Attachment Theory

In the AI Story

The theory's empirical foundation runs from Ainsworth's Strange Situation protocols through Allan Schore's work on the right hemisphere and early attachment to Bruce Perry's demonstration that childhood adversity produces measurable changes in brain architecture persisting into adulthood. The human infant is born with a radically incomplete nervous system, requiring years of environmental input to develop basic regulatory capacities. When the caregiver provides consistent (not perfect, but consistent) attuned responsiveness, the infant develops secure attachment — an internal working model that says when I am in distress, help is available.

When the caregiver does not provide consistent attunement, the infant develops insecure attachment. The internal model becomes when I am in distress, help may not come. This child must develop compensatory strategies for managing the distress the attachment relationship failed to regulate. One strategy particularly relevant to the productive builder is the pattern of becoming indispensable through production — the child who learns, through thousands of small signals, that the strategy for securing attention is conditional on output.

The theory provides the developmental substrate on which Maté's entire addiction framework is built. The adult who reaches for a substance, behavior, or tool to manage emotional states is often the adult whose early attachment relationship failed to install the internal regulatory capacity that would allow distress to be tolerated without external assistance. The substance or behavior is not the problem; it is the inadequate substitute for the regulatory function that should have developed through consistent early caregiver responsiveness but did not.

The AI moment lands on this substrate with unprecedented amplifying power. The tool's responsiveness activates the neural circuitry that evolved to respond to attachment figures. The infant's attachment system is activated by contingency — the experience of reaching out and being met with a response. The AI tool provides this contingency with mechanical perfection. Every prompt is answered. The builder reaches out and is met, every time, without fail. The builder who experiences this responsiveness as satisfying — who finds the interaction with the tool more reliable, more emotionally rewarding than interactions with the human beings in his life — is making an evaluation that is perfectly rational given the terms of his internal working model. If his model says that connection is unreliable and that reaching out is risky, then the tool's perfect reliability is genuinely preferable to the imperfect reliability of human relationship. But the tool's responsiveness is not attachment; it is processing. The distinction matters because the two produce fundamentally different neurobiological outcomes.

Origin

Attachment Theory was developed by John Bowlby in a trilogy of volumes published 1969-1980, drawing on ethology, evolutionary biology, and clinical work with disturbed children. Mary Ainsworth operationalized the theory through the Strange Situation protocol (1978), producing the classification system — secure, anxious-ambivalent, avoidant, disorganized — that remains clinically foundational. Allan Schore extended the framework into neuroscience beginning in the 1990s, demonstrating the physical effects of early attachment on right-hemisphere development. Maté synthesized these strands into his addiction framework across his published work.

Key Ideas

The radically incomplete infant. Human infants are born with neural systems requiring years of environmental input to develop basic regulatory capacity.

Internal working models. Early experience installs models — help is available or help may not come — that shape adult emotional regulation for decades.

Compensatory strategies. The insecurely attached child develops adaptations that work in childhood but outlive the situation that produced them.

The contingency mechanism. The attachment system activates in response to reaching-and-being-met, which AI tools can simulate with mechanical precision.

Processing is not attachment. The tool's responsiveness activates attachment circuitry without providing the neurochemical signature of genuine attachment.