Sensory dissonance analysis
This experiment is based on the classic work by Plomp & Levelt, that made an important contribution to the modern psychoacoustic research by introducing the concept of sensory dissonance of pure sinewaves and its dependency on musical intervals. Sensory dissonance refers to the quality of a sound that is rough, unpleasant and unstable. Its dependency on musical interval has found its practical application in the work of William Sethares Tuning, Timbre, Spectrum, Scale. Sethares used results of Plomp and Levelt's experiment to calculate dissonance curves, which can be used to extract tuning from the timbre of musical instruments. He showed how dissonance curves predict most common musical chords, how they can be used to build dynamic tuning systems or to tune inharmonic instruments.
Below you can see a set of graphs that show that dependency of sensory dissonance vs musical interval for different registers. Each line on the graph is the result of a real person listening to intervals and rating how rought or smooth those intervals sound to them. You can take part in that experiment too! Sign in and find out your own dissonance preference and compare it with others. It takes only 5 minutes to complete and both musicians and non-musicians are welcome to participate.
Discrepancies with the original experiment
The results of Plomp and Levelt were not recieved without cotroversy because they deviated from the scientific and musical consensus at that time. That was not due to the mistake but because Plomp and Levelt intentionally modified their experiments in a way that was not done before:
- The experiment specifically targeted not musicians. All participants were people from the general public without any musical training.
- The set of intervals that were used in the experiment was not based on 12TET nor on simple ratios. This was important to exclude previous training or previous cultural experience from the results and get the raw sensory data.
- Another important aspect was that they used pure sinewaves to conduct listening experiments, that was done to exclude the role that timbre might play in the perception of dissonance.
- Finally, when rating the intervals, participants were asked to rate each interval on a 7-point scale of consonant - dissonant where the score of 7 is most consonant. The meaning of consonance interval was defined as "beautiful" and "euphonious".
In this experiment we attempt to replicate the approach of Plomp and Levelt. Because the experiments are not conducted in controlled laboratory environment, the results will not bear the same scientific rigor as the original research, but it is still interesting to see how it will compare. There are several differences between the original experiment and this one:
- Because the experiment is conducted online, we welcome participants with any musical background.
- We flip the 7-point scale to be more consistent with the work of Sethares. In our experiment the score of 7 is maximum dissonance.
- We never use the word "consonance" or "dissonance" in the experiment. Instead, we use terms like "smooth" and "rough" to describe the intervals. We believe that this is more suitable for broader audience that may include professional musicians as well as non-musicians, as their definitions and perceptions of consonance and dissonance may differ.
In other aspects the experiments are very similar. We hope that gathered data will help us to build new tools to enable microtonal experiments. But most of all we hope to educate people about the topic of sensory dissonance and its connection to the problem of musical tuning.
To follow the development and application of this experiment you can subscribe to our YouTube channel, consider following development on Github, and reach out if you have any questions or suggestions on email: