Optimizing the stimuli to evoke the amplitude modulation following response (AMFR) in neonates

The goal was to identify stimulus features that enhance the detection of the amplitude modulation following response (AMFR) in neonates. The features explored were (1) envelope type, sinusoidal versus a half-wave rectified sinusoid (transposed); (2) best modulation frequency; and (3) spectral content, i.e., tone versus band-pass noise.

Results are based on recordings from 149 babies (80 babies in the neonatal intensive care unit and 69 newborn infants). All had passed hearing screening based on the click-evoked ABR. Babies were not sedated. We used carrier frequencies of approximately 500, 1000, 2000, and 4000 Hz and modulation frequencies between approximately 25 and 98 Hz. For the noise stimuli, we used band-pass noise at center frequencies of 500, 1000, 2000, and 4000 Hz. All stimuli were presented through insert earphones delivered simultaneously to both ears, at intensities ranging from 20 to 70 dB SPL. Magnitude squared coherence, phase coherence, and spectral criteria were used to detect criterion AMFRs. We analyzed four measures: (1) percent of satisfied runs; (2) the amplitude of criterion AMFR; (3) time to detect a criterion AMFR; and (4) response strength (e.g., the value of the magnitude squared coherence when it reached criterion minus the critical value it had to exceed for that number of averages all divided by the critical value).

(1) The AMFRs evoked by transposed tones were larger and detected faster than those to sinusoidal amplitude modulated tones. Consequently, remaining protocols all used the transposed envelopes. (2) The range of effective modulation frequencies was broad (41 to 88 Hz) across carrier frequencies. (3) The AMFRs evoked by transposed noise were faster and more efficient than those to transposed tones.

In neonates, transposed tones are more effective than sinusoidal amplitude modulated tones in evoking the AMFR, modulation frequencies between 41 to 88 Hz are almost equally effective in evoking the AMFR, and band-pass noises are more effective in evoking the AMFR than tones. These three stimulus factors all add incrementally to the efficiency of evoking the AMFR. The short detection times indicate that the AMFR could be an effective tool for hearing screening.