Using Mobile Technology and Real-World Vocalization Samples to Generate Quantitative Metrics of Vocal Communication for Minimally-Speaking Individuals - Using Mobile Technology and Real-World Vocalization Samples to Generate Quantitative Metrics of Vocal Communication for Minimally-Speaking Individuals Over one million individuals remain minimally-speaking past the age of 5, often communicating through vocalizations that contain little to no speech. Yet, little is known about the vocal communication of profoundly affected minimally-speaking individuals. Current speech and communication metrics are often normalized based on typically-developing individuals and fail to adequately capture the skills of minimally-speaking individuals. Moreover, measures that are designed for this population are generally qualitative, relying on caregiver reports or examiner observations from a short clinical visit. The lack of quantitative communication measures that map how vocalizations from minimally-speaking individuals develop over time has profound consequences for researchers, clinicians, families, and the individuals themselves. The lack of quantitative communication measures designed for this population has profound consequences for researchers, clinicians, families, and the individuals: Clinical trials have failed to advance because the outcome measures did not capture the subtle changes in participants’ development, and families of minimally-speaking individuals consistently list communication as one of their top research priorities. This proposal aims to generate a sensitive, quantitative metric of vocal communication for minimally-speaking individuals by tracking their phonological vocalization complexity over time. To achieve this aim, I will enroll n=90 individuals (ages 6-35 years, enrolled equally over the age range) with fewer than 50 spoken words using a smartphone app I previously developed and deployed. This app allows caregivers to record a vocalization and label its meaning with the tap of a button in their homes and natural environments. Caregivers will be prompted to elicit and record as many vocalizations as possible during daily ten-minute intervals for one week every six months for two years. Using this unique database of real-world vocalizations, I will characterize the complexity of vocal communication across and within individuals over time (Aims 1 and 2) and will examine how this metric relates to established assessments (Aim 3). I will also explore acoustic-based metrics that capture vocalization complexity in an objective and scalable manner. The results of this work will allow us to determine sensitive vocal communication milestones for profoundly affected minimally-speaking individuals and appropriate outcome measures for clinical trials and interventions.
