music cognition, anticipation
I'm thinking about anticipation in music. I'll probably use anticipation to mean several things in this entry, but it's part of the process of thinking of more nuanced definitions and explanations.
My idea is to measure the link between musical anticipation and emotional response and how emotional response evolves with the evolution of anticipation. The idea being that roughly three phases may be identified in musical anticipation of a particular piece of music. In the first phase, listeners are acquiring 'an ear' for the song: they are learning the patterns, the structure, and incorporating the music into the musical repertoires in their mind. During this phase, which may be extremely short or non-existent for music incorporating highly familiar structure, there is not yet a lot of anticipation in the listeners mind. During the second phase, the music has become familiar, anticipation is present, and developing, and listeners reach their peak enjoyment with a piece. During the last phase, the music has become stale, anticipation brings less and less enjoyment (if anticipation is the correct word to use here). In some cases, emotional response may become almost non-existent. I need to operationalize what I mean by anticipation.
So to measure the link between musical anticipation and emotional response, we obviously can't just measure emotional response and say it maps anticipation. We need independent measures. For example, we could measure familiarity with a song by asking participants to finish humming the melody, or based on the number of times the song is played (although this is a very crude measure). We could ask listeners to hum along whenever they knew the tune, by recording this (and matching it to the actual tunes), we can get an appraisal of familiarity. Familiarity gives us an idea of the evolution of anticipation - unfamiliar songs that are not hummed are in phase 1, hummed songs may be in phase 2 or 3. Differences in the style and accuracy of the humming may help sort. Additional measures of familiarity may be measured - for example, post experiment, participants may be asked after an interview to hum the songs they can remember - those on their mind would be hummed. Perhaps there are other measures standardly used for musical familiarity?
So familiarity gives us an idea of anticipation. For more complicated songs, different voices could be played in isolation - the more voices participants hum in isolation, the more familiar the song. Next, emotional response. This may simply be a button that participants press when they have a strong emotional response. There may be a sliding scale participants use for the strength of the response. If the experiment runs for multiple days, the scale may be normalized by the responses each day (although this may obscure a tendency to grow tired of music or appreciate it more over time.) So we get a measure of emotional directly. I suppose their are physiological measures as well.
Passages can not be too long, lest results are obscured by participants simply losing interest and not paying attention to the music. Passages may be heard multiple times in one trial and subsequently, in another trial on another day, to gauge how the familiarity and emotional responses changes with exposure in long and short term.
It seems to me there are some evolutionary mechanisms working in the development of a person's relationship with a song.
There is first the positive feeling associated with recognizing something - and this incorporates the "point blank" positive feedback in the brain for recognizing something as well as the positive feedback in recalling (good?) memories associated with that thing. These two levels of positive feedback may in fact be part of the same positive feedback mechanism, especially if the feedback from memories is positive regardless of whether or not they are associated with positive events. So there is a positive feeling associated with recognition which is consistent with preferring music that sounds like other music one is exposed to, and preferring a particular song the more times you listen to it (up to a point).
Another mechanism at work is something that suppresses interest after something has become well-ingrained in a person's mind. Well, I should be careful here. Perhaps better recognition of some elements makes possible the recognition of previously unfamiliar elements. So our musical preferences don't just change, they evolve.
One model we could set up looks like this: For a given musical passage, there is a period where we are learning to recognize the patterns and find proper analogies between emotional states and the musical passage. Once we like a passage, we find its patterns and analogies increasingly recognizable. Perhaps we no longer see the passage some combination of patterns, but increasingly as a single element, a cliche, at which point its power as an analogy with some emotional state is atrophied. Somehow an atrophy occurs. But I don't think that passage no longer serves any function - it now serves as a point of departure - because we now more easily recognize similar passages.
So the positive feedback for recognition should never go away, so there must be some other element of a musical passage that contributes to our enjoyment, lest our enjoyment would never atrophy. I'm thinking that perhaps there needs to be some negative affect somewhere in the process of listening that has the net effect of increasing our enjoyment of a piece. This goes back to the idea of contrasting valence of emotion between the amygdala and the frontal cortex - a perceived neg. emotion in the amygdala and neutral or positive assessment in the frontal cortex. It takes a lot for the amygdala to become familiar with a pattern - and as long as a part of the brain is unfamiliar with a pattern, it will give it a negative appraisal. Meanwhile, the frontal cortex has become familiar with a pattern in music - which one would verify by agreeing that they know a piece of music. But to really know a piece of music, one's amygdala must also know it.
I wonder to what extent the amygdala is capable of processing a pattern in music, as it is such a fast processing unit.
My idea is to measure the link between musical anticipation and emotional response and how emotional response evolves with the evolution of anticipation. The idea being that roughly three phases may be identified in musical anticipation of a particular piece of music. In the first phase, listeners are acquiring 'an ear' for the song: they are learning the patterns, the structure, and incorporating the music into the musical repertoires in their mind. During this phase, which may be extremely short or non-existent for music incorporating highly familiar structure, there is not yet a lot of anticipation in the listeners mind. During the second phase, the music has become familiar, anticipation is present, and developing, and listeners reach their peak enjoyment with a piece. During the last phase, the music has become stale, anticipation brings less and less enjoyment (if anticipation is the correct word to use here). In some cases, emotional response may become almost non-existent. I need to operationalize what I mean by anticipation.
So to measure the link between musical anticipation and emotional response, we obviously can't just measure emotional response and say it maps anticipation. We need independent measures. For example, we could measure familiarity with a song by asking participants to finish humming the melody, or based on the number of times the song is played (although this is a very crude measure). We could ask listeners to hum along whenever they knew the tune, by recording this (and matching it to the actual tunes), we can get an appraisal of familiarity. Familiarity gives us an idea of the evolution of anticipation - unfamiliar songs that are not hummed are in phase 1, hummed songs may be in phase 2 or 3. Differences in the style and accuracy of the humming may help sort. Additional measures of familiarity may be measured - for example, post experiment, participants may be asked after an interview to hum the songs they can remember - those on their mind would be hummed. Perhaps there are other measures standardly used for musical familiarity?
So familiarity gives us an idea of anticipation. For more complicated songs, different voices could be played in isolation - the more voices participants hum in isolation, the more familiar the song. Next, emotional response. This may simply be a button that participants press when they have a strong emotional response. There may be a sliding scale participants use for the strength of the response. If the experiment runs for multiple days, the scale may be normalized by the responses each day (although this may obscure a tendency to grow tired of music or appreciate it more over time.) So we get a measure of emotional directly. I suppose their are physiological measures as well.
Passages can not be too long, lest results are obscured by participants simply losing interest and not paying attention to the music. Passages may be heard multiple times in one trial and subsequently, in another trial on another day, to gauge how the familiarity and emotional responses changes with exposure in long and short term.
It seems to me there are some evolutionary mechanisms working in the development of a person's relationship with a song.
There is first the positive feeling associated with recognizing something - and this incorporates the "point blank" positive feedback in the brain for recognizing something as well as the positive feedback in recalling (good?) memories associated with that thing. These two levels of positive feedback may in fact be part of the same positive feedback mechanism, especially if the feedback from memories is positive regardless of whether or not they are associated with positive events. So there is a positive feeling associated with recognition which is consistent with preferring music that sounds like other music one is exposed to, and preferring a particular song the more times you listen to it (up to a point).
Another mechanism at work is something that suppresses interest after something has become well-ingrained in a person's mind. Well, I should be careful here. Perhaps better recognition of some elements makes possible the recognition of previously unfamiliar elements. So our musical preferences don't just change, they evolve.
One model we could set up looks like this: For a given musical passage, there is a period where we are learning to recognize the patterns and find proper analogies between emotional states and the musical passage. Once we like a passage, we find its patterns and analogies increasingly recognizable. Perhaps we no longer see the passage some combination of patterns, but increasingly as a single element, a cliche, at which point its power as an analogy with some emotional state is atrophied. Somehow an atrophy occurs. But I don't think that passage no longer serves any function - it now serves as a point of departure - because we now more easily recognize similar passages.
So the positive feedback for recognition should never go away, so there must be some other element of a musical passage that contributes to our enjoyment, lest our enjoyment would never atrophy. I'm thinking that perhaps there needs to be some negative affect somewhere in the process of listening that has the net effect of increasing our enjoyment of a piece. This goes back to the idea of contrasting valence of emotion between the amygdala and the frontal cortex - a perceived neg. emotion in the amygdala and neutral or positive assessment in the frontal cortex. It takes a lot for the amygdala to become familiar with a pattern - and as long as a part of the brain is unfamiliar with a pattern, it will give it a negative appraisal. Meanwhile, the frontal cortex has become familiar with a pattern in music - which one would verify by agreeing that they know a piece of music. But to really know a piece of music, one's amygdala must also know it.
I wonder to what extent the amygdala is capable of processing a pattern in music, as it is such a fast processing unit.