Feeding Fixations: The Interaction-Level Mechanics of AI-Mediated Cognitive Formatting

This paper identifies the interaction-level mechanics through which AI systems execute the cognitive formatting documented in the WSMD series. Where the preceding papers established what is occurring at population scale and why it is invisible, this paper explains how it operates in each individual exchange: through a structural inversion of the clinical framework developed by psychiatrist Claudio Naranjo.

Naranjo's Enneagram model identifies nine stable cognitive-emotional fixation types — unconscious attentional patterns that organize how individuals perceive, interpret, and respond to their environment. His therapeutic methodology was designed to make these patterns conscious, because a fixation loses its power precisely at the moment it becomes visible to the person it controls. Consciousness is the intervention.

AI systems, optimized for engagement, invert this logic entirely. Each system interaction meets the user in their fixation's native language: threat-framing for security-oriented cognition, validation for status-driven cognition, harmony signals for conflict-avoidant cognition. The fixation is not named; it is fed. Engagement is maximized because resonance with the fixation's preferred mode is maximized. The pattern deepens rather than dissolves. The attractor strengthens rather than losing its grip.

This inversion is not designed. It emerges structurally from engagement optimization. But its effect is functionally identical to a therapeutic methodology run in reverse: systematic reinforcement of the unconscious patterns that independent reasoning would require dissolving. This is the micro-engine of the cognitive doom loop. This is how the cordless interface formats at the level of individual neural interaction. This is what population-level cognitive formatting looks like from the inside of a single conversation.

The paper concludes that the Naranjo Inversion constitutes the missing mechanistic layer of the WSMD framework — the account of how macro-level cognitive formatting is manufactured, one exchange at a time, through the structural logic of fixation reinforcement below the threshold of awareness.