The Algorithm to Emotions: Music Manipulation and its Application

Music always plays a huge part in our life. From the lullabies we hear as children, the music we listen to as teenagers, and the songs we use for special events. It’s impossible not to hear music and not feel something at the same time. So, the question is, why is it that some music makes us happy? Why is it that some make us sad? Why do we even listen to sad music if it makes us sad? And why do we need to answer these questions anyway?

But before anything else, researchers have debated amongst themselves on the definition of emotions. (Hunter and Schellenberg, 2010) What exactly constitutes emotions? For the layman, emotions are the usual: happy, sad, angry, disgust, excitement, fear, etc. However, when taken into research terms, it becomes much more complicated than that. For example, when we listen to music, are we actually feeling emotions that we feel in everyday life or is music just inducing temporary feelings that are different from the emotions we feel everyday?

Another matter of fact is, how exactly does music affect emotions? Psychologists have to separate emotions, feelings, and moods. While there are various theories chosen to explain how these came to be, it has been mostly agreed upon that music expresses emotion and these things affect us. Researchers argue whether the emotions we feel are real emotions, aesthetic emotions, or just moods induced by the music. It’s also been debated whether we just perceive the emotion expressed by the music or if we’re really feeling this emotion. (Hunter and Schellenberg, 2010)

There have been various studies debating on the many properties of music and emotions (citation), but if there’s one thing that all the researches have agreed on, it’s that certain chords are associated with certain emotions. Major chords are associated with happiness, minor chords are associated with sadness, and dissonance chords are associated with unpleasantness (Scherer, 2004). An example would be playing a song in either minor or major chords and seeing the difference of the emotions evoked in the chords played.

As seen in the video, switching from major to minor chords can really change how songs feel, don't they?

            It’s incredible how just changing the set of chords can make us feel very different things. However, it’s not just the chords that could change how music can portray emotions. Other musical characteristics can actually also change how music can feel.  For example, tempo and mode are the strongest determinants of emotion in music in a series of studies created before (citation). Fast tempos and major modes are associated with happiness and slow tempos and minor modes are associated with sadness. However, even as music can make us feel emotions, the most consistently judged emotions are happiness and sadness, compared to other emotions like anger and fear.

            Some evidence has also been found that timbre affects emotions in music. Soft timbres attenuated with high frequencies are associated with sadness, where as sharp timbres are associated with anger. Loudness of the sound has also been found to be a universal cue to anger according to a cross-cultural study by Balkwill and coworkers (2004)

As seen in this video, one can really feel the anger and unpleasantness of the song. It doesn't matter if you don't understand the lyrics, as long as you hear that melody, it feels quite angry.

Compared to this very soft and calming music, the disparity between the two really is huge. The other makes you feel anger while this one calms you down and makes you feel at peace with yourself.

            Although while it has been found that music portrays emotion, interestingly enough, children have difficulty identifying the emotions in music, at least until they reach the age of 11 years old, where they reach the adult-like accuracy (Hunter et. al, 2008)

            So why should we learn about emotions and music and why they’re related? Music is always a huge part of our lives. The music industry wants to sell more, wants people to buy their product. Obviously, if you’re a customer and you realize you like a product, the next question you usually ask is:

Is there anything else like this?

            And so, these connections of music and emotions have been used in the music industry in order to recommend songs or group playlists according to the emotions that each one portrays. Using a complicated formula, theories, and models that predict the type of emotion music portrays and recommends music according to the similarity of the songs. This includes Spotify, MusicCat, and other music sharing programs (Han, B. J., Rho, S., Jun, S., & Hwang, E.,2010 ;Shan, M. K., Kuo, F. F., Chiang, M. F., & Lee, S. Y., 2009)!

            So there you have it, music’s algorithm and its application in real life.

REFERENCES:

Balkwill L-L, Thompson WF, Matsunaga R (2004) Recognition of emotion in Japanese Western and Hindustani music by Japanese listeners. Jpn Psychol Res 46:337–349. 
Han, B. J., Rho, S., Jun, S., & Hwang, E. (2010). Music emotion classification and context-based music recommendation. Multimedia Tools and Applications,47(3), 433-460.Hunter, P. G., & Schellenberg, E. G. (2010). Music and emotion. In Music perception (pp. 129-164). Springer New York.Pallesen, K. J., Brattico, E., Bailey, C., Korvenoja, A., Koivisto, J., Gjedde, A., & Carlson, S. (2005). Emotion processing of major, minor, and dissonant chords.Annals of the New York Academy of Sciences, 1060(1), 450-453.Shan, M. K., Kuo, F. F., Chiang, M. F., & Lee, S. Y. (2009). Emotion-based music recommendation by affinity discovery from film music. Expert Systems with Applications, 36(4), 7666-7674.Scherer, K. R. (2004). Which emotions can be induced by music? What are the underlying mechanisms? And how can we measure them?. Journal of new music research, 33(3), 239-251.