Supplemental data for "Scanning the Dial"
In 2008, Elias Pampalk and I were lucky to be able to re-edit a printed version of David Perrott and Robert Gjerdingen's classic paper on "Scanning the dial", at the invitation of the Journal of New Music Research. The paper, presented in 1999 at the annual meeting of the Society for Music Perception and Cognition (SMPC),
in Northwestern University (Evanston, IL) had never been published before. We supply here supplemental material for the paper, courtesy of its authors.
Audio data for the original study:
400 audio samples, bundled in one file (.ZIP, 28.3MB)
Naming convention (.XLS file)
Preprint: R. Gjerdingen and D. Perrott, Scanning the Dial: The Rapid Recognition of Music Genres, Journal of New Music Research, 2008, Vol. 37, No. 2, pp. 93–100
History of the paper:
The following is an extract from our introduction to the 2008 re-edition:
J.-J. Aucouturier and E. Pampalk, Introduction: A Little Epistemology of Music
Information Retrieval Research, Journal of New Music Research, 2008, Vol. 37, No. 2, pp. 87–92
The paper is an extended version of ‘‘Scanning the dial: An exploration of factors in the identification of musical style’’, by David Perrott (currently Jury Consultant at Trialgraphix-Kroll Ontrack, New York) and Robert O. Gjerdingen (currently Professor of Music in Northwestern University). This paper never existed in print: it was presented in 1999 at the annual meeting of
the Society for Music Perception and Cognition (SMPC), in Northwestern University (Evanston, IL) on 14–17 August, and there weren’t any associated proceedings. [...]
The study on ‘‘Scanning the dial’’ is well known in the MIR community. It has been referred to very frequently as one of the few experiments measuring human ability to classify musical genre, thus providing a ground-truth to compare automatic genre classification algorithms with. Out of the 24 papers that appeared in the ISMIR conferences from 2000–2007 and included the keyword
‘‘genre’’ in their title, 10 quote ‘‘Scanning the dial’’. A Google search reveals more than 100 conference and journal articles including it as a reference. The first author of these words did quote it too himself, in some of his papers.
Yet, until today, it wasn’t possible to find a written account of this research, neither on the internet nor in the library. The only way for referencing authors to have had a direct encounter with these results is either by attending the 1999 SMPC meeting, or by personal communication with
the authors. There is little value in tracking down the plausibility of both options. A number of authors did indeed contact Gjerdingen and Perrott directly, as revealed by personal communication, but it appears that a probable majority of the works referencing the ‘‘scanning’’ study did so without ever possibly accessing it directly. In amusing support of this,many of the references are plagued with the same recurring typographical errors in the authors’ names.
Only 20% of the referencing ISMIR papers use the correct spelling. Googling ‘‘Gjerdigen (sic) scanning the dial’’ yields 69 hits against 37 with correct spelling. [...]
Contact:
We reproduce this data courtesy of the authors of this study. For all correspondence and scholarly use and references, please write to:
Robert O. Gjerdingen, 711 Elgin Rd., Evanston, IL 60208-1200, USA. E-mail: r-gjerdingen@northwestern.edu
