On popular music and media: Analyzing changes in compositional practices and music listening choice behavior using attention economy principles

LeveilleGauvinHubert_Dissertation2018.pdfOn popular music and media: Analyzing changes in compositional practices and music listening choice behavior
using attention economy principles
Dissertation
Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State
University
By
Graduate Program in Music
The Ohio State University
listening choice behavior in studies that attempt to uncover whether technological
changes have had an impact on songwriting techniques, and whether music listening
choice behavior can be predicted by these compositional devices. The underlying
theory unifying this work is the theory of attention economy: how attention can
be analyzed using supply and demand principles to explain and predict information
content and human behavior.
Two corpus studies were conducted to evaluate whether compositional practices
in popular music have changed in the last decades in a way that is consistent with
the theory of attention economy. The results suggest that, as predicted, composi-
tional practices have changed in a way that favors attention-grabbing behavior. The
strongest change observed was the near elimination of instrumental introductions
between 1986 and 2015.
Building on these results, two behavioral experiments were conducted to test
whether these changes are effective ways to grab listeners’ attention. In the first
experiment, participants were asked to listen to randomly sampled songs in an un-
divided attention setting, while in the second experiment, an independent group of
participants was asked to listen to a subset of the same songs in a divided attention
setting. The results suggest that music listening choice behavior (i.e. how and when a
ii
listener decides to skip when listening to music) is mediated by the listening context,
and that compositional devices, along with musical preferences and familiarity with
a song, can predict choice behavior in listeners.
iii
This dissertation is dedicated to my loving and supportive wife, Vanessa.
iv
Acknowledgments
I would like to thank numerous parties for the help and support with this research
project. In no particular order:
David Huron, advisor on this project, for constantly giving me generous feedback
and comments, and for his invaluable help with the funding process. Eugenia Costa-
Giomi, David Clampitt, Johanna Devaney, and Anna Gawboy, for making the OSU
School of Music an exciting teaching and research institution. Robert Bond, for his
support and help on this project. My office mates and friends, Erin Allen, Claire
Arthur, Andrew Brinkman, SongHui Chon, Nat Condit-Schultz, Dana DeVlieger,
Niels Ch. Hansen, Kirsten Nisula, David Orvek, Lissa Reed, Lindsey Reymore, An-
drea Schiavio, Nicholas Shea, and Caitlyn Trevor. Special thanks to Lindsay War-
renburg for her constant help and support. Thanks to CSML alumni Joe Plazak and
Paul von Hippel. Thanks to Misti Crane for writing a really awesome press release
about my research–what an exciting couple of months that was! I would also like
to thank the many members of the OSU community that offered useful ideas and
suggestions.
v
This research project has been made possible in part by the financial support of
the Fonds de recherche du Québec - Société et culture.
Finally, special thanks to my wife Vanessa for the continuous encouragement and
support. I could not have done it without you.
vi
Vita
2015-present . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Graduate Teaching Associate, The Ohio State University
Publications
Research Publications
Duinker, B. & Léveillé Gauvin, H. (2017). Changing content in flagship music theory journals, 1979–2014. Music Theory Online, 23 (4).
Devaney, J., & Léveillé Gauvin, H. (2017). Encoding music performance data in Humdrum and MEI. International Journal on Digital Libraries, Published Online First October 23, 2017.
Léveillé Gauvin, H. (2017). Drawing listener attention in popular music: Testing five musical features arising from the theory of attention economy. Musicae Scientiae, Published Online First March 1, 2017.
Devaney, J., & Léveillé Gauvin, H. (2016). Representing and linking music perfor- mance data with score information. In B. Fields & K. Page (Eds.), Proceedings of the 3rd International Workshop on Digital Libraries for Musicology (pp. 1-8). New York: ACM ICPS.
vii
Léveillé Gauvin, H., Huron, D., & Shanahan, D. (2016). On the role of semitone intervals in melodic organization: Yearning vs. baby steps. In Proceedings of the 14th International Conference on Music Perception and Cognition (pp. 727-731). San Francisco, CA.
Léveillé Gauvin, H. (2016). Changing use of seventh chords: A replication of Mauch et al. (2015). Empirical Musicology Review, 11 (1): 103-107.
Léveillé Gauvin, H. (2015). “The times they were a-changin’:” A database-driven ap- proach to the evolution of musical syntax in popular music from the 1960s. Empirical Musicology Review, 10 (3): 215-238.
Fields of Study
Major Field: Music
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. Capturing and Sustaining Attention in the Attention Economy Era . . . 4
2.1 On what basis is the music entertaining or engaging? . . . . . . . . 8 2.1.1 Arousal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.1.2 Predictability . . . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 How is the music structured and what role(s) does this structure serve? 11 2.2.1 AABA and Verse-Chorus Forms . . . . . . . . . . . . . . . . 11 2.2.2 Instrumental Introductions . . . . . . . . . . . . . . . . . . 12
2.3 How does the music achieve memorability? . . . . . . . . . . . . . . 15 2.3.1 Repetition & Memory . . . . . . . . . . . . . . . . . . . . . 15 2.3.2 Earworms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.4 How does the use of language contribute to the music’s poetic or emotive appeal? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.4.1 Self-Focus, Narcissism & Anti-Social Behavior . . . . . . . . 19 2.4.2 Seriousness & Meaningfulness . . . . . . . . . . . . . . . . . 20 2.4.3 Sexual Content . . . . . . . . . . . . . . . . . . . . . . . . . 21
ix
2.5 To whom is this music aimed? . . . . . . . . . . . . . . . . . . . . . 21 2.5.1 Style & Genre . . . . . . . . . . . . . . . . . . . . . . . . . 21 2.5.2 Personality Traits . . . . . . . . . . . . . . . . . . . . . . . 24 2.5.3 Tasks & Activities . . . . . . . . . . . . . . . . . . . . . . . 25
2.6 How does the music establish its credibility or authority? . . . . . . 26 2.6.1 Guest Appearances . . . . . . . . . . . . . . . . . . . . . . . 26 2.6.2 Musical Quotation & Sampling . . . . . . . . . . . . . . . . 27
2.7 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3. Testing Five Musical Features Arising from the Theory of Attention Economy 31
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3.2 Study 1: Practices between 1986 and 2015 . . . . . . . . . . . . . . 36
3.2.1 Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 3.2.2 Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.2.3 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.2.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.2.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3.3 Study 2: Within-artist comparisons of success . . . . . . . . . . . . 44 3.3.1 Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.3.2 Sample . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.3.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.4 General Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4. Testing the Influence of Attention Economy Principles, Musical Prefer- ences, and Familiarity on Music Listening Choice Behavior . . . . . . . . 54
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 4.2 Hypotheses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4.3 Experiment 1: Undivided Attention Music Listening Setting . . . . 59
4.3.1 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.3.1.1 Participants . . . . . . . . . . . . . . . . . . . . . . . 60 4.3.1.2 Material . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.3.2 Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 4.3.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 4.3.4 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.4 Experiment 2: Divided Attention Music Listening Setting . . . . . 68 4.4.1 Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
4.4.1.1 Participants . . . . . . . . . . . . . . . . . . . . . . . 69 4.4.1.2 Material . . . . . . . . . . . . . . . . . . . . . . . . . 69 4.4.1.3 Procedure . . . . . . . . . . . . . . . . . . . . . . . . 70
x
4.5 General Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5. General Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.1 Recapitulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 5.2 Implications for the Music Industry . . . . . . . . . . . . . . . . . . 82 5.3 Future Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 5.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
Appendices 88
A. List of the 303 songs analyzed in the first study (Section 3.2) of Chapter 3 88
B. List of the 120 songs analyzed in the second study (Section 3.3) of Chapter 3 97
C. Visual representation of the participants’ skipping times in both the un- divided and divided attention experimental setting (Chapter 4) superim- posed onto each song’s formal diagram . . . . . . . . . . . . . . . . . . . 103
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
2.1 Summary of the literature discussed in this chapter. . . . . . . . . . . 28
3.1 Comparison between first-person singular pronoun usage in the present study and in DeWall et al. (2011). . . . . . . . . . . . . . . . . . . . . 45
3.2 Pearson correlation matrix for five variables. . . . . . . . . . . . . . . 46
3.3 Paired t-test results comparing most popular and less popular songs according to five variables. . . . . . . . . . . . . . . . . . . . . . . . . 50
4.1 Summary of multiple regression for undivided attention. . . . . . . . 63
4.2 Summary of participants’ responses in the undivided attention condi- tion to the prompt “Please briefly explain why you skipped the previous song you heard.” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
4.3 Summary of multiple regression for divided attention. . . . . . . . . . 72
xii
Figure Page
2.1 The attention economy model predicts that, as the disparity between available human attention and information content increases, so does the value of attention. This figure illustrates the attention economy model. The x-axis represents time. The y-axis represents the sum of the information content of all attention-grabbing products (e.g. web pages, images, music, ads) available to a single individual. The dashed- line represents the limited amount of attention a single person has. The scarcity-based increasing value of attention is represented by the darkening of the shaded area: as the gap between information content and human intention increases, so does the value of human attention. 6
2.2 “Blue Moon.” Music by Richard Rogers. Lyrics by Lorenz Hart. Pub- lished in 1935. An example of a song from the ‘Great American Song- book’ written following an AABA structure. . . . . . . . . . . . . . . 13
2.3 Proportion of AABA and Verse-Chorus song forms between 1958 and 1971. The graph shows a decline in popularity of the AABA form in the 1960s in favor of the verse-chorus form (where verse and chorus exhibit different harmonic structures). This shift seems to occur between 1964– 65, with a drastic and quasi-steady decline in frequency of the AABA form over the second half of the decade, going from being present in roughly 40% of the songs in 1964 to being almost nonexistent four years later in 1967. Figure adapted from Léveillé Gauvin (2015, Figure 3.4). 14
xiii
2.4 Proposed cognitive model of Music Listening Choice Behavior (MLCB) featuring a three-stage screening process. In the first, familiarity- related stage (∼ 0–1 sec), listeners may recognize a specific musical work and decide whether the given work is of interest. If the answer is yes, the song passes the familiarity-related screening process and the listening continues, starting the style-related screening process; if the answer no, the song fails to pass the familiarity screening process and the skip button is pressed. In the second, style-related stage (∼ 0–5 sec), listeners quickly recognize the style/genre and assess their inter- est in listening to music of that style/genre. They then assess whether they want to listen to a song belonging to that category. If the an- swer is yes, the song passes the style-related screening process and the listening continues, starting the third, song-related screening process (∼ 5–20 sec); if the answer no, the song fails to pass the style-related screening process and the skip button is pressed. Finally, in the third, song-related stage, listeners assess whether they enjoy they like the musical work, based on their current mood and context. In this stage, characteristics that are specific to this song (such as the melody, the harmony, the lyrics) might come into consideration. Although visually illustrated as three independent stages, the screening processes pre- sented in the MLCB model are likely to temporally overlap with one another. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.1 Relationship between the year a song appeared on the Billboard Year-…