Thank you!
Thank you to all who attended the 2025 SRCLD conference!
Check back soon for 2026 conference details.
Session One: Thursday, May 29 from 3:30 – 5:00 PM
Topic: Communication in Autism
SOP1-1
Socioemotional Inferencing in More and Less Social Texts for Autistic School-Age Children
Meghan Davidson; University of Kansas
Kandace K. Fleming; University of Kansas
Kathy Thiemann-Bourque; University of Kansas
Thomas Gottstein; University of Kansas
Listening comprehension (i.e., comprehension of spoken stories) is difficult for many autistic individuals. In this study, we manipulate text characteristics to compare school-age autistic children’s listening comprehension across less social texts (i.e., texts with limited social information) and more social texts (i.e., texts with social information). Further, we examined if literal (i.e., about information explicitly stated in the text), physical inferential (i.e. inferences about the physical world), and socioemotional inferential (i.e., inferences about characters’ mental or emotional state) question type was differentially affected by text type. In an initial sample of eight school-age autistic children, the main finding was that socioemotional inferencing was specifically poor for the more social texts. This finding has implications for how we assess and interpret comprehension assessments for autistic children.
This study was funded by a grant from NIDCD (5R21DC020786-03).
SOP1-2
Toddlers With and Without Autism Spectrum Disorder Show Large-Scale Differences in Semantic Biases in Vocabulary Composition
Justin Kueser; Boys Town National Research Hospital
Lee Evans; Boys Town National Research Hospital
Sara Kover; University of Washington
Letitia Naigles; University of Connecticut
Children with autism spectrum disorder (ASD) show qualitatively different vocabulary development vs. peers with typical development (TD). The goal of this study is to explore how semantic network organization affects the vocabularies of children with ASD. We ask two questions regarding the vocabularies of toddlers with and without ASD: How does overall semantic network structure differ across groups?, and What biases toward distinct semantic feature types does each group have? We hypothesized differences that may explain the slower vocabulary development in ASD. Participants were 43 18-to-42-month-old toddlers with ASD matched to 47 16-to-23-month-old toddlers with TD on vocabulary size. Individual semantic networks were created for each child and properties of the networks were measured. The ASD group produced vocabulary words in strongly connected semantic neighborhoods, particularly for verbs when semantic connections involved concrete perceptual and idiosyncratic encyclopedic features. Crowded semantic networks may lead to increased competition in lexical retrieval, resulting in slower vocabulary development. Differences in semantic feature biases may contribute to the reduced shape bias, noun vs. verb bias, and idiosyncratic/decontextualized language use in ASD.
Funding: NIH/R01DC007428.
SOP1-3
Do autistic children understand individual words within the delayed echolalia they produce? A pilot case study
Janine Mathee-Scott; Michigan State University
Grace Corrigan; Michigan State University
Emily Lorang; Michigan State University
Zachary Hesse; Michigan State University
Jennifer Johnson; Michigan State University
Courtney E. Venker; Michigan State University
There has been considerable debate related to the characterization of delayed echolalia (i.e., repeated phrases) within autistic children’s expressive and receptive language. One theory, known as Gestalt Language Processing, posits that delayed echolalia in autistic children is indicative of a “gestalt” processing style, suggesting that autistic children who produce delayed echolalia initially do not analyze or understand the individual words within these phrases in the same way that “analytic language processors” do. To date, no empirical studies have tested autistic children’s comprehension of individual words within their delayed echolalia. This pilot study provides proof-of-concept for examining this question. An autistic three-year-old participated in an eye-gaze task, which tested her comprehension of individual nouns and verbs selected from her own delayed echolalia utterances. Results indicate that this participant demonstrated real-time comprehension of twelve target words (ten nouns and two verbs) that she also produced within delayed echolalia. Findings have significant theoretical and clinical implications, suggesting that claims purporting that autistic children who produce delayed echolalia do not analyze individual words within those phrases should be further examined.
Funding sources: NIH R01DC020165; R21DC016102
Session Two: Friday, May 30 from 1:30 – 3:00 PM
Topic: Early Childhood Communication
SOP2-1
The evolving contribution of early and later gestures to subsequent vocabulary outcomes in Mandarin-speaking children with cochlear implants
Ling-Yu Guo; University at Buffalo
Lei Xu; Shandong ENT Hospital, China
Jianfen Luo; Shandong ENT Hospital, China
Min Wang; Shandong ENT Hospital, China
Jinming Li; Shandong ENT Hospital, China
Linda Spencer; New Mexico State University
Children with cochlear implants (CIs) often face challenges in developing vocabulary skills. Although research shows that gestures support vocabulary growth, some early intervention programs for children with CIs do not explicitly emphasize gesture development, citing concerns that gestural (visual) communication may interfere with spoken language. These concerns remain untested. This study examined the role of early and later gesture skills in predicting vocabulary outcomes in children with CIs.
Participants were 50 Mandarin-speaking children with CIs. Parents reported their children’s gesture skills at the time of CI activation and 3, 6, and 12 months afterward using a checklist. They also reported their children’s expressive vocabulary at 12 months post-activation.
Early gesture scores at 3 months after CI activation were a stronger predictor than later gesture scores for children’s expressive vocabulary size at 12 months post-activation. However, later gesture scores at 6 months post-activation were a stronger predictor than early gesture scores for children’s vocabulary size at 12 months post-activation. Targeting specific types of gestures at different stages may maximize vocabulary outcomes in children with CIs.
Funding Source: Faculty Grants for Global and International Research, University at Buffalo
SOP2-2
Associations between vowel acoustics and expressive language in late talkers: An analysis of in-home, naturalistic language samples
Marc Maffei; Boston University
Jana Iverson; Boston University
Speech motor skills have the potential to play a role in predicting language outcomes in late talkers. We derived vowel acoustics (objective, non-invasive metrics of speech motor development) and expressive language variables from recordings of in-home naturalistic language samples of 36-month-old late talkers and typically developing children. Significant group differences were identified related to mean vowel formants. Formant variability was similar between groups. Among late talkers, multiple acoustic variables related to mean formant values and formant variability were correlated with the amount (i.e., total number of utterances, number of different words) and complexity (i.e., mean length of utterance) of spoken language. These correlations were not present in the TD children or the sample as a whole. These findings (1) suggest that naturalistic language samples can be utilized in acoustic analyses of speech motor development; (2) provide a critical groundwork for future studies examining the longitudinal development of speech motor and language skills in late talkers; and (3) motivate future work examining the role of oral motor function in predicting language outcomes.
This work was funded by NIDCD F32 Fellowship DC022496.
SOP2-3
A Deep Learning-Based Automatic Speech Recognition system for both typically developing and at-risk Preschool Children in Naturalistic Early Childhood Settings
John Hansen; The University of Texas at Dallas
Satwik Dutta; The University of Texas at Dallas
Dwight Irvin; The University of Florida
Jay Buzhardt; The University of Kansas
Monitoring child development in terms of speech/language skills has a long-term impact on their overall growth. Given various challenges with direct observation, deploying speech technology will assist in extracting meaningful information for teachers/practitioners. These will help to identify and respond to students in need, immediately impacting their early learning and interest. Young children have significantly different/developing articulation skills as compared to adults and older children, therefore commercially available speech technology solutions might not work. This study takes a deep dive into exploring state of the art Automatic Speech Recognition solutions for low-resource spontaneous preschool (3-5yrs) children with and without developmental delays, being involved in various activities, and interacting with teachers and peers in naturalistic classrooms. This study also performs a subject-wise evaluation of the impact of curated child ASR system on tracking few prominent language learning milestones which include: WH-Words (who, what, where, etc.), Verbs, Pronouns, and Nouns.
This project is supported by National Science Foundation Cyberlearning: Grant #1918032.
Session Three: Saturday, May 31 from 1:30 – 3:00 PM
Topic: Developmental Language Disorder Across Ages and Methods
SOP3-1
The Nature of Word Learning Problems among Elementary Students with Developmental Language Disorder
Karla McGregor; BTNRH
Jacob Oleson; University of Iowa
Timothy Arbisi-Kelm; BTNRH
Nichole Eden; BTNRH
We aimed to determine the nature of the word learning problem associated with developmental language disorder (DLD) and its change from Grades 1-4.
Thirty-eight children with DLD and 45 with typical language development (TLD) were randomly assigned to learn via ostensive naming (direct instruction) or fast mapping (indirect instruction). To capture lexical and semantic aspects of learning, we probed recognition of word forms, form-to-referent links, and semantic categories after instruction.
The TLD group consistently outperformed the DLD group and performed better via ostensive naming than fast mapping. For the DLD group, the benefit of ostensive naming was limited to learning form-to-referent links. The TLD and DLD groups did not differ in rate of growth in either condition or on any outcome, with one exception: the DLD group demonstrated a steeper developmental trajectory for form learning via ostensive naming.
Elementary students with DLD struggle with both lexical and semantic aspects of word learning and do not fully benefit from direct instruction. However, their learning abilities develop at an expected pace over the first four years of school.
Funding: NIH-NIDCD-5R01DC011742-07
SOP3-2
Disfluency Patterns and Speech Rate in Mandarin-Speaking Children with Developmental Language Disorder: Evidence from Narrative Tasks
Luyuan Geng; The Hong Kong Polytechnic University
Li Sheng; The Hong Kong Polytechnic University
Speech fluency is an important dimension to evaluate a person’s language ability. Speech fluency is relatively understudied in children with developmental language disorder (DLD) and existing studies have produced inconsistent findings, partly due to methodological differences in measuring disfluencies (e.g., C-unit/utterance-based vs. word/syllable-based). This study examined 21 Mandarin-speaking children with DLD (52–71 months; Mean = 61.61) and 21 typically-developing (TD) peers (51–70 months; Mean = 61.51) as they told and retold MAIN stories. Audio recordings were analysed in Praat to identify disfluencies—silent pauses (> 250 ms), filled pauses (e.g., “um,” “uh”), repetitions, and self-corrections—and to calculate speech rate as the total language produced per second. Disfluency and speech rates were measured using both C-unit-based and syllable-based methods. Results indicated that syllable-based measurements were more effective than C-unit-based analyses in distinguishing between Mandarin-speaking DLD and TD groups, revealing that children with DLD exhibit more speech disruptions and produce less content per unit of time during the narrative process.
This study was supported by a Hong Kong Research Grants Council General Research Fund Project [15613323].
SOP3-3
Language Processing Network in Adults with Developmental Language Disorder
Noelle Abbott; University of Washington
Gabriel Cler; University of Washington
Developmental Language Disorder (DLD) affects approximately 7% of the population and is characterized by persistent language difficulties that cannot be attributed to other disorders. Despite its prevalence, the neurobiological basis of DLD across the lifespan remains poorly understood. Most neuroimaging research in DLD focuses on children, with few studies examining adults with DLD. This study is the first to investigate language processing in adults with DLD and typically developed (TD) adults using a robust fMRI language localizer task. To determine inclusion in the DLD group, all participants completed the Fidler protocol which identifies adults with DLD based on weighted scores on a modified token task and a spelling test. MRI data were preprocessed using fMRIprep, and within and between-group comparisons of brain activation were conducted using FSL’s FEAT. For this study, we anticipate no significant differences in activation location between groups but expect weaker activation in adults with DLD within core language regions. This study will contribute to our understanding of the neural basis of DLD in adulthood, providing insights into the functional characteristics of the language processing network.
T32 DC005361.
Supported in part by: NIDCD and NICHD, NIH, R13 DC001677, Margarita Kaushanskaya and Audra Sterling, Principal Investigators
University of Wisconsin-Madison – Department of Communication Sciences and Disorders