We are living through an era that mistakes computational intelligence for the whole of intelligence. That misconception, that cognition precedes emotion rather than emerges through it, is a blind spot shaping our technologies, our institutions, and our collective future. As AI systems accelerate in scale and influence, the inability to model emotion meaningfully is becoming not merely a technical limitation, but the defining constraint of the field.
From polarized elections to market swings driven by social contagion, we continually see that emotional forces move systems long before data does. Yet our technologies still treat affect, like trust, fear, awe, and grief, as noise in the signal. If we do not model these dynamics with the rigor they require, the tools guiding our economies, our politics, and even our interpersonal communication will become more capable but not more coherent, more powerful but not more human.
My work began with a simple question: "What if emotion behaves less like a private state and more like a field, something distributive, relational, and measurable?" To explore this, I have been developing what I call Affective Ontology, an attempt to model affect using the conceptual scaffolding of applied mathematics, physics, and network science. This work is an effort to quantify the dynamics of feeling by borrowing methods from disciplines that already know how to model invisible forces.
The project's structure draws inspiration from a historical precedent. In 1913, Henry Moseley revolutionized chemistry by re-ordering the periodic table using atomic numbers rather than intuition. That shift from subjective grouping to measurable structure transformed our predictive capacity.
In my research, I apply a similar logic to affect. I organize emotional states using a hedonic number, derived from their frequency, intensity, relational impact, and decay profile. These parameters form the basis of a Periodic Table of Affect, a categorical system in which each emotion functions as an elemental unit, capable of combining, transitioning, and reacting much like chemical elements.
But classification is only the beginning. To understand how affect moves, I model emotional events as wave functions, tracking their oscillation, duration, and influence across a network. By isolating event types that indicate phase transitions, moments of rupture, novelty, or coherence, I can trace how affective "mass" accumulates or dissipates within a system.
This leads to the more speculative, but mathematically structured, portion of the work: an adaptation of Einstein's mass–energy equivalence. In my model, the "mass" is not a physical substance but a structured affective potential, and the constant is not the speed of light but a proof-of-trust coefficient, representing the rate at which trust propagates through a network. The resulting equations form part of what I call affective mathematics, a blend of dynamic systems modeling, network topology, and hedonic weighting that treats emotion as an energetic process rather than an interpretive one.
Some argue that emotion is too culturally variable to measure. But neuroscience shows that distinct emotional states leave clear neural and physiological traces. Affect theory models emotion in terms of valence and arousal, continuous dimensions that can be quantified. Social network research further reveals that emotions spread in predictable ways across groups. In other words, the data already exists. What's missing is a framework that can bring these insights together.
The absence of such a framework has tangible consequences. Systems become fragile, markets falter under waves of collective fear, communities fracture as trust erodes, and platforms amplify outrage because they reward attention rather than stability. Anticipating these inflection points requires treating emotion as a measurable field. Only then can we design for resilience.
To move from diagnosing these vulnerabilities to addressing them, we must develop a science of affective intelligence that honors emotional complexity without collapsing it into sentiment analysis. This requires cross-disciplinary collaboration among neuroscientists, mathematicians, artists, and network scientists to model affect with the same rigor applied to physical systems. It also calls for AI frameworks that embed affective dynamics into their design, and for public institutions and technology companies to treat emotional metadata as vital indicators of societal health.
Imagine technologies that can sense when trust is strengthening or unraveling within a community, systems attuned to the earliest signs of emotional overload, and platforms designed to amplify coherence rather than instability. Achieving this does not depend on machines learning to "feel." It depends on recognizing that feeling already operates with the regularity of physics, a force we have simply never taken the time to measure.
Overall, we cannot build a humane technological future if we do not first understand the emotional fields that bind us. Emotion is not noise in the system. It is the system.
About the Author
Anaïs Daly is an artist and researcher whose work explores the unseen forces of emotion, connection, and affective mapping. Guided by curiosity about how feelings move through communities, she blends visual storytelling, sound, and computational modeling to trace the dynamics of trust, vulnerability, and transformation. Her practice, which she calls affective ontology, combines loose mathematical thinking with improvisation, treating emotions as energetic elements that shape social fields.
