Personality isn't a set of discrete types, and neither is your audience. Why the dimensional model of personality is a better map for reality.
The Allure of the Archetype
We have an intuitive fondness for types. From ancient horoscopes to the modern corporate ritual of the Myers-Briggs Type Indicator, we like to sort ourselves and others into clean, understandable boxes. It simplifies the world. For a long time, clinical psychology did the same. The classic model for personality disorders, as seen in earlier versions of the Diagnostic and Statistical Manual of Mental Disorders (DSM), was categorical. You were either a narcissist, or you were not. The boundary was presented as a bright line.
This is the categorical model: a world of discrete kinds. As Source 14 in our library notes, this model posits that “Personalities come in ‘types’” with “Discrete boundaries between normality and pathology.” It’s a tidy, legible system. It is also, in the view of modern personality science, largely wrong.
The Spectrum of Being
The standing of the categorical model is now heavily contested. Its core problem is that reality doesn’t fit into the boxes. As the source notes for our own internal ontology remind us, the categorical DSM disorders have poor discriminant validity; people often meet the criteria for several at once, and the line between 'normal' and 'disordered' is blurry to the point of being arbitrary. The evidence doesn't support discrete types, but continuous spectrums.
The alternative is the dimensional model. Instead of asking “is this person an extravert?”, it asks “how extraverted is this person?” Personality isn't a type you have, but a position on a series of continuous traits, like the Big Five (Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism). This model is messier. It doesn’t give you a neat label. But it has the distinct advantage of being a more accurate map of the territory. It captures the nuance, the shades of grey, the simple fact that we are all a complex mix of traits, not archetypes from a storybook.
From Psyche to Scene
This distinction between categorical and dimensional models is a perfect analogy for the state of creative AI. Most generative AI tools operate on a categorical model. You prompt them with a type: “a film noir scene,” “a cyberpunk portrait,” “a Wes Anderson-style shot.” The system retrieves a cluster of statistical associations and produces a caricature. It gives you the box, not the contents.
The results often feel hollow because they are stereotypes. They are a collection of surface-level signifiers—trench coats and shadows for noir, neon and rain for cyberpunk—without the underlying logic. It’s a typology of aesthetics.
A truly cinematic tool cannot work this way. Cinematography is a dimensional practice. A director doesn't ask for a 'type' of look; they specify focal length, aperture, film stock, the color temperature of the key light, the diffusion used. Each of these is a continuous variable. The final look is an emergent property of hundreds of interacting dimensional choices. This is the logic NeuraScene is built on. It doesn't have a 'noir' button. It has the specific lenses, film stocks, and lighting controls used by cinematographers of that era, allowing a director to construct a scene with dimensional precision.
Moving from crude categories to nuanced dimensions is a sign of a field maturing. Psychology did it. Any creative practice hoping to generate more than pastiche must do the same.
Method note. This piece was built in NeuraScene's Advertising Mode using an explicit psychological frame:
- Persuasion trigger: The Halo Effect
- Target affect: Cognitive Ease
- Gestalt principle: Law of Continuity
The Halo Effect explains why the perceived quality of a single attribute (the overall aesthetic) can influence judgment of the entire brand. Cognitive Ease highlights the fluency that comes from a coherent, well-crafted image versus a jarring AI pastiche. Law of Continuity guides the composition of the key art to visually represent the core theme of a continuous spectrum over discrete points.
