Development of a methodology and theoretical framework for melodic discrimination

Abstract

The ability to discriminate two melodies as differing or identical is arguably the most basic musical task, yet modern research on melody discrimination has stagnated without reaching psychologically important conclusions. I reviewed previous research and identified the barriers to progress as the lack of a sophisticated theoretical framework, and inadequate experimental and data-analytic methodologies. My work addresses these barriers.The theoretical framework proposed distinguishes between the attributes describing a melody and the units in which those attributes are organised. This makes clear the need to ensure that corresponding units are compared in the comparison of melodies.The characteristics of the melodic domain make traditional experimental designs and analyses particularly inappropriate. Discriminability measures could only be calculated, analysed, and interpreted in terms of the summary properties of groups of melody pairs rather than the properties of the individual melodies. I developed an extension to signal detection analysis that allows the discriminability of individual melody pairs to be calculated and analysed in terms of the individual melodies involved.The first experiment was designed as an exploratory study to find previously unreported effects. It demonstrated effects related to contour reversal, overall contour shape, particular pitch transitions, pitch range, and cadence.The second and third experiments further investigated the contour reversal effect. This effect was convincingly demonstrated in these experiments while I developed my methods of data analysis. The third experiment also demonstrated effects associated with the repetition of a note across a contour reversal and with rhythmic stress.In the fourth experiment I used the new methodology to attack directly the question of the relative importance of interval-based and note-based attributes. This experiment demonstrated that simple measures of the number of notes and intervals changed are inadequate to account for melody discrimination performance. I found effects related to: repetition across a contour reversal, the tonic, the major triad, the commonness of particular pitch transitions, and the process of matching corresponding units. These effects were more naturally described in terms of note-based attributes than interval-based attributes.