Mixing Memory

What Does Your Music Say About You?

I went to a high school at a time (one not that different from most others, I imagine) when musical preferences were a good clue to social group membership. There were, for example, the punks who listened to, well, punk; the stoners who listened to Pink Floyd’s “The Wall” over and over and over again; the hipsters, who listened to what was the hip music of the time (grunge); and the “popular” kids who listened to pop, pop rock, and country (I went to high school in Nashville, where country music was the popular music). We all assumed that what a person listened to could tell you a lot about who they were, and it wasn’t uncommon for the question, “So, what sort of music do you listen to?” to be one of the first asked upon meeting someone new. It turns out our assumption was probably right. People’s musical preferences good indicators of their personality make up. At least, that’s the finding of research coming out of the lab of social psychologist Sam Gosling, of animal personality fame.

Gosling is not new to intuitively plausible personality research. He’s found, for example, that people’s personality judgments based only on observing a person’s bedroom or obvious are pretty accurate1, which fits with my experience of people accurately judging my personality based only on the fact that my desk looks like it has been hit by an F5 tornado. In the last few years, he’s turned to music, with the following goals2:

The fundamental question guiding our research program is, Why do people listen to music? Although the answer to this question is undoubtedly complex and beyond the scope of a single article, we attempt to shed some light on the issue by examining music preferences. In this research we take the first crucial steps to developing a theory of music preferences–a theory that will ultimately explain when, where, how, and why people listen to music. (p. 1236)

It turns out that, despite the fact that most of us believe that musical preferences say something about who we are, social psychologists haven’t really been all that interested in studying music. Rentfrow and Gosling write

At this very moment, in homes, offices, cars, restaurants, and clubs around the world, people are listening to music. Despite its prevalence in everyday life, however, the sound of music has remained mute within social and personality psychology. Indeed, of the nearly 11,000 articles published between 1965 and 2002 in the leading social and personality journals, music was listed as an index term (or subject heading) in only seven articles. (p. 1236)

Seven articles! This simply will not do (I should note that cognitive psychologists have been studying the hell out of music for some time, ’cause we get people). So Rentfrow and Gosling start down the road to remedying this by looking at the relationship between individual differences in musical preferences and personality traits.

They started off by looking at people’s beliefs about the importance of music “in people’s everyday lives.” To do this, they gave university undergrads questionnaires that asked them to rate the importance of various activities, including listening to music, to indicate how often they participated in those activities, and to rate how much those activities said about themselves and other people. As you can see from the graph below (Rentfrow and Gosling’s Figure 3), participants believed that music revealed as much or more about themselves and others than other activities. In act, hobbies were the only types of activities that revealed as much about people as their musical preferences.

Now confident that people really do use musical preferences as indicators of personal qualities, Rentfrow and Gosling next sought to map out the dimensions of those preferences. They started by identifying different musical genres, using a free-association task with five judges, and then getting more information from music stores. This process yielded 80 different musical genres, which were then divided into fourteen, with 66 subgenres. Using these genres and subgenres, Rentfrow and Gosling developed a questionnaire they called the “Short Test of Musical Preferences,” or STOMP, which they gave to participants along with several personality measures. Participants also completed the STOMP a second time three weeks after their first testing, to measure the test’s retest reliability.

The STOMP results were fed into a factor analysis, which yielded four factors, or dimensions, of musical preferences. Rentfrow and Gosling labeled these dimensions “Reflective and Complex,” which included the genres blues, jazz, classical, and folk; “Intense and Rebellious,” which included rock, alternative, and heavy metal; “Upbeat and Conventional,” including country, sound tracks, religious, and pop; and “Energetic and Rhythmic,” including rap and hip/hop, soul and funk, as well as electronica and dance. The correlations between these factors from the first testing and the retesting were high (between .77 and .82), indicating that STOMP and its resulting factors were reliable. Subsequent testing with more diverse samples (the original sample was comprised entirely of university undergrads) indicated that the factors were generalizable as well.

Finally, they looked at the correlation between the different dimensions of musical preferences and different personality traits using several different measures of personality, including the Big Five Index, tests of social dominance, and tests of communication styles. The results indicated that the different dimensions of musical preferences do in fact correlate with different personality features. Here’s a summary of the results (from pp. 1248-1249):

In short, people who listen to jazz are smart, liberal, adventurous, and poor; people who listen to heavy metal are smart, liberal, adventurous, athletic, and prone to social dominance; people who listen to Madonna or the “Dancing With Wolves” soundtrack are agreeable, conscientious, conservative, rich, happy, dumb, emotionally unstable, and hot; and people who listen to hip hop are extraverted, agreeable, liberal, athletic, and hot. Well, those are the tendencies at least (I’ve known some smart Madonna fans, though I have to say that they were pretty emotionally unstable).

Having established that musical preferences do reveal information about personality differences, Rentfrow and Gosling conducted a set of follow up studies designed to look at whether and how people actually use information about musical preferences to make personality judgments3. First, they looked at whether people talk about music when getting to know each other. To do this, they used the following method with college undergrads:

Participants were introduced to a study of how individuals get to know one another over the Internet. Each was instructed to interact with another participant for 6 weeks using an on-line bulletin-board system. Half the participants were assigned to same-sex pairs, and half to opposite-sex pairs. Participants were given no specific instructions about what to talk about. Instead, they were encouraged to talk about anything that they thought would enable them to get to know one another.

Analyzing the transcripts of these conversations, they found that from the first encounters, music was discussed more often than all other activities combined! Don’t believe me? Here’s the graph (their Figure 1):

As you can see, it wasn’t until the sixth week of the participants’ interactions that the amount of talk about all of the other activities combined equaled the amount of talk about music. Clearly, then, people are bringing up music a lot when they’re getting to know someone, indicating that they believe it will tell people about themselves. But are our perceptions based on musical preferences accurate? To determine this, Rentfrow and Gosling had 74 undergrads provide a list of their top ten favorite songs. A second set of participants was then asked to rate the first set of participants on several personality dimensions, based solely on listening to the individuals’ ten favorite songs. These ratings were then compared to the first set of participants’ scores on the same personality dimensions. The correlations between self-ratings and the observers (who’d only listened to their favorite songs!) ranged between .11 and .38 for each of the Big Five personality dimensions, with all but one of the correlations (emotional stability) being .27 or above. In other words, personality judgments based solely on musical preferences were pretty damn accurate. That’s impressive!

There you have it, then. Social psychologists (two of them, at least) have jumped head first into the waters of music research, learning that our music says a lot about who we are, and that we can make pretty accurate judgments about what people are like based solely on the music they like. I now feel somewhat less guilty about not being friends in high school with people who listened to country music. We probably wouldn’t have gotten along, ’cause our personalities would have clashed.

1Gosling, S.D., Ko, S. J., Mannarelli, T., & Morris, M.E. (2002). A Room with a cue: Judgments of personality based on offices and bedrooms. Journal of Personality and Social Psychology, 82, 379-398.
2Rentfrow, P.J., & Gosling, S.D. (2003). The do re mi’s of everyday life: The structure and personality correlates of music preferences. Journal of Personality and Social Psychology, 84, 1236-1256.
3Rentfrow, P.J., & Gosling, S.D. (2006). Message in a Ballad: The role of music preferences in interpersonal perception. Psychological Science, 17, 236-242.