Using IBM Watson Text Sentiment Analysis, I analysed the “Hot 100 Chart” of Billboard to see if there is anything to be said about the aesthetic preferences of hit music listeners in terms of the psycholinguistic properties of lyrics. Turns out that showing strong emotion can partially account for chart success. Full report is on the way..
In this post follows my article “A Different Definition of Music”, an earlier version of which had been published with the greek edition of Mensa Magazine, in April 2011.
A Different Definition of Music, and Applications in Research
Examining music, one can observe that one of it’s most fundamental characteristics is the activation of the listener’s imagination, and especially the spontaneous creation of mental imagery. But what is not instantly recognized is that the potential plasticity of that imagery differentiates music from most of artificial mental entrainment means (arts), as one of Passive Creativity. The present article discusses the use of music as a mean of studying the behavior of brain entrainment, and therefore the way of meaningful mathematical research of high creativity as manifested in the “masterpieces of art”.
A Different Definition Of Music
If we were to put the collective understanding of music in one sentence, this could be averaged in the next, generally acceptable definition: “Music is the art form whose mean is the sound”. Likewise, “Art is the product or the procedure of the deliberate organization of symbolic elements, in such a way that it affects and influences one or more of the emotions, senses and cognition.” Even though this definition is effective in giving information about music, I personally find it, along with the rest of definitions that I have read, somewhat deceiving regarding the essence of music. This definition fails to address that we subconsciously categorize the sonic stimulus as music, if and only if it succeeds in what the definition of art describes about the affecting and influence. With this in mind, the term “music” is just a description of the stimulus, totally connected with the quality of its entrainment. For logical reasons, we should transfer this description of the stimulus to a description of the quality of the mental entrainment that the stimulus induces. To make this more clear, let’s think of this example: If we somehow find a man in our house that has solved a plumbing problem that it had well enough, it’s reasonable to just think that “this is a professional result”, and not as much to also hypothesize that “this man works as a plumber”. So, we conclude to a new definition for music, which is: “Music is the aesthetically mentally significant result of hearing”. This means that music is not the product described in the definition of art, but in reality, it’s the product of this product (and not only that). In other words, it’s not the sound that is considered as music, but the special cognitive result that sound has on the brain.
With this definition of music, it becomes obvious that it’s existence is not described sufficiently binarily (“This is music” / “This is not”). In truth, music becomes a variable belonging in a numerical set of values that count the power of the stimulus-produced “mental entrainment”.
On The Subjectivity Of Quality
It is common knowledge that a song doesn’t have a specific “quality level”, but that it depends on the listener’s preferences. I think that this idea is not right. What depends on the listener’s preferences is only the perceived percentage of quality, and not the quality per se. At first, let us realize that the mental paths that a specific sound stimulus tries to induce are the same for everyone. We know that the emotional direction of a major chord generally does not differ between people, and also that some chord progressions “make more sense” than others. More generally, it could be stated that even emotionally complex structures, such as melodies, also intend to influence emotion in specific ways, with their efficacy as predetermined as the one of a chord progression. Despite this fact, the percentage and it’s distribution over listening sessions, regarding entrainment that a music piece produces, can differ completely from one person to another. For example, the attention that someone will give to the stimulus, their mental capability, their psychology, their experiences, what is “hot at the moment” in the music culture, and their level of ability to appreciate new music by the first listen, are only some of the factors that significantly influence the percentage of perceived quality, and make it vary a lot between people and between musical pieces. While it is generally easy to feel how good or bad a chord progression is, the exponentially more complex nature of a melody, or -worse- a full piece, even while fully predetermined is so hard to comprehend, that given the aforementioned factors of influence we are led to believe that ultimately there is no prefixed quality in the composition, but rather, that it is only a matter of personal preference.
Essence Over Numbers
Research using maths and statistics on what makes music uplifting, what differentiates a “hit song” from the rest, and so on, has always been popular. We often read titles on media like “Science found the secret of Beatles’ Music”, “Researchers discovered which are the best songs to make you feel happy” etc. Which initially sounds interesting, but scientifically, they generally give little or no new information. Most of the time it is about some simple statistical comparisons between songs, for example “Songs that people classified as happy were faster, were in major key and were more high pitch than average.” Not new for most songwriters. Or, conclusions like “Top 40 songs had a median note frequency of 440Hz, higher than the median note frequency found in songs that didn’t make it to Top 40”. But this information is hardly interpretable. While this kind of research is always interesting, I think that in order to give more important results, it requires that the researcher will first have a clear hypothesis in mind, and work in relation to it. And this hypothesis will be formed through the researcher’s personal thoughts on the mechanisms of the song in the first place. Also, in the one or the other form, it will always be concerned about the concept of aesthetic quality level. We should, then, find a way that enables us to confidently make assumptions that could possibly lead to important results regarding our understanding of aesthetic quality in music.
Switching Gears With Passive Creativity
Certainly, due to the factors discussed in the previous section is, of course, very hard for us to state that we could potentially fully realize the quality of a musical piece. Someone will then wonder: “If a song can mean something entirely different for each person, how could we say that a specific song has a very specific quality, and how that Song X is <<objectively better>> than Song Y?”. The answer is, that we definitely would have to find a way to increase our current capacity of music perception a lot, in order to overcome our personal biases and limitations that hinder our level of understanding of music quality in a composition. But what is this way? The answer on this occurred to me many years ago, when I was a high school student. I gradually developed a technique that helped me utilize the brain’s mental imagery forming ability, in order to completely change the way I perceived music while listening to it. This technique has commons with the widely known “speed reading” technique. When we are reading text, we tend to read in a word-to-word, line-to-line fashion, while verbally repeating the words we read with our inner voice. But people who have developed the ability of speed reading are able to cross the page with their eyes, massively collect the words through their peripheral vision, bypass the “linear” aural processing, and directly utilize the “multi-band” mental imagery function. This means that when they read the word “war”, instead of silently saying “war”, like most of us do, they instantly see a war-themed picture in their minds, which takes a lot less time to happen. With this speed of word intake, they are able to read exponentially faster than the rest of us, and still get the meaning correctly through vivid mental pictures. It is the same change of processing way that we can implement while listening to music, to rise our comprehension of the quality of the piece exponentially, and give our research and maths a clear direction toward more meaningful results.
The Active Listening Technique
Just as we read text without special mental preparation, like speed readers do, while listening to music we also don’t do special preparation for the session. But this is what we have to do, to vastly grow the amount of information we intake, and sufficiently decode music quality by utilizing imagery function, using what I call “Active Listening”. This is how the Active Listening technique works: At the start of a listening session, we close our eyes and set a random mental image in our mind, be it a specific real-life picture, or abstract shapes, or anything we can imagine (the starting point of imagery does not matter). As we have set this image and press “Play” to start the music, the aural stimulus influences it through its duration, while we concentrate on our mental imagery. In this setting we observe the movement and the evolution of the mental pictures, as they occur automatically because of the stimulus, forming a kind of mental imagery movie. These mental imagery moves are like the ones we make when we try for example to “move variable X to the other side of the equation”. The difference is that the stimulus does it for us, and we just observe the moves, and it happens in a random visual space instead of a piece of paper with equations on. The more complex and dense these moves are, the more mental entrainment occurs, and the more aesthetic quality the music has. (Remember the definition of music we gave at the start.) This is how we can make music make more clear sense to us, and know what a music piece has to give and why, before we start comparing data and building models -as harsh as this could sound- almost in random. This is how we can concentrate on examining structures and factors inside a melody or piece that we already know they make sense regarding it’s music quality, and then test our hypothesis per usual. At this point I would like to provide my email address, in order to collect experiences of people who tried the Active Listening technique. My address is email@example.com, and I welcome you to mail me your results of trying to fully experience the quality of a music piece by using it.