Masking Effects of Speech Competing Messages

Initially two experiments were performed on eight subjects to determine the effect of semantic content or meaning of a competing speech message on the identification of synthetic sentence material. The competing message was reversed and reproduced in a backward mode in an attempt to eliminate its semantic content or meaning. No important differences were found between the performance-intensity functions in the forward or backward mode. The results of 2 additional experiments with 12 subjects demonstrated the presence of a plateau or notch in the slope of the performance-intensity functions obtained in English or foreign language competing messages. This plateau effect occurred primarily when three experimental conditions were present: (1) the same speaker delivered the primary and competing messages, (2) the relative intensity levels of both the primary and competing messages were equal, (3) the temporal pattern of the competition was similar to that of the primary message. The semantic content of the competing message did not contribute to the plateau in the slope of the performance-intensity function, but the plateau was primarily due to the similarity of the temporal patterns of the primary and competing messages which occurred when the intensities of both messages were equal.