Yuan et al. (2020)
Contents
Source Details
Yuan et al. (2020) | |
Title: | Perceptual vs. Automated Judgements of Music Copyright Infringement |
Author(s): | Yuan, Y., Oishi, S., Cronin, C., Müllensiefen, C., Atkinson, Q., Fujii, S., Savage, P. |
Year: | 2020 |
Citation: | Yuan, Y., Oishi, S., Cronin, C., Müllensiefen, D., Atkinson, Q., Fujii, S., & Savage, P. E. (2020) Perceptual vs. automated judgments of music copyright infringement. PsyAriXiv preprints. Available: https://doi.org/10.31234/osf.io/tq7v5 |
Link(s): | Open Access |
Key Related Studies: | |
Discipline: | |
Linked by: |
About the Data | |
Data Description: | The study uses 17 court decisions where substantial similarity of melody was in question (14 from the US and 3 from Japan). Thereafter, the researchers conducted an online perceptual experiment where 20 participants judged for themselves the substantial similarity of the melodies in question. Following this, the researchers compared the participants perceptions with the results of automatic analysis using two different algorithms. |
Data Type: | Primary and Secondary data |
Secondary Data Sources: | |
Data Collection Methods: | |
Data Analysis Methods: | |
Industry(ies): | |
Country(ies): | |
Cross Country Study?: | No |
Comparative Study?: | No |
Literature review?: | No |
Government or policy study?: | No |
Time Period(s) of Collection: | |
Funder(s): |
|
Abstract
“Music copyright lawsuits often result in multimillion dollar settlements, yet there are few objective guidelines for applying copyright law in infringement claims involving musical works. Recent research has attempted to develop objective methods based on automated similarity algorithms, but there remains almost no data on the role of perceived similarity in music copyright decisions despite its crucial role in copyright law. We collected perceptual data from 20 participants for 17 past copyright cases from the USA and Japan after editing the disputed sections to contain either full audio, melody only, or lyrics only. Due to the historical emphasis in legal opinions on melody as the key criterion for deciding infringement, we predicted that listening to melody-only versions would result in perceptual judgments that more closely matched actual past legal decisions. Surprisingly, however, we found no significant differences between the three conditions, with participants matching past decisions in between 50-60% of cases in all three conditions. Automated algorithms designed to calculate melodic and audio similarity produced comparable results: both algorithms were able to match past decisions with identical accuracy of 71% (12/17 cases). Analysis of cases that were difficult to classify suggests that melody, lyrics, and other factors sometimes interact in complex ways difficult to capture using quantitative metrics. We propose directions for further investigation of the role of similarity in music copyright law using larger samples of cases and enhanced methods, including some developed for purposes of cover-song detection. Our results contribute to important practical debates, such as whether jury members should be allowed to listen to full audio recordings during copyright cases.”
Main Results of the Study
More than half of the human participants reported a substantial music experience in the cases surveyed. Participant perceptions matched court opinions across full-audio, music-only and lyrics-at rates of 58%, 54% and 49% respectively. By contrast, both of the automated algorithms used matched 12 out of 17 judgements (71%), suggesting a high level of accuracy. The study suggests that the lower levels of human participant accuracy may be as a result of the controversial nature of some of the cases observed (e.g. ‘He’s So Fine’ and ‘Blurred Lines’).
Overall, melodic similarity was found to play the most significant role in perceptions of overall music similarity. Where melody is listened to in isolation, participants have more difficulty in detecting infringement. Instead, full audio recordings improve accuracy of judgement, suggesting that such recordings, rather than sheet music, may be more helpful in legal cases.
Policy Implications as Stated By Author
The study does not make any explicit policy recommendations, instead suggesting that algorithms should not replace human judgements in substantial similarity cases.
Coverage of Study
Datasets
Sample size: | 20 |
Level of aggregation: | Individuals |
Period of material under study: |
Sample size: | 17 |
Level of aggregation: | Court cases |
Period of material under study: |