Cognitive Daily

Psychologists have known for decades that people perceive music as happier when it’s played faster, and in a major key (mode). Take a listen to the following sound clips I created using a synthesized flute. Each plays the same melody three times—first in a major mode, then a minor mode, then a “whole tone” middle ground. The only difference between the two clips is that the second clip is played twice as fast.

Clip 1 (slow)
Clip 2 (fast)

For most people, the second clip sounds happier than the first overall, and the major mode portion sounds happiest within each clip. But what matters most—the speed (tempo), or the mode?

When Kate Hevner first reported on this phenomenon in the 1930s, she demonstrated the relationship between tempo and mode and happiness/sadness judgements, declaring tempo to be the more important of the two. After listening to the clips, you may agree with Hevner, but her data didn’t offer a compelling reason to believe that tempo matters more.

Much more recently, Lise Gagnon and Isabelle Peretz found a way to determine which impacts our perception of emotions more—mode or tempo. They developed 8 different melodic sequences (one of which I duplicated in the clips above), and then adapted them to major, minor and whole tone modes. These sequences were then played for paid participants in four different conditions:

  • Mode change: All sequences were played at the same tempo, in each of the different modes—major, minor, and whole tone
  • Tempo change: All sequences were played in whole tone mode, at a fast, middle and slow tempo
  • Convergent: Fast tempo sequences were played in major mode, middle tempo sequences were played in whole tone mode, and slow tempo sequences were played in minor mode
  • Divergent: Fast tempo sequences were played in minor mode, middle tempo sequences were played in whole tone mode, and slow tempo sequences were played in major mode

After hearing each sequence, participants rated it on a scale of 1 (happiest) to 10 (saddest). The Mode and Tempo change conditions were included simply to replicate earlier work by Hevner and others, and as expected, the major mode and faster tempo were associated with happier ratings. Gagnon and Peretz expected to find stronger happiness and sadness ratings in the Convergent condition, because mode and tempo effects were combined. As they expected, in the Convergent condition, participants rated combined fast, major mode sequences as significantly happier than either major mode or fast tempo by themselves.

The Divergent condition was the key to the experiment—when, for example, a fast tempo was combined with a minor key, or a slow tempo was combined with a major mode, tempo always took precedence. Slow tempo, even when associated with a major mode, was rated sad, and fast tempo, even when associated with a minor mode, was rated happy.

But what if the only reason for this difference is that the tempo changes were more dramatic than the mode changes? Gagnon and Peretz had doubled the speed of the sequences, from 110 beats per minute for the slow tempo to 220 for fast tempo. What if the tempo change were made less dramatic—would the results still hold? The researchers figured out a way to generate tempo changes that were roughly equivalent to mode changes: they tested the sequences with a new group of volunteers, but this time they used many different tempos, until they found a set of three whose happiness and sadness ratings were roughly equal to those found in the Mode Change condition. This time, the fast tempo was reduced to only 160 beats per minute.

They now repeated the entire experiment using the smaller range of tempos. Even though participants had rated tempo-only and mode-only changes as the same, in the Divergent condition, tempo once again took precedence. So it appears that tempo is more important than mode in determining whether a musical selection is happy or sad.

Gagnon, L., & Peretz, I. (2003). Mode and tempo relative contributions to “happy-sad” judgements in equitone melodies. Cognition and Emotion, 17(1), 25-40.


  1. #1 Alex Young
    October 26, 2005

    “They developed 8 different melodic sequences …” – how is this what the test subjects would describe as “music”? There’s more to music than minor, major, tempo and cadence. As such, a more holistic approach to this research would probably yield different results.

    Have you ever read Emotion and Meaning in Music by Leonard B. Meyer? It’s old, but he covers a broad spectrum of music, and alludes to the fact that a great deal of emotional response from music comes from embellishments performers add (consider trills in baroque music, off-key notes in Jazz or blues). The devil is in the details 🙂

  2. #2 Dave Munger
    October 26, 2005

    “As such, a more holistic approach to this research would probably yield different results.”

    Almost certainly, Alex. The problem then would be determining what emotional responses were caused by particular aspects of the music. In fact, Hevner’s original 1935 study did consider “real” music (she used excerpts from the classical canon), but using her methodology, it would be impossible to isolate the impact of tempo or mode.

    I would argue that Gagnon and Peretz’s experiment, by showing that we have emotional responses to even the simplest melodies, suggests that some aspects of our reaction to music are caused by the most basic elements of sound. This does not discount that a wide range of emotional responses may be elicited through more complex features of music.

    Neither does it suggest that a 110 b.p.m. song could never be seen as “sad,” since as you point out, there is much more to music than tempo and mode.

  3. #3 Larry Beck
    October 26, 2005

    Is music all about emotions or feelings? Is there a difference between emotions and feelings? Yes, there is, and I believe music is more about feelings than emotions. That said, this study seems to depend upon how much musical training the person has had, because I can listen to the slow movement of Dvoark’s New World Symphony and feel happy, regardless of the slow tempo. Just like I can listen to the last movement of Tchaikowsky’s 4th symphony and feel sad, even though it’s played at a break neck tempo. Because I know the background of the two pieces of music, it affects how I feel about them.

  4. #4 Dave Munger
    October 26, 2005


    The participants in the study had no musical training—however, I believe there have been other studies with trained musicians that offer similar results.

    I agree with all your points about the complexity of music; however, it’s also important to understand what concepts hold up even in the simplest instances of music. It would be difficult to conduct a controlled study of the New World Symphony, for example, because any change in the melody would be a fundamental alteration of the work itself.

    re: feelings/emotions. Why can’t it be about both? And isn’t it a matter of opinion, rather than controlled study, as to which is more important?

  5. #5 Scott Spiegelberg
    October 26, 2005

    Some studies by Emmanuel Bigand and others have attempted to isolate induced emotions from communicated emotion. This is similar to Larry’s distinction, with induced emotion as his “feeling.” With induced emotions, it is much more difficult to get universal claims, whereas communicated emotion (or inherent emotion) is more stable.

  6. #6 Larry Beck
    October 27, 2005

    Bennett Reimer’s “A Philosphy of Music Education” is where I “learned” how to really appreciate music for what it really is, a means of non-verbal communication that bypasses our senses to directly affect our perceptions of reality. I’ve spent a lot of time pondering about the effect music has on people, for good and bad. Why is it that I can be eating lunch in a food court in a busy, noisy mall, and be brought to tears when Dan Fogelberg’s “Leader of the Band” is played? Because my Dad was a musician? Or because the song is well crafted? Probably a little bit of both. Or Mozart’s Clarinet Quintet. Because it was featured in the last episode of MASH? Or because it’s a well crafted beautiful piece of music? (But what determines well crafted and beautiful?) I know this: music is one of those gifts from God that we should get down on our knees and thank Him for every evening.

  7. #7 Jeff Dunn
    November 10, 2005

    I would point out that reporting that a particular musical passage as “sad” is not the same thing as “feeling sad” for a legitimate personal reason. “Sad” and “happy” are probably the wrong words to use for musical feelings–in most situations. The trouble is, we don’t have a vocabulary for musical emotions like the Eskimos do for snow. Perhaps “energized” and “reflective” might be better to describe the affect of fast vs. slow music.

New comments have been disabled.