Just listening to music, despite the hype associated with the “Mozart Effect,” appears to have little influence on IQ or other abilities. It does seem to make us more aroused and put us in a better mood, which can improve performance on tests, but it doesn’t actually make us any smarter. But what about actual long-term training in music? Clearly musical training can make us better able to perform and appreciate music, but can it also improve our performance in areas? With its mathematically based notation system, music has been shown to improve mathematical reasoning skills. But surely music is more than just math. What other abilities does it improve?
A group led by William Thompson reasoned that since so many studies have linked music and emotion (we’ve discussed some of these studies here, here, and here), perhaps musical training can help us perceive emotions in others. We know music can express emotion, and obviously facial expressions can, too. More recent research has suggested that the musical aspects of speech itself—speech prosody—also express emotion. Some theoreticians have speculated that music itself originated with mothers’ desire to communicate emotionally with their children through songs. So perhaps musical training itself can help us interpret speech prosody.
Thompson et al. developed a set of experiments to see if there is a relationship between musical training and recognizing speech prosody. They recorded a speaker saying ordinary sentences such as “the chairs are made of wood” while expressing four different emotions: happy, sad, angry, and fearful. Then they transformed the sentences into musical sequences by calculating the average pitch of each syllable. In their first experiment, they tested musically trained and untrained adults. The untrained group had had no formal music instruction, while the trained group had taken at least 8 years and averaged over 13 years of music lessons. Participants listened to the tone sequences and then indicated whether they thought the emotion of the corresponding sentence was angry, sad, happy, or fearful. Here are the results:
The musically trained group was significantly better at identifying the emotion from the sequences than the untrained group. While both groups were not as accurate with the fearful and angry sequences, in every case, the trained group was better at detecting emotion from speech prosody.
But this experiment on its own doesn’t show that musical training causes people to be better at interpreting emotion—it might be that the people who’ve stuck to music lessons for all those years chose music in the first place because of their superior ability to perceive emotion.
To address these concerns, Thompson et al. turned to a group of 7-year-olds that were part of a larger study on the impact of music instruction. These kids had been randomly assigned to groups receiving a year of lessons at the age of six—either drama, keyboard, singing, or no lessons (the kids in the “no lessons” group received a free year of instruction the following year). Since they were randomly assigned, there was no chance that more emotionally sensitive kids had self-selected music lessons.
Though the children were given a simpler test than the adults (they only had to distinguish between two emotions at a time—happy/sad, or anger/fear), the results this time were less clear. There was no difference between the groups when distinguishing between happy and sad. However, in the more difficult anger/fear condition, the results were as follows:
The kids who took keyboard or drama lessons performed statistically better than those with no lessons. However, the result for singing lessons was neither statistically different from the keyboard or drama group, nor the control group. So drama lessons—as may be expected—help perception of speech prosody, as do keyboard lessons, but rather unexpectedly, one year of singing lessons did not lead to an improvement.
Thompson et al. believe that this result for singing lessons can be explained by the different methods used for teaching singing to children compared to keyboard lessons. Singers generally are asked to use pitches that do not correspond to those used in normal prosodic speech—so emotion is not conveyed the same way in song as it is in speaking. Also, singers are often not asked to generate precise pitch—a wide range of pitches around the desired note is usually tolerated, whereas with keyboards, the pitch produced by each key is always identical. Finally, it’s possible that just one year of singing instruction is insufficient to affect perception of speech prosody.
Nonetheless, the results of the child study with keyboarding, combined with the adults who have had more years of instruction, suggest that music training does confer real benefits in perceiving emotion.
Thompson, W.F., Schellenberg, E.G., & Husain, G. (2004). Decoding speech prosody: Do music lessons help? Emotion, 4(1), 46-64.