Frequency Specificity of Narrowband Chirp and 2-1-2 Stimuli: Spectral Analyses

PURPOSE: Narrowband (NB) chirps have recently been of great interest in audiology for the electrophysiological assessment of hearing in infants, but clinicians have little information about these new stimuli and how they compare to more standard stimuli, such as 2-1-2-cycle tones. The current study examined the frequency specificity of NB chirps by comparing the spectral characteristics of 500-, 1000-, 2000-, and 4000-Hz NB CE-Chirp LS ("level-specific" version) stimuli with those of the 2-1-2 tones. METHOD: Spectral characteristics, including center frequency (CF), bandwidth, and stimulus energy changes after stopband filtering, were compared. Bandwidth was computed as the difference between the upper and lower frequencies at the -20 and -3 dB cutoff points of the main lobe; CF was determined as the geometric mean of the upper and lower frequencies at the -20 and -3 dB cutoff points. RESULTS: At 100 dB peSPL, the bandwidths of the 500-, 1000-, and 2000-Hz NB CE-Chirp LS acoustic spectra were 1.8-1.9 times wider than those of the acoustic spectra for the 2-1-2 tones; the 4000-Hz NB CE-Chirp LS bandwidths were 1.4-1.5 times wider than those of the 2-1-2 tones. The energy of the NB CE-Chirp LS stimuli was concentrated within ±0.75 octave of the CF, compared to ±0.5 octave for 2-1-2 tones. These spectral characteristics did not change with stimulus intensity. CONCLUSIONS: The NB CE-Chirp LS stimuli demonstrated poorer frequency specificity compared with that of the 2-1-2 tones. NB-chirp thresholds may therefore underestimate true thresholds, especially in steeply sloping hearing loss. Further studies are needed to investigate the cochlear place specificity of auditory brainstem responses to the NB CE-Chirp LS stimuli before recommending them for clinical use.