Objective: To improve understanding of normal responses in infants by comparing air conduction (AC) and bone conduction (BC) auditory thresholds using both the auditory steady state response (ASSR) and behavioral testing methods in normal-hearing infants (6 to 18 months of age) and adults. At present, there are no correction factors available for estimating BC behavioral thresholds from BC ASSR thresholds, which is a barrier to clinical implementation of the ASSR. In addition, previous studies have reported infant-adult differences in AC and BC sensitivity, which suggest a "maturational" air-bone gap (ABG) that is not attributable to a conductive pathology; no study has yet compared AC and BC thresholds for either ASSR or behavioral methods in the same individuals. The objectives of the present study are: (1) to compare BC thresholds between methods and provide the initial step toward positing correction factors to predict BC behavioral thresholds, (2) to directly compare AC and BC thresholds to provide an accurate estimate of the maturational ABG, (3) to determine preliminary normal levels for BC and AC ASSRs to exponentially amplitude modulated stimuli, and (4) to investigate infant-adult differences in AC and BC thresholds using ASSRs and behavioral assessment tools.
Design: Participants were 23 infants (6.5 to 19.0 months of age) and 12 adults (17 to 50 years of age) with normal hearing. Thresholds were estimated at 500, 1000, 2000, and 4000 Hz using air- and bone-conducted stimuli for ASSRs and behavioral testing. The ASSR stimuli were exponential envelope modulated (amplitude modulation [AM]) at modulation frequencies of 78, 85, 93, and 101 Hz for 500, 1000, 2000, and 4000 Hz, respectively, presented simultaneously. Frequency-modulated (warble tone) stimuli were used for behavioral testing for both infants and adults, respectively. All stimuli were calibrated in dB HL. Thresholds were compared across frequency and between stimulus presentation modes, between age groups and assessment method. Normal levels for AC and BC ASSRs to AM stimuli were also calculated.
Results: The findings indicated that BC thresholds were, on average, 7 to 16 dB poorer for ASSR compared with visual reinforcement audiometry (VRA), but varied widely across infants. For infants, mean ABGs of 14 to 17 dB were found for low-frequency ASSR thresholds but mean ABGs for VRA thresholds were less than 10 dB. The preliminary normal levels for ASSR AM stimuli at 500, 1000, 2000, and 4000 Hz, respectively, were: (i) AC: 30, 30, 20, and 20 dB HL, and (ii) BC: 20, 20, 30, and 30 dB HL. There was a tendency for infant and adult ASSR thresholds to differ for BC, but not for AC. Behavioral thresholds for AC and BC were similar between infants and adults and across frequency.
Conclusions: Infant-adult and AC-BC threshold differences are greater for ASSRs compared with behavioral measures. The results support the presence of a clinically significant maturational ABG in the low frequencies for infant ASSRs but not for VRA. The findings also show a significant offset between BC ASSR and BC VRA thresholds and large intersubject variability.