What if preference is not a single answer?
What if “I like this” is only the final surface of a much richer event — one in which the eyes, the body, attention, memory and conscious choice do not always say the same thing?
This question has slowly emerged at the intersection of several projects at Delvira Art n Tech.
SYNTHESIS began as an experiment in evolutionary design: a person enters a space of generated forms, attention becomes part of the selection process, selected forms breed and mutate, and eventually one of them can become a physical object. The original question was simple and seductive:
What if attention could become a fitness function?
But almost immediately, a harder question appeared.
Does looking mean liking?
Not necessarily.
A person may look longer because something is beautiful.Or because it is confusing.Or unfamiliar.Or disturbing.Or strangely magnetic.
Likewise, a quickening heartbeat may accompany attraction — but also surprise, uncertainty, effort or tension.
The more carefully we look, the less aesthetic preference resembles a single signal.
And that changes what SYNTHESIS might be for.
From detecting preference to observing its formation
The naive version of a biometric creative system would be easy to describe:
heart rate rises → the person likes the formgaze lingers → the person prefers the form
But a human being is not that simple.
Suppose one form repeatedly pulls the gaze back toward itself, but the person never consciously chooses it.
Another form produces a strong bodily response, yet the participant says, “No, I don’t like it.”
A third is selected almost immediately, without long inspection or noticeable arousal.
Which one is the true preference?
Perhaps that is the wrong question.
Perhaps the more interesting object of study is the relationship between the signals.
Aesthetic preference may not be a single signal at all, but an emergent pattern across:
attention, physiological activation, and conscious choice.
SYNTHESIS can therefore ask not only what someone chooses, but:
Where do these layers converge?Where do they diverge?And what becomes visible in the gap between them?
This reframes the system completely.
SYNTHESIS is no longer merely an evolutionary design tool.
It becomes an instrument for observing preference while it is forming.
The silent body and the speaking self
This is where SYNTHESIS meets Matilda.
The Matilda archetype began from a different problem: there are things inside us that are difficult to access directly. When an inner process is personified — given a name, voice and position outside the everyday “I” — material that was previously inaccessible can become available to consciousness. Matilda functions as a deliberate inner interlocutor: a constructed Other through which the self can hear itself differently.
In Matilda Meets the Machine, the parallel becomes explicit.
Matilda externalizes inner meaning through language.
SYNTHESIS externalizes nonverbal response through form.
One listens to the speaking psyche.
The other listens to the silent body.
One may return a sentence you did not know you knew.
The other may return a form you did not know you were attending to.
And yet neither should be mistaken for an oracle.
Matilda is not an infallible voice of truth.
A biometric signal is not a verdict about who you “really are.”
A pulse is not a self.
A gaze is not a preference.
A generated object is not a diagnosis.
Each is better understood as a mirror: a way of making one layer of experience visible enough to examine.
The moment the mirrors disagree
And perhaps the most interesting moment is not when all the mirrors agree.
It is when they do not.
Imagine a participant leaving a SYNTHESIS session saying:
I thought I was choosing what I liked.But my body responded to one form,my eyes kept returning to another,and consciously I chose a third.
That is not noise.
That is information.
The divergence may reveal that the word “liking” is too coarse.
Maybe one form created calm.
One created fascination.
One created desire.
One created cognitive friction.
One felt familiar.
One felt alive.
One was chosen because it was beautiful.
Another was never chosen, yet remained impossible to ignore.
The task then becomes not to force all these responses into one score, but to learn to distinguish them.
And this is where SYNTHESIS touches another Delvira idea:
the resolution of perception.
What does “resolution” mean here?
In The Resolution of the Soul, the central argument is that understanding depends on distinctions.
A person with only the category “I feel bad” has low resolution.
A person who can distinguish grief from shame, fatigue from anxiety, envy from loneliness, has a finer vocabulary of inner experience.
The outer world has not necessarily changed.
The number of distinctions available to perception has changed.
The same principle may apply to aesthetic experience.
At low resolution:
I like it / I don’t like it.
At higher resolution:
I am drawn to it.It unsettles me.It feels familiar.I admire it but do not want it.I keep looking at it without knowing why.My body reacts, but my conscious taste rejects it.I would choose it for the room, but not for myself.I do not find it beautiful, yet I cannot forget it.
The object has not necessarily become more complex.
The observer has acquired more axes along which to perceive it.
That is resolution.
The blind spot of attention
There is another important connection.
The Resolution of the Soul distinguishes between two failures of understanding.
Sometimes we cannot perceive a distinction because we do not yet possess the vocabulary for it.
But sometimes the distinction is available — we simply do not look there.
The article calls this the blind spot of resonance: the capacity exists, but attention habitually passes over it.
Now imagine SYNTHESIS not merely recording where someone looks, but revealing patterns in what repeatedly escapes conscious selection.
A participant may say:
“I never choose these forms.”
Yet across several generations, their gaze repeatedly returns to the same morphological quality.
A particular curvature.
A density.
An asymmetry.
A surface tension.
A relationship between void and mass.
Suddenly the system can ask:
Is this preference absent?Or has it remained outside the participant’s conscious vocabulary?
That is a much more interesting question than “Which shape won?”
SYNTHESIS as a perceptual training ground
This suggests a broader purpose for the project.
SYNTHESIS can become a training ground for the resolution of perception.
Not because the machine tells a person what they truly like.
But because it creates conditions in which the person can compare several normally hidden layers of response.
The participant can observe:
what I consciously choosewhat keeps attracting my attentionwhat activates my bodywhat I rejectwhat I return towhat changes across time
Then the system gives those differences back in visible form.
The goal is not prediction alone.
The goal is discernment.
To leave a session not merely with a generated object, but with a finer vocabulary of one’s own perception.
This changes the role of the algorithm
In a conventional recommendation system, the goal is usually:
predict what the user will like.
In SYNTHESIS, the more interesting ambition may be:
help the participant notice how liking happens.
This is a fundamentally different relationship with technology.
Instead of hiding complexity behind a recommendation, the system can expose complexity.
Instead of saying:
“We know your preference.”
it says:
“Here are the different traces your preference left while it was forming.”
The algorithm is therefore not an authority.
It is not even primarily a designer.
It is a perceptual mirror.
And the quality of that mirror is measured not only by whether it predicts the final choice, but by whether it helps reveal distinctions the person could not previously see.
Where Matilda and SYNTHESIS meet
Matilda and SYNTHESIS now appear less like two separate experiments.
They are two methods of creating a useful distance from ourselves.
Matilda makes an inner voice temporarily Other so that we can hear it.
SYNTHESIS makes patterns of attention and bodily response temporarily visible so that we can examine them.
One externalizes intuition into dialogue.
The other externalizes attention into geometry.
Both create a loop:
inner process → external form → renewed perception → deeper inner process
And both depend on the same humility.
The mirror must not be mistaken for the self.
Its value lies precisely in allowing us to see something from another angle.
Perhaps preference is relational
This may be the deepest shift.
We usually imagine preference as a property stored inside a person:
I like X.
But what if liking is less like a stored value and more like an event?
A relationship between:
a particular person,a particular form,a particular body state,a particular context,a particular history,and a particular moment of attention.
Then preference is not merely retrieved.
It is formed.
And if it is formed, it can be observed as a process.
Not to control it.
Not to reduce it to data.
But to understand its texture.
The new question
So the question behind SYNTHESIS changes.
It is no longer only:
Can attention generate form?
Nor even:
Can biometrics predict aesthetic preference?
The deeper question becomes:
How does liking come into being?
What happens before the conscious “yes”?
What happens when the eyes say yes and the voice says no?
What happens when the body responds to something the mind rejects?
What happens when all three align?
And can repeatedly seeing these relationships increase our ability to notice finer distinctions in our own experience?
If so, SYNTHESIS may produce something more interesting than an optimized object.
It may produce a more precise observer.
And perhaps that is the real artifact.
The final form can still be printed.
But the deeper materialization happens elsewhere:
the person leaves with a higher-resolution way of seeing.
SYNTHESIS creates conditions in which a human being can observe beauty becoming a choice.