Development and validation of the pediatric AzBio sentence lists

Area of Science:

• Audiology
• Speech-Language Pathology
• Hearing Science

Background:

• Evaluating speech perception in pediatric populations with hearing loss presents unique challenges.
• Existing adult-oriented materials may be inappropriate in difficulty or content for children.
• A need exists for validated, equivalent sentence lists for pediatric speech perception assessment.

Purpose of the Study:

• To create and validate a new set of equivalent sentence lists for assessing speech perception in listeners with hearing loss.
• To ensure lists are appropriate for pediatric use and suitable for adults with severe hearing impairment.
• To enable reliable evaluation of speech-understanding performance over time and across conditions.

Main Methods:

• Developed 16 equivalent sentence lists (20 sentences each) from the Pediatric AzBio corpus using cochlear implant simulation for intelligibility estimation.
• Validated list equivalency by testing adults and children with hearing loss.
• Employed a binomial distribution model to account for list variability and generate confidence intervals.

Main Results:

• Created 16 lists with average intelligibility ranging from 78.4% to 78.7%.
• Validation testing showed no significant differences in performance across lists for adult and pediatric cochlear implant users.
• A binomial model accurately predicted performance, capturing 94% of responses within 95% confidence intervals.

Conclusions:

• Cochlear implant simulation effectively estimates sentence intelligibility for evaluating hearing aid and cochlear implant users.
• The developed method for constructing equivalent lists and estimating variability is validated.
• The AzBio Pediatric Sentence Lists are validated as equivalent and appropriate for assessing speech understanding in children and adults with hearing loss.