Facebook
Twitter
Pocket
Reddit
Email
Artistic appreciation is a deeply subjective process, perhaps the most essentially personal thing that humans do. But are there some explanations for why we like what we do? Why, for instance, does a particular song get popular?
Some of it has to do with the quality of the music—and by quality, I mean that there’s something about the music itself the resonates with people and makes them want to listen to it.
One thing that people seem to like in music is when it addresses themes we care about—particularly songs that involve romance. There is a debate over what kind of meaning is conveyed by instrumental music, but music with lyrics usually involves human themes: 92 percent of the most popular tracks from 2009 had “reproductive” messages (pdf)—involving courtship, attraction, and, of course, sex itself—with an average of 10.49 per song. Bestselling tracks contained even more. And these kinds of messages were found in songs dating back at least 400 years!
People also seem to like music that is relevant to the times that they live in—in particular, to contemporary fears. In difficult social and economic times, popular music has longer songs, more meaningful content, is more comforting, romantic, and slower, according to one 2009 study. This reflects, perhaps, people’s search for comfort and meaning when they are frightened.
Some Accounting for Taste
Of course, not everybody likes the same music. To some degree, people’s preferences are predictable: Our stereotypes about who likes what kind of music tend to be accurate. People seem to prefer music that matches their personalities.1 People who like rock are more likely to be anxious and artistic, and not very conservative or friendly. Classical fans tend to be friendly, emotionally stable, and conscientious. People are pretty good at guessing somebody’s personality traits from what kind of music he or she likes (pdf).
One study (pdf) found that people who have higher IQs, are more intellectual, and crave new experiences tend to appreciate music in a more cerebral way, focusing on complexity, and listening with a critical ear. The more neurotic and introverted and less conscientious folks use music more for emotional regulation, like for a pick-me-up after a hard day.
Where we often get a sense immediately if we like a painting or not, many of us have learned that music, although we sometimes get a strong reaction at first, often takes repeated listenings before we know for sure whether we like it or not. Indeed, experiments show that repeated exposure to a particular track increases positive feelings toward it. Music is meant to be heard again and again. We don’t think it strange if someone tells us they love a novel that they have only read once, but we would be surprised if someone mentions a song that he or she loves, yet has only listened to one time.
But there are other forces at work that have a powerful influence on whether a particular song gets popular.
Everyone Is (in Thrall to) a Critic
The culture that we are embedded in has a large effect on what music we find compelling: We use the opinions of others to help us determine what we like. It is possible that there is a bias to trust authority figures when it comes to judging art, but this power is not limited to professional critics; sometimes we see the mere popularity of a cultural product as a signal of its quality. This is not entirely irrational: Although some people take pride in listening to music that nobody else knows about, it’s also nice to consume the same music and movies as your friends so you can experience them and talk about them together.
But can this influence get out of hand? It sure seems so, because some songs become popular even though they are terrible (of course, nobody can agree on which songs are actually terrible). It feels, at least, like some songs are popular just because they happen to become popular, rather than as a result of their inherent quality. Can we find any evidence that some songs get popular just because they happen to get some boost soon after their releases?
In the real world, it’s tough to tease out objective song quality from the domino effect of song popularity. So the only way to figure it out is with an experiment where you can control the variables. This is what psychologists Matthew Salganik and Duncan Watts did several years ago: They conducted an online experiment (pdf) in which 12,207 people listened to, rated, and downloaded unknown songs from unknown bands. Some of these people were able to see the download rates of each song (the social-influence group), and others were not (the independent group). Would people use the download rates as a sign of quality, or would the sound of the music determine which songs ended up most well-liked?
The researchers found there was indeed an influence of popularity. In the social-influence group, the most downloaded songs ended up being the most popular. A person in the social-influence group was about six times more likely to listen to the most popular song than to listen to one in the middle rankings. Interestingly, the least popular songs were also more likely to be listened to than the middling ones, perhaps because they were easier to find at the very bottom or as part of an explicit anti-conformist behavior.
To further explore the issue, the researchers also created an “inverted” group, in which the least popular songs were made to look like they were the most popular, and vice versa. Would the bad songs (as rated by people in the independent condition) end up as the most popular, just because they were given a head start?
It turns out that social influence had a real effect. Songs that were rated objectively bad in the independent condition were listened to more often in the inverted condition. Likewise, objectively good songs (again, as chosen by the independent group) fared more poorly.
However, over time the good songs “recouped much of their original popularity,” and bad songs at the top rankings gradually fell, suggesting that song popularity was not simple a matter of copying the crowd; the overall song quality was also a factor. Also, in the inverted conditions, people left the experiment earlier, and there were fewer downloads over all. Perhaps they were frustrated by the mismatch of their own preferences with other people’s.
So even though people can be swayed by the crowd, particularly as a guideline for what to try, some songs really are better than others.
Footnote:
1. Winner, E. (1982). Invented Worlds: The Psychology of the Arts. Harvard University Press. Pages 72–73.
Jim Davies is the author or Riveted: The Science of Why Jokes Make Us Laugh, Movies Make Us Cry, and Religion Makes Us Feel One with the Universe. He teaches cognitive science at Carleton University.
Get the Nautilus newsletter
Cutting-edge science, unraveled by the very brightest living thinkers.
