Do we all imagine the same thing when we listen to music, or are our experiences hopelessly subjective? In other words, is music really a universal language?

To explore these questions, an international team of researchers (including a classical pianist, a rock drummer and a concert bass player) asked hundreds of people what stories they imagine while listening to instrumental music. The results appeared recently in the Proceedings of the National Academy of Sciences.

The researchers, led by Elizabeth Margulis of Princeton and Devin McAuley of Michigan State University, found that listeners in Michigan and Arkansas imagined very similar scenes, while listeners in China imagined completely different stories. different.

“These results paint a more complex picture of the power of music,” said music teacher Margulis. which uses theoretical, behavioral and neuroimaging methodologies to study the dynamic experience of listeners. “Music can generate remarkably similar stories in the minds of listeners, but the degree to which these imaginary narratives are shared depends on the degree to which culture is shared among listeners.”

The 622 participants came from three regions on two continents: two suburban college towns in Central America – one in Arkansas and the other in Michigan – and a group from Dimen, a village in rural China where the language main one is Dong, a tonal language unrelated to Mandarin, and where locals have little access to Western media.

All three groups of listeners – in Arkansas, Michigan and Dimen – heard the same 32 musical stimuli: 60-second snippets of instrumental music, half Western music and half Chinese music, all without words. After each musical excerpt, they provided free-response descriptions of the stories they envisioned as they listened.

The results were startling. Listeners in Arkansas and Michigan described very similar stories, often using the same words, while listeners in Dimen imagined stories similar to each other but very different from those of American listeners.

This audio clip, identified by researchers only as W9, is from the opening strains of Ferde Grofé’s “Grand Canyon Suite”.

For example, a musical passage identified only as W9 evoked a sunrise over a forest, with animals waking and birds chirping for American listeners, while those of Dimen imagined a man blowing a leaf on a mountain, singing a song to his beloved. For musical passage C16, listeners in Arkansas and Michigan described a cowboy, sitting alone in the desert sun, gazing at a deserted town; Dimen participants imagined a man in ancient times sadly contemplating the loss of his beloved.

Quantifying the similarities between the free-response stories required huge amounts of natural language data processing. The tools and strategies they developed will be useful in future studies, said Margulis, who is also director of the Music Cognition Lab at Princeton. “Being able to map these semantic overlaps, using natural language processing tools, is exciting and very promising for future studies that, like this one, straddle the boundary between the humanities and the sciences.”

“It’s amazing,” said co-writer Benjamin Kubit, drummer and postdoctoral research associate formerly at the Princeton Neuroscience Institute and now in the Department of Music. “You can take two random people who grew up in a similar environment, play them a song they’ve never heard before, ask them to imagine a story, and you’ll find similarities. However, if those two people don’t share a culture or geographic location, you won’t see the same kind of similarity in experience, so while we imagine music can bring people together, the reverse can also be true. it can distinguish between groups of people with different backgrounds or cultures. »

Although the researchers carefully ensured that the tracks they chose had never appeared in a movie soundtrack or any other setting that would prescribe visuals, the same music triggered very similar visuals in hundreds of people. listeners – unless they grew up in a different cultural context.

“It’s amazing to me that some of these visceral, hard-to-articulate, imaginary responses we have to music can actually be shared widely,” Margulis said. “There’s something about it that’s really confusing and compelling, especially because the way we encounter music in 2022 is often lonely, through headphones. But it turns out it’s still a shared experience, almost like a shared dream. I find it really surprising and fascinating with the caveat, of course, that this is not universally shared, but depends on a common set of cultural experiences.”

Co-author Cara Turnbull, a concert bassist turned graduate student in musicology, said: “It’s just fascinating how much our upbringing shapes us as individuals while giving us enough shared experiences that we relate to this medium in both unique and shared ways.”

“Narratives Imagined in Response to Instrumental Music Reveal a Culture-bound Intersubjectivity,” by Elizabeth H. Margulis, Patrick CM Wong, Cara Turnbull, Benjamin M. Kubit, and J. Devin McAuley, appears in the January 21 issue of Proceedings of the National Academy of Sciences (DOI: 10.1073/pnas.2110406119). This research was supported by the Division of Behavioral and Cognitive Sciences of the National Science Foundation, award numbers 1734063 (PI: JDM) and 1734025 (PI: EHM).