Sounds from the North

Differentiating musicians using a fast, musical multi-feature paradigm

Posted on 17. Feb, 2010 by admin in Publications

This is a poster at the MMN 09 Fifth Conference on Mismatch Negativity (MMN) and its Clinical and Scientific Applications. Budapest, Hungary, April 04 – 07,  2009. This fast multi-feature method was designed by the people of the CBRU at Helsinki University. I contributed to the paper by analysing the collected EEG data using standard MMN analysis and intersubject correlation of the ERP signals to identify possible clusters. The code will be released on a different post. The article is currently submitted.

Differentiating musicians using a fast, musical multi-feature paradigm

Peter Vuust1,2*, Elvira Brattico3,4, Miia Seppänen3,4, S Pakarinen3,4, R Näätänen3,5,2, E Glerean3,4 and M Tervaniemi3,4

1. Royal Academy of Music, Aarhus, Denmark
2. Center of Functionally Integrative Neuroscience, Aarhus, Denmark
3. Cognitive Brain Research Unit, Department of Psychology, University of Helsinki, Helsinki, Finland
4. Finnish Centre of Excellence in Interdisciplinary Music Research, University of Jyväskylä, Jyväskylä, Finland
5. Department of Psychology, University of Tartu, Tartu, Estonia

The Mismatch negativity (MMN) is a component of the auditory event-related potential (ERP) related to change in different sound features such as pitch, timbre, location, intensity, and rhythm. Importantly, the MMN is sensitive to discrimination learning and to musical expertise. Recently it has been shown that dependent on the performance and practice strategies, musicians develop enhanced auditory abilities reflected in shorter latency of the MMN to contour changes when they employ aural performance and practice strategies. Additionally, MMN findings have revealed that fine pitch changes are automatically detected by violinists only and that localization changes are more efficiently processed by band musicians than control subjects. These results suggest a sensitivity of the MMN to acoustic features specific of a musical genre or musical practice. The traditional MMN paradigms used so far, however, are both time consuming and far away from sounding musical.

Here, we present a fast multi-feature MMN paradigm, in which six types of acoustic changes relevant for musical processing in different musical genres are presented in the same sound sequence which tackles the two challenges mentioned above. Specifically, the six musical features studied were: pitch mistuning, intensity, timbre, sound-source location, rhythm and a pitch slide typical for improvisational music. In contrast to the recently developed multi-feature paradigm, the present paradigm is closer to real music. It uses a musical figure, well-known in many genres of music, played with sample piano sounds: the Alberti bass, an accompaniment originally encountered in classical music such as Mozart’s sonatas or Beethoven’s rondos but later adopted in other contemporary musical genres. Using this paradigm, the length of the experiment is reduced to about 20 minutes, and the sequences sound like chord sequences (arpeggios).

We measured ERPs in subjects classified in 3 groups according to their main genre of performance and education: jazz musicians, rock, and classical. In addition, a group of non-musicians was also measured. We found MMNs to most of the 6 deviant musical features as well as differences between the subject groups. Interestingly, these differences confirmed our expectation that the MMN reflects the auditory skills specific to a musical genre.

This shows that auditory pattern change influences pre-attentive perception of musical notes, not just sound, differently in different musical populations. This study provides a novel way of investigating pre-attentive neural mechanisms involved during online music listening.

Tags: EEG, matlab, MMN, music