Evaluating an Objective Measure of Language in Minimally Verbal Autism: Automated Language ENvironment Analysis (LENA) in Phelan-McDermid Syndrome

Despite advances in the study and treatment of autism spectrum disorder (ASD), up to 30% of individuals with ASD remain minimally-verbal. Further research is hindered by the lack of an objective tool for measuring expressive language in this population. This study evaluates the use of an automated language processor called Language ENvironment Analysis (LENA) in Phelan-McDermid syndrome (PMS), a monogenic form of ASD characterized by severe language delays. LENA Software quantifies and analyzes language output and is able to count the number of times a child vocalizes during a given period and then filter out vegetative sounds such as respiration and fixed signals like cries. Approximately 542 hours of audio recording were collected using LENA in the home environment of 18 children with PMS. Selections of audio were evaluated by human transcribers and compared to LENA-generated transcriptions for agreement. LENA accuracy was adequate in a subset of children with PMS, specifically younger children and those with fewer stereotypic vocalizations. While LENA has previously been used in populations with ASD or language impairment alone, this study provides the first evaluation of its use in individuals with a single gene form of ASD and minimal verbal abilities.