Context-specific Knowledge Is the “Key” to Salsa Music

ABSTRACT

Introduction

Previous research has shown ways in which both formal training and informal exposure affect perceptual experience and the development of musical abilities. Here we asked what types of training and exposure are necessary to acquire the context-specific knowledge associated with expertise. We specifically focused on the perception of salsa music: a genre that is rich in rhythmic complexity, but has received relatively little attention in experimental settings.

Methods

We examined specific groups within the exposure and training populations: those with musical training in the production of salsa rhythms (Study 1) and “native listeners” who grew up listening to salsa music without formal training (Study 2). Using two clave patterns (3–2 and 2–3 son clave) and three constructed alternatives, we asked participants to choose the correct clave pattern for a variety of music excerpts.

Results

We found that informal listening exposure was not enough to detect the salsa–clave pairings. Instead, proficiency was only developed when training and exposure were both domain-specific.

Discussion

Our results show the importance of deliberate training and the degree to which expertise comes to fruition through context-specific focus, thus helping to illuminate the complex relationship between the local and the universal in musical-cultural experience.

KEYWORDS:

• Cross-cultural
• expertise
• music perception
• music training
• rhythm
• syncopation

Acknowledgments

We would like to thank Michael Kubovy for helpful comments on our research design and interpretation, Robert Jospé for allowing his “Learn to Groove” students to take part in our experiment, and research assistants in the Object and Word Learning Lab at the University of Miami for running the “native” salsa listeners. The experiments were designed and data for Study 1 were collected while LMG was a graduate student at University of Virginia, and data for Study 2 were collected when LMG was a postdoc at Villanova University funded by a Mendel Science Experience Postdoctoral Fellowship.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Log-linear analyses model cell counts from contingency tables, thus helping to understand the relationship between levels of categorical variables without one factor being the independent and the other being a dependent variable. Degrees of freedom are based on n-1 for whichever factor has the most levels. Thus, in our analyses, five response categories would lead to df = 4. If we only compare correct vs. incorrect responses, the two categories would lead to df = 1.

2. In Study 1 Session 2, 3 of the 6 training participants chose only natural patterns and the other 3 chose natural patterns at least 75% of the time. In the no training group, only 3 participants chose natural patterns at least 67% of the time; the rest were at chance. In Study 2, 10 of the 23 native listeners showed chance performance, with the other 13 choosing the natural patterns at least 50% of the time (and only 4 choosing natural patterns at least 75% of the time.)