A comparison of two methods for the measurement of neural sensitivity to interaural intensity differences

Two alternative methods for the measurement of neural sensitivity to interaural intensity differences (IIDs) were used to obtain IID-sensitivity functions for samples of excitatory-inhibitory (EI) neurons from the central nucleus of the inferior colliculus and the primary auditory cortex of the cat. In one, the EMI-constant method, intensity was held constant in the ear providing excitatory input and varied above and below that level in the other ear. In the alternative (ABI-constant) method, intensity at the two ears was varied symmetrically about a constant base intensity, in a manner roughly approximating the pattern of changes that occur when a free-field stimulus is moved in azimuth from the median sagittal plane. For neurons with monotonic or near-monotonic rate-intensity functions for the excitatory ear, the two methods generated IID-sensitivity functions that were identical or near-identical by a number of quantitative criteria. For neurons with non-monotonic rate-intensity functions, however, the IID functions generated by the two methods were very different: those produced by the EMI-constant method were monotonic, whereas those generated by the ABI-constant method were non-monotonic and sharply peaked. The advantages and disadvantages of the two methods, and the implications of the results for the neural encoding of IIDs and for the azimuthal sensitivity of EI neurons with non-monotonic rate-intensity functions, are discussed.