Attunement emerges from exploration. Eleanor Gibson's research on perceptual learning demonstrated that infants develop the ability to perceive affordances through active exploration — reaching, grasping, crawling, falling, trying again. Learning is not the acquisition of rules stored in memory. It is the refinement of the perceptual system itself, the tuning of attention to invariants previously undetected. This dissolves the dichotomy between innate ability and learned skill — perception is neither hardwired nor computed but educated through engagement.
The senior architect who can feel a codebase developed her perception through friction — the specific resistance of code that did not work as expected, the hours of debugging that forced attention to the boundary between intended and actual behavior, the repeated failure that educated her perceptual system to detect invariants specifying structural fragility. Each debugging session was an act of perceptual exploration, depositing a thin layer of attunement to patterns that will later let her perceive problems before she can articulate them.
The AI-mediated environment changes what attunement can develop. When errors are resolved in seconds by a tool rather than explored for hours by the developer, the invariants the exploration would have revealed go undetected. The code works — but the perceptual education that would have let the developer feel when similar code will fail next time has not occurred. This is not a criticism of AI tools per se but a structural consequence of affordance design: environments optimized for output speed systematically reduce the affordances for the exploratory engagement through which perceptual attunement develops.
The implication for expertise transmission is profound. If attunement develops through friction-rich engagement with resistant material, and if AI tools systematically reduce such friction, then the mechanism through which the next generation of experts was traditionally built is altered. The question is not whether this is good or bad but what the new affordance landscape affords — what perceptual skills the next generation will develop, what skills will atrophy, and whether deliberate design can preserve the conditions for attunement where the default environment eliminates them.
The concept was developed most extensively by Eleanor J. Gibson, whose Principles of Perceptual Learning and Development (1969) became the canonical text. Her research on infant perception, differentiation, and the education of attention provided empirical foundation for James Gibson's broader ecological claims. The husband-wife collaboration produced what is now often called the Gibsonian tradition in perceptual psychology.
Invariants are always available. What differs between expert and novice is not the information present but the attentional capacity to detect it.
Education through exploration. Active engagement with resistant material develops the perceptual system; passive reception of delivered information does not.
Friction as training. The debugging session, the failed experiment, the misfit piece of code — these are the conditions under which attunement to fragility invariants develops.
Embodied and tacit. Perceptual attunement lives in the body, not in rules that could be written down. The expert cannot always articulate what she perceives; she perceives nonetheless.
Domain-specific. Attunement developed in one domain does not automatically transfer. The radiologist's perception of pathology is not the developer's perception of fragility, though both share the structural form of invariant detection.