The study included six experiments on responses to harmonic and melodic intervals, using two complementary methods: event-related potentials (ERP) and verbal responses. Three experiments were performed with each method, using the same musical material: (1) isolated harmonic intervals (m 2nd, m 3rd, p 5th); (2) nine pairs of melodic intervals comprising combinations of three intervals (prime, ascending 2nd, ascending M 6th); (3) 27 pairs of harmonic intervals: nine combinations of the three intervals from Experiment 1, and the three intervals from Experiment 2. All subjects in the ERP experiments were knowledgeable in music, whereas in the verbal experiments, subjects varied in their cultural background (Western or Arab), age, and musical knowledge. The results showed specific responses to intervals, even without context, indicating that intervals may be viewed as meaningful words, even when presented in isolation. Furthermore, the results of comparisons within and between experiments confirmed earlier assumptions on the contribution of musical elements to tension or excitement, as opposed to relaxation, and underscored the importance of cognitive principles such as the "inverted U function," contrast, levels of consciousness, and hierarchy.
This study investigates the possibility of measuring cognitive responses to musical stimuli. The experiments were based on measurements of the event- related potential (ERP) of three electroencephalographic electrodes. The musical stimuli consisted of five-tone pitch patterns ( constant intensity, duration, and timbre), based on the Western tonal system. Subjects compared reference patterns with comparison patterns, which were either identical to the reference pattern or nonidentical because of predetermined changes in the comparison patterns. The patterns were presented in auditory or visual mode in a slow or a fast version. The results show a striking cognitive response to nonidentical tones in the comparison patterns and to a lesser extent to exceptional tones even in the reference patterns. Different levels of response were detected according to the type of pattern. We also found some evidence of factors contributing to "subjective equivalence" between different patterns. The correlation between results from this study and those from earlier studies based on the same measuring technique with different kinds of input data or using different methods and techniques is discussed.