Validating Motor Skills as Early Autism Indicators
Overview
The Infant Brain Imaging Study (IBIS) Network, a group of researchers and clinicians conducting longitudinal studies across the U.S., was interested in understanding the motor profiles of young children. Specifically, they wanted to validate whether motor skills of their participants could serve as early indicators of autism.
Why? Early identification of autism through motor skills is crucial for timely interventions, which can improve life outcomes for children. This research aims to fill the gap in understanding how motor skills develop in young children under 3 years old and whether they can predict autism.
Impact: Understanding early motor development in children is key to identifying early indicators of autism, which can lead to timely interventions and better developmental outcomes.
My Role: As the Lead Researcher, I was responsible for designing the research question, analyzing the data, and presenting the results.
Stakeholders
The IBIS Network is a collaborative group of researchers and clinicians conducting ongoing longitudinal studies across the U.S. to better understand early signs of autism. Their research involves large and complex datasets, including behavioral assessments, parent-report surveys, and MRI scans.
IBIS Network Map
Research process
I worked closely with stakeholders from the IBIS Network, including clinicians and researchers, to define our research objectives. Together, we wanted to validate whether motor skills could serve as early indicators of autism in children under 3 years old.
After discussions, we chose online parent-report surveys for their ability to reach a larger sample size, easy for participants to complete, and ability to capture a broader range of motor data. This decision aligned with both research goals and practical clinical needs.
During the recruitment phase, I identified several pain points in the survey design, including overly long surveys, incorrect scoring, and the ability for participants to skip key questions. To improve data quality and participant experience, I refined the survey by splitting it into shorter sections, correcting the scoring system, and making key questions unskippable. These refinements enhanced user engagement and increased survey completion rates.
Regular meetings with IBIS stakeholders helped refine the objectives and ensured the data collected was relevant for both clinical and research applications.
Literature Review
Autism is a developmental disability characterized by social-communication challenges, repetitive behaviors, and restricted interests. While most children are diagnosed after the age of 3, research shows that signs of autism can often appear earlier.
Motor skills play a crucial role in early childhood development, supporting language, cognitive, and overall growth. Early identification of autism, particularly through motor skill assessments, can significantly improve the effectiveness of interventions, potentially leading to better outcomes in education, language development, and life trajectories for autistic children.
Hypothesis:
Research has shown that young autistic children often face challenges in motor skills. This project aims to validate whether motor skill difficulties in young children are early indicators of autism, using data from the IBIS network.
Goal: To determine if motor skills are associated with early signs of autism before the age of 3.
Hypothesis: We hypothesize that younger autistic children will exhibit lower motor skills compared to their non-autistic peers.
Methods
Examples of questions
We conducted an online parent-report survey to assess children's motor development.
We recruited parents of children under 2 years old across the United States. These children either had an older autistic sibling or an older non-autistic sibling.
The survey was administered when the children were 12, 18, and 24 months old.
Tools
I used R for descriptive statistics to analyze participant demographics and applied linear mixed models to compare developmental differences across groups. Linear mixed models were chosen because they accommodate repeated measures and allow for the inclusion of covariates, controlling for external factors.
In both general research and UX research contexts, these skills are directly transferable. Whether analyzing user data from surveys, usability tests, or behavioral studies, I can identify trends and control for external variables to derive meaningful insights. This helps inform research-driven recommendations, optimize user experience, and support evidence-based decision-making.
This approach was ideal for our research as it aligned with our objectives, was straightforward to use, and provided easily interpretable results that guided actionable insights.
Challenges
One major challenge was missing data, as some parents skipped survey questions that were crucial for analysis. Initially, participants were allowed to skip questions, which affected the completeness of the data. After consulting with statisticians, I decided to exclude incomplete surveys from the analysis and modified the survey design to prevent skipping key questions.
I also found some surveys were scored incorrectly, so I re-scored them to ensure data accuracy. Additionally, the survey was too long, causing participant fatigue. To improve completion rates and user experience, I redesigned the survey by breaking it into shorter sections.
For the statistical analysis, I consulted with statisticians and confirmed that linear mixed models were the most suitable and straightforward approach for our data.
Outcomes
Our research demonstrated that autistic children in the first two years of life exhibit lower motor skills compared to their non-autistic peers. The parent-report surveys successfully captured these motor differences early on, even before a formal autism diagnosis.
We also validated that IBIS participants who were later diagnosed with autism consistently showed lower motor skills during early development.
These findings were published in Autism Research (2024), further validating our work and contributing to the broader research, clinical, and autism communities. Read our full story here.
Impact
Our research has influenced early autism evaluations, shaping clinical practices for clinicians and pediatricians. Published in Autism Research (2024), the study highlights motor skills as early indicators of autism, enabling earlier and more effective interventions.
To improve data quality and participant experience, we redesigned the parent-report surveys by shortening them, preventing skipped questions, and correcting scoring issues. These improvements enhanced IBIS’s ongoing data collection, ensuring more accurate and actionable insights.
By incorporating motor assessments, clinicians can now identify autism sooner, leading to improved developmental outcomes. We also proposed standardized online surveys to help parents detect early signs and seek timely evaluations.
This research has expanded IBIS’s focus on early autism identification and contributed to further studies, including those using MRI data. It also played a key role in securing the renewal of multi-million-dollar research grants for the IBIS network.
Our publication as shown: https://pubmed.ncbi.nlm.nih.gov/38204321/
Interested in learning more about my research or collaborating? Feel free to contact me or check out my other projects.