Sunday, August 28, 2016

Pattern and Timing in Sex, Music and Narrative -- The Deep Structure of Aesthetic Experience

This long, somewhat rambling post explores the cognitive and physiological underlayer of various aspects of human experience and aesthetics -- touching three examples: sex, music and narrative.

Timing is Not Everything, But It's a Lot

It's sometimes said that in music, timing is everything.   

Now, this is often said in other domains too -- for instance, in business, timing is very important ... the right startup idea introduced too late or too early will have minimal or no success, but the right startup introduced at the right time can become a billion-dollar "unicorn."

But I think there is a core to the human psychology of timing in music, which is more specific than the observation that "timing is generally very important in complex systems."  It seems to me that the emotional impact of music on humans, depends on specific aspects of the human mind/brain's arousal and expectation subsystems, and the way they react to the ratios of different time intervals.  

Here I will fairly loosely sketch a model of the human mind's arousal/expectation subsystem, which I will call Timarcex (for TIMing of ARousal, Comparison and EXpectation).  And I will explain how Timarcex -- as it intersects with other aspects of psychophysiology, including pattern recognition in particular -- underlies various aspects of sexual, musical and narrative experience.

From a science view, this is mostly speculation, which ultimately needs to be backed up (or not) via empirical studies, simulation models, and so forth.

Timing and Sexual Arousal

Just for fun, let's start with sexuality.... 

Let's think about the role of unpredictable timing in sexual arousal … specifically in the arousal of a human being to the point of orgasm.  There has been a great deal written on this topic already; my goal here is to explain some of what is already well known, in a more fine-grained way based on how the human mind-brain handles timing related to arousal and expectation.   (The same basic dynamics apply to both males and females in this context, but they are generally more easily observable in females because of females' generally longer and more erratic paths to orgasm.)

For simplicity let us divide the path to orgasm into four phases: initial arousal, pre-orgasm, orgasm and post-orgasm.  Obviously this is a discrete approximation of an underlying continuum.

(Yeah, I do realize that other modalities of sexual experience exist, for example tantric sex in which one doesn't achieve traditional orgasm, but moves through different "altered states" of consciousness in which slower, deep waves of pleasure move through one.    The tantric sex experience is more complex and could also be analyzed cognitively and physiologically, but I won't get into that here.  My point here isn't to do a thorough job of cognitive-physiological sexology, but rather to use sexuality as a way of illustrating certain general points, that also come up e.g. in music and narrative and other areas.   Looking at these dynamics in the context of sex highlights their basic physiological/emotional nature.)

The first point I want to look at is the differential desirability of randomness at different stages in the arousal process.

In the pre-orgasm phase, extremely regular and predictable movements are generally desirable, as long as they’re pleasurable movements for the particular person involved.

But in the earlier phase, where arousal is still being built up, random variations in timing are desirable.  That is, a specific pleasurable motion every 5 seconds is worse than the same pleasurable motion delivered with a MEAN interval of 5 seconds but random variation of, say, 2 seconds variance on either side.  My suspicion is that this is true if the movements are delivered by a purely automated tool, but even more true if the movements are delivered by an agent that has emotions or is perceived to have emotions.

So, before going further, let's explore: What is this all about?  Why would it be better to start with more randomness in movement timing during initial arousal, and then introduce more consistency of timing during pre-orgasm?

To understand this, I suggest, we can model human pleasure as having at least three aspects:

·      Raw underlying pleasure
·      Comparison of pleasure to immediately-past pleasure
·      Comparison of pleasure to (e.g. cognitively) expected pleasure

To make things a little clearer, let's refer to "raw underlying pleasure" as pleasure0, and use the term pleasure1 to refer to the pleasure that involves the three components

·      Current pleasure0
·      Comparison of current pleasure0 to immediately-past pleasure0
·      Comparison of pleasure0 to (e.g. cognitively) expected pleasure0

We may then recurse, and define pleasure2 in terms of pleasure1, etc.    

We may also look at averages, e.g. define pleasure* as a weighted average of pleasure0, pleasure1, pleasure2, etc..... 

One can then also look at pleasure*1, defined from pleasure* just as pleasure1 is defined from pleasure0 -- one can even build up a whole ordinal omega hierarchy of pleasures -- but probably this is unnecessary in the human-psychology or early-stage-AGI context.  (I look forward to becoming an uploaded supermind that can experience multiple ordinal infinities of pleasure types....)

I suppose that pleasure* is about right as a grounding of the intuitive concept of "pleasure"; and I suppose that approximating pleasure* in terms of pleasure0, pleasure1 and pleasure2 is probably good enough for almost call cases.

In the sexual case, the "raw pleasure" pleasure0 is mainly physiological in origin.  In other cases, this same role may be caused by other factors -- e.g. by various system goals being fulfilled.   If a system has a strongly weighted goal of learning new things, and it learns something new and exciting, then this may cause a burst of pleasure, which we would categorize as pleasure0.   What makes an instance of pleasure an instance of pleasure0 is that it is not constructed recursively via comparison or expectation of pleasure.

So, let us return to the scenario above (building arousal, in the initial-arousal phase, via the tension induced from random timing variations in pleasure).  The key point here is that: When the pleasure comes a few seconds early, then one is getting MORE THAN EXPECTED ... which feels extra-pleasurable. 

For instance: perhaps a person expects to get a burst of pleasure0 after 5 seconds, but instead gets it after 3.  As our pleasure1 is increased based on (among other factors) the difference between the pleasure0 expected and the pleasure0 obtained, then getting e.g. pleasure0 at 3 seconds when one only expected it at 5 seconds is GOOD and will increase pleasure1, at that moment (the 3 second moment) when it is received.  

But what about when the pleasure comes a few seconds later than expected?  Well, by the time 7 seconds have rolled around, if one was expecting a burst of pleasure0 after 5 seconds, one is feeling pretty eager and disappointed and frustrated.  After all, the pleasure0 was supposed to come at 5 seconds, right?!  But wait a moment – that sometimes makes it all the more pleasurable when it comes along after 7 seconds!   In this case, the pleasure0 is not just delivering pleasure0 -- it’s delivering relief from frustration!   In terms of our model: When the pleasure0 comes after 7 seconds instead of the expected 5, then the component of pleasure1 that responds to the contrast of current pleasure0 with recent past pleasure0 kicks in, and causes a burst of pleasure1.  Also, the component of pleasure2 that responds to the contrast of current pleasure1 with past pleasure1 kicks in, and provides a burst of pleasure2.  Woo hoo!

On the other hand, what if the pleasure0 simply shifted from a regular schedule of every 5 seconds to a regular schedule of every 3 seconds or every 7 seconds?   Then
expectations would just get recalibrated – and the new “on time” would be 3 or 7 seconds instead of 5.   If 3 seconds is what’s expected, then 3 seconds isn’t excitingly soon anymore – 1 second is excitingly soon … and 7 seconds isn’t a tantalizing delay anymore… it’s a highly frustrating delay that starts making one bored or angry or confused…. 

The delay from 5 to 7 seconds may be just enough to create a little frustration, which is rapidly vanquished in a poof of release and pleasure2 alongside pleasure1 when the pleasure0 finally comes.   But a delay from 3 to 7 seconds might be soooo long that frustration builds up to the point where the pleasure0 doesn’t release it when it finally comes – perhaps the delay is long enough to trigger some cognitive process thinking “WTF is going on, is the pleasure0 ever gonna come again?”   This brings up an important point: the pleasure hierarchy described above is not the only thing going on.  There are also other emotions like anger and displeasure happening, with their own dynamics.  

One may look at a hierarchy of pain0, pain1, pain2, ... pain* , comparable to the pleasure hierarchy described above.   It may be that in people with some forms of pain asymbolia, pain0 is intact but the higher orders of the pain hierarchy are not operational, resulting in them knowing that a pain signal has been received by their body, but not reacting to this pain signal in the normal way.   The timing of the comparisons and expectations in the pain hierarchy may be different from those in the pleasure hierarchy.  Also, the relations between the pain-hierarchy internal timing parameters and the pleasure-hierarchy internal timing parameters may vary from person to person.

Getting back to the example -- given the particular parameters of typical human response, 3 seconds or 5 seconds between intense bursts of pleasure0  may be long enough for some anticipation to build and then get fulfilled.  If the pleasure came after half a second the experience would basically be one of pleasure never stopping – which in a way would be more pleasurable, but in a way would be less pleasurable, because the process of building an expectation and then getting it fulfilled would not be there… and at the cognitive rather than purely physiological level, pleasure is largely about expectations and their fulfillment and disappointment…

Now, human sexual arousal is not one-dimensional, there’s not just one “pleasure button” being pushed at various intervals. Rather, there are various different tissues which are stimulated by different things, and which get pleased, roused to anticipation, and become frustrated upon neglect – with different time-scales.  So what one has in a real situation of human early-stage sexual arousal is the above story, but on a variety of different tissues, each with their own characteristic time-scales.  There is a sort of symphony of anticipation, fulfillment and frustration, as at any given time some tissues may be satisfied, some may be anticipating, and some may be frustrated….

And then of course, in human sexual arousal, there is also an overlay of more complex emotions besides pleasure and its expectation and frustration.    When the other party seems eager and keeps getting ahead of what seems to be the natural rhythm, that may make one feel loved or at least desired; a lackluster or overly robotic rhythm may make one feel that the other party isn’t really into it; etc.  The arousal of these emotions is more variable among different people, whereas the basic logic of arousal via expectation and frustration is more generic.   A person’s specific physiological and emotional makeup, layered on top of the basic biology of human physical pleasure and the math of expectation fulfillment and frustration, add up to form that person’s personal sexual pleasure calculus…

So -- What would cause regularity of stimulation timing to become more desirable than random stimulation timing, when the initial arousal phase is passed and one enters into the pre-orgasm phase?   What happens, I hypothesize, is that, as the initial arousal phase proceeds, the physiology of the sex response kicks in, and on average each instance of pleasure0 becomes greater than the last.   In part this is due to the specifics of the human body's sex-related chemistry; and it part it's a generic aspect of human psychophysiology -- getting  more pleasure puts other things out of one's mind, it stops other cognitive and emotional processes ... but some of these other cognitive and emotional processes may have been causing stress, which may have been holding back the body from physically experiencing pleasure.   So quite generically, as pleasure escalates and occupies more of the mind's attentional focus, it drives out other stress-producing processes and actually increases the amount of pleasure0 that is physiologically experienced.

So then, as the initial-arousal phase continues, once the amount of pleasure0 per stimulus burst becomes large enough, the "current pleasure0" component of the pleasure1 becomes dominant, and the other components become more minor by comparison.   One reason not to have constant pleasure0 by this stage, is because of the underlying physiology, which experiences pain if stimulated for too long, but experiences pleasure if stimulated for a short time.   In this phase, the optimal strategy is pretty much for the body to receive as much pleasure0 as it can physiologically take, not messing around with mind-games regarding expectation and comparison.

By the orgasm phase, then, the pleasure0 is at a constant high level.  But interestingly, orgasmic pleasure is often experienced as moving through the body in waves.   A wave movement, with continuous rising and falling of intensity, still does provide comparison dynamics, with potential for the indirect terms in the definition of higher-order pleasure to play a role.

In the story I'm telling here, the logic of human sexual arousal toward orgasm follows from the pleasure hierarchy, with lesser contributions also from the pain hierarchy and other aspects of human response (such as the ability of pleasure to drown out stress-producing cognitive and emotional processes and hence increase pleasure response).  

While I have presented this Timarcex (timing of arousal, comparison, expectation) model here verbally, I am sure it would be possible to present it equationally as well.  If one did so, one would find that the particular parameters involved in defining the pleasure and pain hierarchies, would impact the particular nature of the sexual experience being modeled.

To construct such a mathematical model I would start with the synthesis of Joscha Bach's / Dietrich Dorner's Psi model and Scherer's Component Process Model , tthat we have been using to model emotion in OpenCog.   But to unfold these ideas in that context would require more depth and detail than would make sense to give in this (already over-long) blog post; this will be worked out and written about later.


I have discussed above "arousal" in a sexual context, where it has a certain specific meaning -- but the term also has a more general meaning in psychology.  In general, "arousal" refers to the propensity of a system to undertake (internal and/or external) activity.     This activity may be on the part of the whole system, working toward some goal or involved in some complex systemic attractor; or it may be on the part of various parts of the system acting independently.

Arousal is sometimes contrasted with relaxation, but this has to be done with care.  Arousal in the sense of high activity is of course an opposite of relaxation in the sense of low activity.   However, if "relaxation" means a lack of tension, then it is possible to be highly aroused and highly relaxed at the same time (tantric sex would be an example, as would certain other meditative states or flow states).   Similarly, arousal is not synonymous with "excitement" in the emotional sense; though in ordinary states of consciousness, the two do often correspond. 

In a state that is highly aroused and relaxed at the same time, the different parts of the mind-body system are carrying out coordinated activity in a manner that does not involve the various parts causing each other significant tension (no "knots"!).

In a sexual context, what generally happens is that pleasure0 increases over the course of an interaction, according to dynamics outlined above and others; and a high degree of pleasure0 is one way to effect a high level of arousal in human bodies, based on typical physiological mechanisms.   A similar dynamic occurs with music and narrative, though there the pleasure0 is often obtained via fulfillment of other goals (e.g. perception of novel patterns, empathy with a character) rather than via raw physiological pleasure.

Patterns and Pleasure

The next ingredient to introduce into the story is PATTERNS.   Continuing with the sex example for continuity's sake, consider that, in a sexual context, stimulation is not just a matter of delivering pleasure to one or another tissue; rather, pleasure may be delivered in specific patterns of touch geometry and intensity, and the body will then come to expect these patterns.   Examples would be a “French corkscrew twist” movement of one sex organ within or around the other; or a stroke of the hand repeatedly up and down the lover’s hip; etc. etc.  

Such a movement pattern will typically have a  certain timing, and can be accelerated or delayed, inducing dynamics of expectation, fulfillment, frustration, tension, etc.   Furthermore, movement patterns impact each other, as they may involve adjacent, overlapping or more subtly cross-connected parts of the body … so that fulfillment of one may cause deferral of another, etc.  Kissing on the neck faster and faster may cause what’s happening with the genitals to temporarily get slower and softer, etc. – an example of pleasure via acceleration in one place, correlated with pleasure via temporary frustration and then fulfillment in another place.

Patterns and Pleasure in Music

"I want to make love to my guitar, instead of just the same old everyday woman"
-- Jimi Hendrix

And this complexity brings us finally back to music.   In music we have many complex patterns unfolding through time, each eliciting expectations regarding what will come next.  The symmetries of musical scales and chords are instrumental in forming these expectations.  When a chord pattern has been repeating, one expects it will continue. When a melodic line has been going up a certain scale, one expects it will continue.  When a melodic line has been going up and down within a certain scale, one thinks it will stay within that scale.  When two melodic patterns have occurred, each immediately followed by their inversions … and then a third one occurs, one expects it will be immediately followed by its inversion.  And so forth.

So what we have in music is  a collection of interlocking patterns, each giving rise to expectations.  Each of these expectations may happen a little sooner or a little later than expected, thus causing extra bursts of pleasure in accordance with the basic logic of anticipation, frustration and pleasure.   Making one pattern continue sooner or more intensely, will generally cause some other pattern to continue later or less intensely — thus weaving a complex web of fulfillments and frustrations, which reminds us of the complex webs of fulfillment and expectation in our inner and outer lives.

A series of notes in time displaying an increasing amount of pattern, will often have the property of fulfilling an increasing number of the expectations elicited based on the previously occurring patterns.   So on the whole it will give the feeling of increasing pleasure – increasing fulfillment of expectation.  

But of course it’s hard for this to go on too long, because creating an increasing amount of pattern is difficult, assuming these patterns must have some emotionally evocative content in themselves, and that all the patterns are created via arranging the basic notes, chords and scales in the vocabulary of modern music.   So typically in a substantial piece of music, after a build-up in which the time-series of musical events generates a greater and greater density of patterns, things relax.   Here the intellect is being less stimulated, but if the relaxation comes with continuation of a pattern that is emotionally resonant, then the experience is still satisfying.   We see in this case the combination of the primal/visceral and the mathematical in music.   The mathematics of pattern emergence and tension buildup are critical, but there is also a simple matter of resonance with the rhythm of human bodily life – there is something innately satisfying, to anyone with a human body, about quieting down and proceeding at a slow regular rhythm after a huge burst of increasingly complexly patterned activity.

 Connecting Some Dots Regarding Musical Aesthetics

My introduction to the math of music aesthetics was Meyer's good old theory of "surprising fulfillment of expectations", from his book Emotion and Meaning in Music

A friend commented to me on Meyer's theory that it seemed to contradict the pleasure he got from listening to the same songs over and over again: "When I know very well where things are going (and where they'll end up), the strong emotion
can't really be connected with expectation and surprise, can it?"   However, I think he is not correct, because the working memory can still be surprised even if the long term memory knows what's going to happen.

The considerations I have presented here are conceptually consistent with Meyer's thinking, but add a lot of detail, and provide a broader perspective.  In particular, I would suggest that the expectations and the surprise we experience in music are a mix of
  • basic mathematical music-structure factors, as in the vein of combinatorial music theory etc. etc.
  • human psychology factors as modeled e.g. in Joscha Bach's work and in the Component Process Model
  • human body-rhythm patterns, to do with the way the rhythms of different subsystems of our physical bodies overlay on each other, which we perceive all the time sub and semi consciously and which influence our perception of music; obviously even classical music without a drumbeat is highly rhythmic...
  • more abstract cognitive pattern recognition

So a good melody packs up "surprising fulfillment of expectations" where the expectation and fulfillment are assessed according to a combination of the above 4 factors ... and if you leave any of the above 4 factors out you're going to get lots of "false positives" ...

The Timarcex approach, as sketched and hinted above, provides a way to model the combination and intersection of these factors in producing a human pleasure response.  

First, there is raw pleasure (pleasure0) coming from timbre and from rhythm -- e.g. a sound that reminds one of birdsong or a mother's voice, or a rhythm that reminds one of the heartbeat or the breath or the sound of feet pitty-patting, or the thunder in the sky, etc.   There is also raw pleasure coming from the recognition of complex patterns -- because humans appear to have "recognize patterns" as a basic goal, and any fulfillment of any of our basic goals.  

Certain auditory patterns are easy and natural for us to recognize, based on the nature of our auditory cortex.  Chords and scales have emerged historically because they are easy patterns for our brains to recognize.   Then, as we have heard them more and more, they become easier and easier to recognize, because we are recognizing new chordal and scalar patterns using brains that are already attuned to such patterns.   

It is important that what we have here is a system of patterns that gives the human mind a lot of novel potential patterns to experiment with and perceive, and also that a lot of these novel patterns are simple enough for the human mind to perceive RAPIDLY.    Mathematics, for instance, is also a system of patterns that gives the human mind a lot of novel potential patterns to experiment with and perceive; and it also builds up incrementally, so that the more math one knows, the easier it is to recognize more and more mathematical patterns.  However, the patterns involved in math are not FAST to perceive, unless you're Ramanujan (and OK, even for him not all the interesting mathematical patterns were fast to perceive).   Perceiving mathematical patterns generally involves thinking, and the timing required for this sort of thinking is hard to predict.  So making music involving complex mathematical patterns would be difficult.   But chordal and scalar patterns are complex and various in their own way, yet are also very rapid for the human mind/brain to perceive -- so they can be arranged in TIME, and can be used to elicit expectations that can then be toyed with in the context of Timarcex dynamics.  

I noted above that sexual arousal doesn't just involve bursts of pleasure0 delivered on particular schedules -- it involves complex patterns of pleasurable and (in some cases) painful stimulation delivered to various parts of the body in complex spatiotemporal patterns.   However, the imprecision of the movements involved makes it difficult to transmit or recognize highly complex patterns in this context.  The type of precision embodied in tonal music makes it very easy for the human mind to precisely recognize simple patterns, and then shift its attention to the complex patterns that come out of these simple patterns, etc.   A sufficiently intelligent mind could recognize even more complex patterns in purely atonal music; but tonality breaks sound down into discreta, which makes it easier for a limited-capability pattern recognition system like the human mind-brain to recognize a diversity of complex patterns.  

Music seems almost unique among human activities in that it involves the delivery of reasonably complex patterns, that are designed to be recognized fast enough by the human mind-brain that they can be part of Timarcex-type dynamics, similar to those involved in sexual arousal.   This analogue has been parodied many times -- for instance, Frank Zappa used to make fun of his keyboardists when they ended every solo by moving their fingers faster and faster toward the high notes at the right of the keyboard.  Zappa felt they were emulating an orgasm, emulating the blow job they hoped to get from some groupie after the set was over.   But Zappa's parody, as was often the case with him, had some accuracy to it.    Many jazz or instrumental-rock solos do follow a pattern of

  • Beginning with a predictable pattern, in timing and often also in melody
  • Then deviating more and more -- providing more Timarcex-based pleasure, based on variations in timing as well as based on introduction of new patterns
  • Eventually, once the pleasure has built up enough and the listener's mind is overcome with the joy of the solo and isn't paying attention to anything else, returning to the original melodic theme.  Sometimes the original melodic theme is returned to in timing before melody; sometimes in melody before timing.

The pattern of playing  variations on the melodic and timing themes of the solo FASTER and FASTER toward the end, plays on the fact that by this point these patterns are extremely well imprinted in the listener's mind, so they can recognize them clearly even if they come really quickly -- and that by this time, the Timarex parameters are at settings where rapid bursts of pleasure0 from pattern recognition will give a lot of pleasure1.   Escalating speed in this context can lead to increase in pleasure2 and pleasure3 as well, etc.

Obviously what I have so sketchily and awkwardly described above is just one way that some solos emerge -- there is a huge diversity to the psychological structure of different jazz and instrumental-rock solos, and doubtless the same is true in other genres of music with which I am less familiar.  One could create vast theories of psychomusicology associated with particular musical genres, unfolding the intersections between elicitation and recognition of various sorts of patterns on various levels, and the "psychological timing games" I have labeled Timarcex and rambled about above in the context of sexual arousal.

Surprising Fulfillment of Expectations as a Case of Increasing Structural Complexity

These considerations also relate somewhat to Juergen Schmidhuber's theory of beauty as increase in compressibility, see

Schmidhuber's theory is not quite right but it's insightful nonetheless.  A deviation from the norm that sounds awesome usually is one that causes some meaningful but
not-wholly-expected pattern to come about ... i.e. an increase in compressibility ...

However, my view is that an increase in the number of compressing patterns
is perhaps more important than an increase in compressibility  (or more properly, an increase in the total amount of compressing pattern ... the "structural complexity" as I called it way back in my first book "The Structure of Intelligence," written in 1989-90, published in 1991.  Structural complexity is total amount of compressing pattern, but subtracting off for overlaps in a certain way.)

"Surprising fulfillment of expectations" in a musical context means:

·      the previous expectation is fulfilled, i.e the previous compressing pattern continues
·      a new, surprising factor is found ... but probably one that also leads to a new pattern in the overall perceived music, hence a new/different compressing pattern

So surprising fulfillment of expectations is one route to get to "increase in the number of compressing patterns" ... while also playing nicely with Timarcex dynamics.

Within familiar human music genres, "music theory" constrains the set of ways that one can get increase in the number of compressing patterns, without getting too complex.    But this kind of constraint is necessary in a music context given that, for music to be music, the pattern recognition involved needs to be done in real time.

The Structure of Narrative

Having dispensed with sex and music (well OK not quite, but hey --), let us now turn to narrative structure.   The classical way of structuring a screenplay, or almost any story, is based on a structure like:

·      Protagonist is introduced
·      Protagonist meets challenge
·      Protagonist overcomes challenge, ideally in a somewhat surprising way
·      Protagonist is seen to grow and change as a result of overcoming the challenge

In a movie we can have real-time Timarcex effects like in music or sex --- especially in an action scene or a sex scene, where the timing may be carefully worked out to play on the viewer's Timarcex dynamics.

In a book-reading experience we don't have "real time" dynamics like in music, but there is still an imaginal time axis: for instance, we know that events are occurring within a human's lifespan, or within a limited time-frame involved in a particular situation.   Timarcex dynamics can be elicited in the reader's mind, relative to these imaginal time axes -- because while reading, the reader may be replaying the already-known and predicted activities of the protagonist in their mind.  If the reader is replaying the life of a character in their mind, over a 3-7 second period, then changes in the life of the character will change the nature of this 3-7 second "inner semantic movie", which may cause Timarcex-related effects.    

So -- Even if a character's love affair unfolds over 10 years in the imaginal time-axis of a novel, the reader still reviews this whole 10-year affair in a few seconds in their mind -- so when the reader learns about a new triumph or a new frustration, they experience it as a satisfaction or frustration or surprise within their few-seconds replay.   In this way the experience of reading narrative is subtler than the experience of listening to music or watching an action or sex scene in a movie -- the Timarcex aspect is there, but it relies on the construction of short-term internal temporal experiences based on imaginal time-axes semantically inferred from the narrative.

A common experience when experiencing a narrative is to empathize with the protagonist.   (This is not the only possibility, obviously.  As an example of an alternative, one may instead envision oneself in some sort of relationship with the protagonist.)   One, at least temporarily and while experiencing the narrative, almost can't help to some extent "putting oneself in the protagonist's shoes."   This is how human beings are wired -- we understand other human beings in part by modeling their thoughts and behaviors using our neural "mirror system", which re-uses much of the neural wiring we use to be ourselves, to model the other.

The basic pleasure0 of reading about a character meeting a challenge and overcoming it, has to do partly with our identification with the character: We feel that we, to a certain extent, are facing the character's challenges and then overcoming them.   (We can see this even in lowbrow forms of literature like pornography.  The porn novel or film almost always begins with at least a slight amount of framing.   Instead of going straight into the "old in-and-out", we see the people involved first ... perhaps we see a woman walk into a man's house, and he greets her and she says something and then she takes her clothes off and they get down to it.   We have even here an element of the traditional narrative structure -- the challenge of getting the hot woman to screw the guy is posed, and it is met immediately in a satisfying way... so the viewer can feel like "I am that guy who was just sitting in his house, saw that beautiful girl walk in, faced the unconscious challenge of whether I could get her to screw me, and then HEY! immediately surmounted that challenge and succeeded!")

But a story in which a character simply meets a challenge and then straightforwardly overcomes it is not that interesting.   Overcoming challenges in a surprising way makes a much better story.  And so we are back to "surprising fulfillment of expectations" again.   Just like in the initial-arousal phase of sex, just like in music.  

A great story has an increase in structural complexity as it goes along -- more and more patterns get built up subtly and then gradually or suddenly recognized by the reader.   But for Timarcex to help, the build-up of structural complexity has to happen such a way that, when the reader reviews the events of the narrative in their mind as they read/view/listen, the introduction of new events causes various levels of pleasure in a timing-specific way, just like in sex or music.   The surprising fulfillment of expectations is one way to provide this.

Losing Oneself, Gaining One's New Self

The final stage of the typical narrative, in which the protagonist changes and grows as a result of meeting their challenge, analogizes with sexuality and music in a subtle way related to the persistence and transcendence of the self.    The nature of sexual arousal is such that, by the time orgasm is reached, the pleasure0 is so overwhelming that the self-model is entirely pushed out of the mind's attentional focus.  One "loses oneself" in the experience.  If one is relaxed enough or passionate enough, this may happen early in the experience, well before orgasmic release (or in tantric sex, one loses the self via meditative-type activity, and the pleasure0 and mutual-giving-taking aspects of the sexual activity then push the meditative state in certain directions).

In music, one can also lose oneself for various reasons.  Rhythm can do it: entraining the body in a rhythm can make the internal body movements so emphatic and overwhelming that everything else gets pushed out of the attentional focus.   Sufficient evocation of particular emotions by a melody can do it -- if the emotion evoked becomes strong enough.   If a melody evokes the habitual pattern of rise and fall of a particular emotion, and the rhythm of the music simultaneously emulates the body-rhythms often experienced while experiencing that emotion, then a quite overwhelming, self-suppressing music-appreciation experience may ensue.   A sufficiently complex combination of musical patterns can also do the trick -- if there is enough emotional meat there to attract the listener's mind, then complex patterns playing around the emotional structure can absorb the listener's processing power.    Some music combines strong melodic and rhythmic emotion elicitation with processing-power-exhausting rapid-fire complexity (the end of One Word by the Mahavishu Orchestra comes to mind, or many passages in Coltrane's "A Love Supreme" or Buckethead's  "Hold Me Forever" ... or Hendrix's "Red House" solo on "In the West" -- etc. etc. etc.).

In the traditional narrative structure, the protagonist loses aspects of their own self and gains a new and better self.  The self-model is suppressed and overcome.   After orgasmic or tantric sex, and after listening to an absorbing piece of music, you emerge a new and different person as well -- more peaceful, or more excited ... in any case, your emotional parameters are re-set.   Part of the overall process of having sex, or listening to intense music, or being a character in a story (or empathizing deeply with a character in a story), is that after all the Timarcexish pattern play is over, you emerge in a quite different state.

The precise state that a narrative leaves one in, after the experience is done, obviously depends on a lot of things.   Hollywood movies and romance and action/adventure and mystery novels have an obsession with "happy endings" -- they want to leave you in a state where stress and challenges have been overcome to reveal joy ... both because leaving you in a state of joy is a good thing, and because this leaves a frame of "bad things will be overcome by good things" in your mind, which may guide you to have more positive experiences later on.   On the other hand, tragedies are also possible, and some critics (e.g. Nietzsche) have felt they are morally and aesthetically superior.  As Nietzsche saw it, tragedies remind us that life is great in spite of the bad parts, that everything is connected and to embrace the good in life we also have to embrace the terrible.

Similarly, some music ends all of a sudden, leaving you dazed and confused -- other music lets you down gradually, bringing you back to Earth with some slow, soothing melodies after the fast, crazy ones have passed.  You can make up the sex analogy for yourself...

Brief Summary (Enough for Now!)

OK -- I don't really have a grand conclusion here but I've run out of time for writing just now.  (I'm on a flight from Istanbul to Addis Ababa and am going to land now....).

So lacking a grand conclusion I'll re-summarize ;)

I suppose the basic point is that human aesthetic experience is rooted in basic human emotional experience.   And human emotional experience is grounded in what I've labeled Timarcex -- the particular way we construct hierarchies of pleasure and pain feelings based on comparison and expectation.   And our expectations are based on what patterns we can recognize, given the time and resources available and given the fund of patterns already in our minds (and the specific set of patterns elicited via our recent experience).

Finally: strongly time-bound aesthetic experiences like sex and music and action-scenes in movies, are founded on systems of patterns that can be recognized rapidly by the human experiencer, so that instances of pattern recognition and expectation can be arranged judiciously in relation to the Timarcex parameters of the human experiencer.   Less time-bound aesthetic experiences like reading a novel are founded on surprising fulfillment of expectations in different ways that aren't so tied to Timarcex parameters --- but are also to some extent based on short-term emotional experience guided by Timarcex parameters, given that human consumers of narratives are often rehearsing narratives in their minds in a rapid way ... so the mental rehearsal of a narrative becomes more time-bound like a fragment of musical or sexual activity.

How Specifically Human is All This?

An interesting question from an AGI point of view is: How much of all this is just specific to human beings, versus how much has any sort of meaning from a more general perspective?   At first blush it seems to be a complex mix.  

Experiencing satisfaction based on a change in the amount of raw satisfaction -- this is just temporal difference learning, basically.

Experiencing satisfaction based on expectation of future satisfaction -- this is basic to intelligence as well.

So in a broad sense, it seems the Timarcex pleasure and pain hierarchies are fairly general, natural aspects of intelligence.   However, the specifics of how Timarcex is implemented in the human brain, seems not to have any extremely special power where intelligence is concerned.  Indeed, one suspects humans could become more intelligent if they had more explicit cognitive control over the parameters of the relevant subsystems of their brains.

The importance of "surprising fulfillment of expectations" seems to come from the intersection of Timarcex with the more basic goal of "increasing structural complexity."   The latter basically means "wanting more patterns."   Arguably, the quest to experience more and more patterns is a foundational aspect of intelligence.   If so, then the quest for surprising fulfillment of expectations would seem to be an inevitable aspect of intelligent goal-seeking systems.

So what we have here in sex, music and narrative are -- surprise, surprise -- specifically human manifestations of some generic aspects of general intelligence.

Quod Et Handwavium, once again...

(Hey, was that a grand conclusion?  Maybe, sort of, after all....)

1 comment:

Desmond said...

Maybe I've overlooked, but what about chemical and physical aspects of the subsystems? It takes some time for synapses to reset, hormones to flush, as well as pleasure receptor tissues to reset, light receptor cones to fire again, etc. Also, it takes a while for a narrative to play out (either when read, listened to, or watched). Music needs language of the instruments and melodies to play out during expectation setup.