Reading Group Meeting 14.01.2014 – Two papers on oscillator models for rhythm perception
by Emmanouil • January 13, 2014 • News, Reading Group
1) Velasco, M. J., & Large, E. W. (2011). Pulse Detection in Syncopated Rhythms using Neural Oscillators. In 12th International Society for Music Information Retrieval Conference (pp. 185–190).
Pulse and meter are remarkable in part because these perceived periodicities can arise from rhythmic stimuli that are not periodic. This phenomenon is most striking in syncopated rhythms, found in many genres of music, including music of non-Western cultures. In general, syncopated rhythms may have energy at frequencies that do not correspond to perceived pulse or meter, and perceived metrical frequencies that are weak or absent in the objective rhythmic stimulus. In this paper, we consider syncopated rhythms that contain little or no energy at the pulse frequency. We used 16 rhythms (3 simple, 13 syncopated) to test a model of pulse/meter perception based on nonlinear resonance, comparing the nonlinear resonance model with a linear analysis. Both models displayed the ability to differentiate between duple and triple meters, however, only the nonlinear model exhibited resonance at the pulse frequency for the most challenging syncopated rhythms. This result suggests that nonlinear resonance may provide a viable approach to pulse detection in syncopated rhythms.
2) Bååth, R., Lagerstedt, E., & Gärdenfors, P. (2013). An Oscillator Model of Categorical Rhythm Perception. In M. Knauff, M. Pauen, N. Sebanz, & I. Wachsmuth (Eds.), Proceedings of the 35th Annual Conference of the Cognitive Science Society (pp. 1803–1808). Austin, TX: Cognitive Science Society.
Categorical perception is a well studied phenomenon in, for example, colour perception, phonetics and music. In this article we implement a dynamical systems model of categorical rhythm perception based on the resonance theory of rhythm perception developed by Large (2010). This model is used to simulate the categorical choices of participants in two experiments of Desain and Honing (2003). The model is able to accurately replicate the experimental data. Our results supports that resonance theory is a viable model of rhythm perception and they show that by viewing rhythm perception as a dynamical system it is possible to model properties of categorical perception.