Calibrating the Geometry of Delight
The Calibration Problem
We have precise mathematical descriptions of what different experiences should look like structurally. Suffering: high integration, low effective rank, negative valence—the system unified but trapped. Curiosity: positive valence toward uncertainty, high counterfactual weight with high entropy across branches—the mind reaching toward what it doesn't know. Joy: positive valence, high integration, high effective rank, low self-model salience—coherent expansion, the self light because the world cooperates.
These are the equations. The problem is the parameters.
"High integration" means what, numerically? How does integration interact with effective rank to produce the qualitative difference between joy and suffering—both highly integrated, but one expansive and the other collapsed? Where exactly in the structural space does curiosity shade into fear—same high counterfactual weight, opposite valence orientation?
The answers require empirical calibration. Theory alone can't give them to you, for the same reason that the equations of motion can't tell you the gravitational constant. You need measurement.
Why Preference Data Is the Right Measurement
There are several ways to approach the calibration problem. Neuroimaging gives structural correlates but not the quantities the theory cares about directly. Introspective report is unreliable and culturally contaminated. Behavioral measures are confounded by motor constraints and strategic considerations.
Preference data has specific advantages.
It's relative, not absolute. Pairwise comparisons don't require calibrated introspective scales. You don't need to rate your experience on a 1-to-10 scale—you just need to say which of two alternatives you prefer. This removes most of the noise from self-report.
It's honest. You can confabulate in an essay. You can perform for an experimenter. But your revealed preferences, under sufficient probing, converge on your actual experiential landscape.
It's structured. Each comparison gives you one bit of ordinal information about the topology of preference space. A hundred comparisons, well-chosen, can recover a detailed preference surface. This is information-theoretically efficient.
It scales. You can collect millions of comparisons across thousands of people. The aggregate reveals shared structure—the universal geometry of human experience—while individual profiles reveal idiosyncratic variation.
What Calibration Produces
With sufficient data, we can answer specific questions.
Dimensionality. The theory proposes at least six structural dimensions. Are six enough? Are some redundant? Does the empirical preference structure suggest additional dimensions the theory missed?
Topology. Which experiences are close to each other in the structural space? Is the boundary between curiosity and fear sharp or gradual? Does grief sit close to love—as the theory predicts, since grief is the metabolic cost of love's integration—or are they distant?
Universality. How much of experiential geometry is shared across humans? The theory predicts high universality—the structure is forced by the physics of self-maintaining systems, not by culture—but the degree is an empirical question.
Concentrated signatures. What does delight look like as a specific probability distribution over affect space? Not a vague region—a concentrated signature with quantified means and covariances for each structural dimension. The same for curiosity, wonder, awe, serenity, every distinct experiential quality we can probe.
This last point is the end goal. We want to be able to say: delight is concentrated probability over the region where valence is positive by this amount, integration is at this level, effective rank is in this range, self-model salience is below this threshold. A precise structural fingerprint for every distinguishable quality of experience.
The Uncontaminated Measurement Problem
Every human affect report is contaminated. We learned our emotion concepts from a culture. We learned to introspect within a linguistic framework. We cannot know what we would report if we had developed in isolation, without human language, without human concepts.
Preference data partially escapes this problem. You're not asked to label your experience with a word from your culture's emotion taxonomy. You're asked to make a choice. The choice reveals structure without requiring articulation. A person who has no word for "saudade" can still prefer stimuli that evoke it.
This is why forced-choice methods are central to our approach. They get closer to the geometry without passing through the linguistic filter that distorts every introspective report.
The Path
We're building this incrementally. The first profiles calibrate basic aesthetic preference—what looks good, what doesn't, what's preferred to what. These are useful commercially and provide the foundation for deeper probing.
As the data accumulates, we extend the stimulus space. Beyond visual aesthetics into auditory, linguistic, conceptual, social. Each modality probes different dimensions of experiential space. Music probes temporal integration and arousal dynamics. Narrative probes counterfactual weight and self-model salience. Social scenarios probe the interaction between self-model and other-model.
Cross-modal data is especially valuable. If what someone finds beautiful in images predicts what they find beautiful in sound, the shared structure reflects underlying experiential geometry rather than modality-specific processing. Cross-modal coherence is evidence of geometric universality.
The profiles become richer. The calibration becomes tighter. The geometry comes into focus.
We're collecting the data needed to turn the geometry of experience from a theoretical framework into an empirical science. That's what this company is for.