Conference proceeding
Learning the Language of Affect: A New Model That Predicts Perceived Affect Across Beethoven's Piano Sonatas Using Music-Theoretic Parameters
Proceedings of the ICMPC-SMPC 2016 Joint Conference, pp.79-85
2016
Abstract
This paper examines affective expression in Beethoven's piano sonatas from a structural perspective by building multiple regression models using music-theoretic parameters as predictors of perceived affective expression. Thirty 5-second excerpts from the Beethoven piano sonatas were selected by polling musicians for those moments that were the most emotionally expressive in the repertoire. These excerpts were divided into a training set and a test set, matched by compositional period and inner/outer movement. 68 participants were randomly divided into one of the two groups in which they listened to each excerpt in that group and rated each on 9 affective dimensions. Each excerpt was analyzed for 14 different musical parameters, used as predictors in a regression model for each affective dimension. The models from the training set were applied to the same musical parameters for excerpts of the test set as a way of predicting participant ratings of affect. These predictions were tested against actual participant ratings as a way to test the generalizability of the models. Of the nine affective dimension models, six produced predictions that were significantly positively correlated with mean participant ratings in the test set, but all nine models were significantly better at predicting ratings than chance. The regression models were able to account for between 9-71% (Mean = 34.9%) of the variance in participant response. However, all nine reverse models, generated from the test set and tested on ratings from the training set were significantly positively correlated with mean participant ratings, and all were likewise better at predicting ratings than chance.
Details
- Title: Subtitle
- Learning the Language of Affect: A New Model That Predicts Perceived Affect Across Beethoven's Piano Sonatas Using Music-Theoretic Parameters
- Creators
- Joshua Albrecht - University of Mary Hardin–Baylor
- Resource Type
- Conference proceeding
- Publication Details
- Proceedings of the ICMPC-SMPC 2016 Joint Conference, pp.79-85
- Publisher
- San Francisco, CA
- Copyright
- Copyright © 2016 the author
- Language
- English
- Date published
- 2016
- Academic Unit
- School of Music
- Record Identifier
- 9985219828202771
Metrics
1 File views/ downloads
1 Record Views