Measuring speech recognition in children with cochlear implants in a virtual classroom.

Purpose: To determine the feasibility of using a virtual auditory test material to evaluate reverberation and noise effects on speech recognition of pediatric cochlear implant (CI) users and to compare their performance with that of children with normal hearing.

Method: Virtual test materials representing nonreverberant and reverberant environments were used to measure speech recognition of 7 children with CIs in quiet and in noise, and of 18 children with normal hearing in the quiet condition. Performance of CI users in noise (signal-to-noise ratio resulting in 50% performance) was compared to normative data from a previous study (Neuman, Wroblewski, Hajicek, & Rubinstein, 2010). For CI users, stimuli were sent directly to the CI speech processor via auxiliary input, whereas children with normal hearing were tested using insert phones.

Results: The speech recognition of children with CIs decreased significantly in the reverberant condition. There were individual differences in susceptibility to reverberation. Children with CIs also required higher signal-to-noise ratios than children with normal hearing in the reverberant condition.

Conclusion: Direct connect testing with reverberant test materials allows assessment of speech recognition under conditions typical of classrooms and could be useful in identifying children with CIs whose performance decreases significantly in the presence of reverberation and noise.