The psychophysical detection threshold of a low-frequency tone masked by broadband noise is reduced by ≤ 15 dB by inversion of the tone in one ear (called the binaural masking level difference: BMLD). The contribution of 120 low-frequency neurons (best frequencies 168-2,090 Hz) in the inferior colliculus (ICC) of the guinea pig to binaural unmasking of 500-Hz tones masked by broadband noise was examined. We measured rate-level functions of the responses to identical signals (So) and noise (No) at the two ears (NoSo) and to identical noise but with the signal inverted at one ear (NoSπ): the noise was 7-15 dB suprathreshold. The masked threshold was estimated by the standard separation, 'D'. The neural BMLD was estimated as the difference between the masked thresholds for NoSo and NoSπ. The presence of So and Sπ tones was indicated by discharge rate increases in 55.3% of neurons. In 36.4% of neurons, the presence of So tones was indicated by an increase in discharge rate and Sπ tones by a decrease. In 6.8% of neurons, both So and Sπ tones caused a decrease in discharge rate. In only 1.5% of neurons was So indicated by a decrease and Sπ by an increase in discharge rate. Responses to the binaural configurations were consistent with the neuron's interaural delay sensitivities; 34.4% of neurons showing increases in discharge rate to both So and Sπ tones gave positive BMLDs ≤3 dB (Sπ tones were detected at lower levels than So), whereas 37.3% gave negative BMLDs ≤ dB. For neurons in which So signals caused an increase in the discharge rate and Sπ a decrease, 72.7% gave positive BMLDs ≤3 dB and only 4.5% gave negative BMLDs ≤3 dB. The results suggest that the responses of single ICC neurons are consistent with the psychophysical BMLDs for NoSo versus NoSπ at 500 Hz, and with current binaural interaction models based on coincidence detection. The neurons likely to contribute to the psychophysical BMLD are those with BFs near 500 Hz, but detection of So and Sπ tones may depend on different populations of neurons.
|Number of pages||22|
|Journal||Journal of Neurophysiology|
|Publication status||Published - 1997|