Fitting Frequency-Lowering Signal Processing Applying the American Academy of Audiology Pediatric Amplification Guideline: Updates and Protocols.

BACKGROUND: Although guidelines for fitting hearing aids for children are well developed and have strong basis in evidence, specific protocols for fitting and verifying technologies can supplement such guidelines. One such technology is frequency-lowering signal processing. Children require access to a broad bandwidth of speech to detect and use all phonemes including female /s/. When access through conventional amplification is not possible, the use of frequency-lowering signal processing may be considered as a means to overcome limitations. Fitting and verification protocols are needed to better define candidacy determination and options for assessing and fine tuning frequency-lowering signal processing for individuals. PURPOSE: This work aims to (1) describe a set of calibrated phonemes that can be used to characterize the variation in different brands of frequency-lowering processors in hearing aids and the verification with these signals and (2) determine whether verification with these signal are predictive of perceptual changes associated with changes in the strength of frequency-lowering signal processing. Finally, we aimed to develop a fitting protocol for use in pediatric clinical practice. STUDY SAMPLE: Study 1 used a sample of six hearing aids spanning four types of frequency lowering algorithms for an electroacoustic evaluation. Study 2 included 21 adults who had hearing loss (mean age 66 yr). DATA COLLECTION AND ANALYSIS: Simulated fricatives were designed to mimic the level and frequency shape of female fricatives extracted from two sources of speech. These signals were used to verify the frequency-lowering effects of four distinct types of frequency-lowering signal processors available in commercial hearing aids, and verification measures were compared to extracted fricatives made in a reference system. In a second study, the simulated fricatives were used within a probe microphone measurement system to verify a wide range of frequency compression settings in a commercial hearing aid, and 27 adult listeners were tested at each setting. The relation between the hearing aid verification measures and the listener's ability to detect and discriminate between fricatives was examined. RESULTS: Verification measures made with the simulated fricatives agreed to within 4 dB, on average, and tended to mimic the frequency response shape of fricatives presented in a running speech context. Some processors showed a greater aided response level for fricatives in running speech than fricatives presented in isolation. Results with listeners indicated that verified settings that provided a positive sensation level of /s/ and that maximized the frequency difference between /s/ and /∫/ tended to have the best performance. CONCLUSIONS: Frequency-lowering signal processors have measureable effects on the high-frequency fricative content of speech, particularly female /s/. It is possible to measure these effects either with a simple strategy that presents an isolated simulated fricative and measures the aided frequency response or with a more complex system that extracts fricatives from running speech. For some processors, a more accurate result may be achieved with a running speech system. In listeners, the aided frequency location and sensation level of fricatives may be helpful in predicting whether a specific hearing aid fitting, with or without frequency-lowering, will support access to the fricatives of speech.

Effects of semantic and acoustic context on nonword detection in children with hearing loss.

PURPOSE: Children with hearing loss (HL) are known to have smaller receptive vocabularies than children with normal hearing (NH). This may be due, in part, of their reduced exposure to new words and their slower rate of word learning. A necessary prerequisite to lexical development is the detection of new words in conversation. The purpose of this study was to examine the effects of HL on children's ability to detect the presence of nonwords within sentences that varied in semantic and acoustic context. METHODS: Twenty-nine children with NH and 16 children with HL between the ages of 7 and 13 years participated. The children listened to short sentences and reported the number of nonwords detected, ranging from zero to two nonwords, in each sentence. The structure of the sentences was either meaningful or nonsensical to the children to reveal the effects of semantic context. The effects of acoustic context were revealed by presenting the sentences in quiet, steady-state noise, and in multi-talker babble. RESULTS: Significant effects of age (older > younger), hearing (NH > HL), and listening condition (quiet > noise and babble) were observed. Also, nonword detection was better for semantically meaningful sentences than for nonsense sentences. Error analyses revealed that the children with NH tended to underestimate the number of nonwords in meaningful sentences but not in nonsense sentences. The children with HL, however, were more likely to underestimate the number of nonwords than were the children with NH for both meaningful and nonsense sentences. These error patterns were observed in each listening condition. CONCLUSIONS: Error patterns suggest that children with HL apply strong repair strategies during speech perception, which may limit their opportunities to learn new words.