Contributions of monotic intelligibility to dichotic performance

The contributions of monotic intelligibility to the right ear advantage for speech presented dichotically were examined in 3 experiments. The first experiment was mainly a replication of an earlier study by Speaks and Bissonette. Signal level in the right ear (RE) was attenuated and, separately, signal level in the left ear was amplified relative to 4 different reference intensities. Dichotic performance was not sensitive to absolute interaural-intensitive differences, but rather to something approaching an absolute sound pressure level. Thus, the RE advantage was affected only when signal level in one ear was sufficiently low (less than 60 dB SPL) to be near or below the "knee" of the monotic intelligibility function. That contention received support from the second experiment in which RE attenuation was introduced in both quiet and noise. Noise shifted the monotic intelliagibility function, and, as a consequence, less RE attenuation was required to cancel the RE advantage in noise than in quiet. The third experiment was conducted on a listener whose monotic intelligibility functions for the 2 ears were separated by 20 dB. The size and direction of the dichotic ear advantage, even though assessed at intensities that yielded 100% monotic intelligibility for both ears, depended on the proximity of the test intensity to the knees of the 2 monotic intelligibility functions. The results of all 3 experiments suggest that performance on a dichotic task does not involve central-level interactions exclusively.