Sam Mehr and Manvir Singh from Harvard’s Music Lab talk with us about their research suggesting that people across the world can detect the social purpose of other cultures’ songs based only on how they sound. Their open-access article, “Form and function in human song” was published in Current Biology on February 5th 2018. The article was co-authored with Hunter York, Luke Glowacki, and Max M. Krasnow.
[powerpress]
Websites
- The Natural History of Song Project
- Demonstration of the first experiment
- Sam & Manvir on NPR’s Science Friday
- Alan Lomax’s The Global Jukebox
- Sandra Trehub at University of Toronto
- Headphone Screening Test
- Sam and Max’s 2017 article “Parent-offspring conflict and the evolution of infant-directed song” in Evolution and Human Behavior.
Press
- “A Study Suggests That People Can Hear Universal Traits in Music” in The Atlantic
- “Can You Tell a Lullaby From a Love Song? Find Out Now” in New York Times
- “Where does music come from?” in Christian Science Monitor
- “Songs In The Key Of Humanity” in Harvard Gazette
- “Perhaps Music Really Is A Universal Language” in Pacific Standard
- “Music may be the food of love, but oddly, is not its language” in The Economist
- “Your Brain Knows What Songs Are For, No Matter Where They Came From” in Smithsonian Magazine
- “Music Really Is A Universal Language” in phys.org
Bonus clips
Patrons of Parsing Science gain exclusive access to bonus clips from all our episodes and can also download mp3s of every individual episode.
Support us for as little as $1 per month at Patreon. Cancel anytime.
[dropdown]
[dropdown-option] Clips available to patrons include … [/dropdown-option]
[dropdown-option] Full episode with available download [/dropdown-option]
[dropdown-option] Bonafide theories are falsifiable theories [/dropdown-option]
[dropdown-option] Copyright considerations in studies such as theirs [/dropdown-option]
[dropdown-option] Earlier attempts in literature to identify universals [/dropdown-option]
[dropdown-option] Checking for headphone use [/dropdown-option]
[dropdown-option] On form and function of healing songs [/dropdown-option]
[dropdown-option] Music acquisition in other species [/dropdown-option]
[dropdown-option] Potential use of data science in Natural History of Music projects [/dropdown-option]
[dropdown-option] The Center for Open Science’s Preregistration Challenge [/dropdown-option]
[dropdown-option] Purpose of prestudy [/dropdown-option]
[dropdown-option] Timing of prestudy [/dropdown-option]
[dropdown-option] The principal components [/dropdown-option]
[dropdown-option] On Sam’s research into vocalizing to babies [/dropdown-option]
[dropdown-option] Why vocal-only music was studied [/dropdown-option]
[/dropdown]
Patrons can access bonus content here.
We’re not a registered tax-exempt organization, so unfortunately gifts aren’t tax deductible.
Hosts / Producers
Ryan Watkins & Doug Leigh
How to Cite
Watkins, R., Leigh, D., Mehr, S., & Singh, M.. (2018, February 20). Parsing Science – Universals in Songs. figshare. https://doi.org/10.6084/m9.figshare.5915416.v2
Music
What’s The Angle? by Shane Ivers
In their study, Sam and Manvir’s team investigated how song’s features or form relates to their social function. To do so, they first asked people to guess what songs’ social purposes were … by listening to short excerpts of vocal music that were collected throughout the world. Later, they had a new group of people to do the same, but also instructed them to identify what features they could hear in the songs. We began our conversation by asking Manvir to explain how they carried this out.
The song excerpts used in the study were drawn from the Natural History of Song project, a systematic index of the signature of vocal music from across the world. Here, Sam and Manvir explain how the Natural History of Song discography is organized and what the project is set out to accomplish.
In the sense of there being a form of musical expression that can be intuited even by listeners who don’t understand the language used in the song, Sam and Manvir’s experiments suggest that universal links exist between form and function in vocal music. They talk with us next about what’s important about this conclusion, as well as how their findings go beyond what has been found in previous studies.
To measure listeners’ confidence in their evaluation of each song’s functions, Sam and Manvir used a six-point scale ranging from “definitely not” being used for each of various purposes, to it being “definitely” used for each of those purposes. Ryan and I wondered how they came up with this approach to measuring people’s ability to determine songs’ social function and what the advantages of doing so are.
Sam and Manvir used Amazon.com’s Mechanical Turk – an online marketplace for requesting research participants. As several of our previous guests have used “mTurk” for similar purposes, we were interesting learning what their experience was in recruiting participants through the site.
Doug and I were curious how it is that it came about that people from differing cultures are still able to discern the intent of one another’s music. Manvir gives us his impressions on the question.
In their first experiment, 750 people across 60 countries listened to song excerpts drawn from the Natural History of Song’s discography then rated those songs’ functions. In their second experiment, people were additionally asked to assess 10 factors related to those songs forms as well. Ryan and I followed up on Manvir’s explanation of form and function by asking Sam to tell us why they asked listeners to rate both the contextual features of songs, as well as musical ones.
Sam and Manvir were interested in learning if people could infer the social function of songs as being for dancing, to soothe a baby, to heal illness, or to express love to another person based solely on the song’s features. Doug and I asked Manvir if a song’s function can be thought of as it’s genre, with its form being how it sounds to a listener as well as where these terms originated.
Sam and Manvir used a statistical technique called “principal components analysis” – or PCA – to determine the number of factors that uniquely explain the relationships among the variables measured in a study, as well as the strength of those associations. While the method doesn’t name the factors, we asked how they would describe the two principal components that they identified.
Lastly, Doug and I wanted to learn what other recent developments in the study of psychology we might benefit from knowing more about.
The first experiment was preregistered with the Center for Open Science as part of its “Preregistration Challenge,” which rewards researchers with $1000 if their study is accepted for publication by a participating journal. Since Ryan and I had previously interviewed social psychologist Brian Nosek – the co-founder and director of the Center for Open Science – we were interested in learning of Sam and Manvir’s thoughts on preregistration.