MSV comprehension was stronger than production, and MSV production was greater during narrative retell than generation, though several pairwise comparisons approached significance. These preliminary results suggest that autistic children are stronger at comprehending MSVs than producing MSVs; although, there are weaknesses in MSV use across Narrative Comprehension, Retell, and Generation.

Funding: National Institute on Deafness and Other Communication Disorders (NIDCD), R21DC020786