Changing musical emotion: A computational rule system for modifying score and performance

Steven R. Livingstone, Ralf Muhlberger, Andrew R. Brown, William Forde Thompson

Research output: Contribution to journalArticleResearchpeer-review

55 Citations (Scopus)


CMERS system architecture has been implemented in the programming language scheme, and it uses the improvised music programming environment with the objective to provide researchers with a tool for testing the relationships between musical features and emotion. A music work represented in CMERS uses the music object hierarchy that is based on GTTM's grouping structure and is automatically generated from the phrase boundary markup and MIDI file. The Mode rule type of CMERS converts a note into those of the parallel mode and no change in pitch height occurs when converting to the parallel mode. It is reported that the odds of correctness with CMERS are approximately five times greater than that of DM. The repeated-measures analysis of variance for valence shows a significant difference between systems with F (1, 17) = 45.49, p < .0005 and the interaction between system and quadrant is significant with F (3, 51) = 4.23, p = .01, which indicates that CMERS is extensively more effective at correctly influencing valence than DM.

Original languageEnglish
Pages (from-to)41–64
JournalComputer Music Journal
Issue number1
Publication statusPublished - Mar 2010
Externally publishedYes


Dive into the research topics of 'Changing musical emotion: A computational rule system for modifying score and performance'. Together they form a unique fingerprint.

Cite this