Session 13 - Language
Examining the feasibility of audio and video caregiver-collected language measures
Marianne Elmquist1, Audra Sterling1, Claudia Schabes1, Miriam Kornelis2, Lizbeth Finestack2 (1University of Wisconsin-Madison. 2University of Minnesota-Twin Cities)
Email of first author: melmquist@wisc.edu
Background: Researchers and clinicians are challenged to identify appropriate outcome measures to evaluate language growth of young children with Down Syndrome (DS). We examined the feasibility of two methods for caregiver-collected language measures: caregiver-child interaction videos and audio-recordings with the Language ENvironment Analysis System (LENA). Our research questions are as follows:
- What is the feasibility of caregivers of young children with DS self-collecting audio recordings for communication sample analysis?
- What is the feasibility of caregivers of young children with DS self-collecting video recordings for communication sample analysis?
- Do caregiver-collected audio or video measures demonstrate more feasibility?
Methods: Fifteen caregivers and their children with DS between 2 and 5 years of age participated in this study. Participants recorded two 16-hour LENA recordings and two 15-minute videos of caregiver/child interactions at two timepoints spaced one month apart. We used descriptive statistics to compare the recording fidelity and feasibility of both procedures.
Results: Overall our results show that caregivers were able to self-record audio and video caregiver-child interactions. All LENA audio recordings (100%) were collected with an average of 14hrs. Most 15-minute videos (96.43%) were successfully recorded. Caregivers were visible 93.26% of the time and children were visible 95.40% of the time. However, caregivers often needed help sharing videos with researchers, only 20.69% of videos were independently uploaded to an online server.
Conclusion: Results from our study have the potential to change how researchers and clinicians collect caregiver-child interactions in a manner that is reliable, valid, and convenient. However, more research is needed on how to ensure caregivers are able to share videos with researchers independently.
Early Communication Development in Infants and Toddlers with Down Syndrome
Authors: Daniela Fanta1, Kai-Yun K. Huang1, & Laura J. Mattie1
- Department of Speech & Hearing Science, University of Illinois at Urbana-Champaign, USA
Email: dcf2@illinois.edu, ljhahn@illinois.edu
Individuals with Down syndrome experience challenges in developing and acquiring communication and language (difficulties in expressive language, relative strengths in receptive language and nonverbal social communication1, 2). Documenting performance on standardized early communication screening measures can provide a quick snapshot of skills to help clinicians determine the need for services and inform intervention targets. The present study addressed the following research questions:
What is the pattern of early communication on the Communication and Symbolic Behavior Scales-Caregiver Questionnaire (CSBS-CQ) compared to the normative mean?
What percentage of infants and toddlers with Down syndrome are in the “range of concern” on the CSBS-CQ?
Method: Thirty-six mothers completed the CSBS-CQ and Vineland Adaptive Behavior Scales for their infants and toddlers (M=17.66, SD=5.61). The CSBS-CQ is a standardized questionnaire that examines communication across three domains (Social, Speech, Symbolic) with seven subdomains. A Total Communication Composite is calculated with a range of concern being a standard score of 81 or less. The range of concern for domains and subdomains is 6 or less.
Results/Conclusions: Young children had significantly lower Total Communication (M=78.69; SD=7.65) than the normative mean, t(35)=-16.72, p<.001, with 61.1% of children in the range of concern. They had significantly lower scores than the normative mean across the three domains (Social [M=7.25; SD=2.45], Symbolic [M=6.17; SD=1.95], and Speech [M=6.50; SD=1.94]) and on 6 of the 7 subdomains (p’s <.001). 66.6% were in the range of concern for the Symbolic domain, 50% were in the range of concern for the Speech domain, and 36.1% were in the range of concern for the Social domain. Thus, this measure can provide a quick snapshot of skills for clinicians to target in interventions. Case examples will be presented.
References:
Fidler, D. J., Philofsky, A., & Hepburn, S. L. (2007). Language phenotypes and intervention planning: Bridging research and practice. Mental Retardation and Developmental Disabilities Research Reviews, 13, 47–57.
Luyster, R. J., Seery, A., Talbott, M. R., & Tager-Flusberg, H. (2011). Identifying early-risk markers and developmental trajectories for language impairment in neurodevelopmental disorders. Developmental Disabilities Research Reviews, 17(2), 151–159. https://doi.org/10.1002/ddrr.1109
Characterization of Communication Profiles of Toddlers with Down Syndrome and Autism
Anna Stewart1,2, Alexis Monk2, Tanisha Chanda3, Maggie Norberg1, Katherine Pawlowski1, Nicole Baumer4, Carol Wilkinson1
1Boston Children’s Hospital, 2Harvard University, 3Georgia Institute of Technology, 4Children’s Hospital Colorado
Email: annastewartbch@gmail.com
Background:
Children with neurodevelopmental disorders (NDDs), including Down syndrome (DS) and autism spectrum disorder (ASD), face diverse communication challenges. DS is associated with receptive and expressive language impairments due to oral motor difficulties and intellectual disability. Additionally, 15–30% of children with DS also meet criteria for ASD, further diversifying their communication profiles. Traditional assessments often exclude minimally verbal children, whereas natural language sampling (NLS) offers a more inclusive approach. However, no standardized NLS transcription protocol exists for children with limited expressive language and articulation challenges, particularly one that incorporates pre-speech and non-speech vocalizations.
Objective:
This pilot study compared the communication profiles of toddlers with DS and those with autism and a co-occurring language impairment (ASD-LI), and examined relationships between NLS measures and standardized assessments.
Method:
68 toddlers (DS: n = 34, mean age = 42.5 months; ASD-LI: n = 34, mean age = 50.8 months) enrolled in a longitudinal study. Participants with ASD-LI scored below 80 on the Preschool Language Scales (PLS-5). Expressive and receptive language were assessed using the PLS-5, while NLS data were transcribed using a modified Systematic Analysis of Language Transcripts (SALT) protocol.
Results:
Both groups exhibited similar challenges in expressive and receptive language. NLS measures, including intelligibility and lexical diversity, strongly correlated with standardized assessments (Spearman R = 0.64–0.83). While overall intelligibility was low (14.7% for DS; 30% for ASD-LI), language samples were far from silent, averaging 119.7 utterances (DS) and 106.4 (ASD-LI). Children with higher intelligibility scored higher on language assessments and showed greater lexical diversity.
Conclusion:
Despite a younger mean age, the DS group demonstrated similar communication challenges to the ASD-LI group. Low intelligibility likely impacts standardized language assessments, underscoring the need for transcription protocols that capture pre-speech and non-speech sounds to better assess communication and inform interventions in the DS population.
