Ascending Friction Thesis

In the AI Story

The distinction between cognitive growth and cognitive drift is the thesis's practically critical feature. Growth occurs when freed resources are invested in progressively more demanding operations, producing neuroplastic strengthening of the circuits those operations engage. Drift occurs when freed resources are not invested — when the workflow beyond the automated tasks is routine and cognitively undemanding — and the prefrontal cortex, receiving no recruiting demand, reduces its overall activity into a mild sustained hypofrontality that feels fluent but does not involve deep engagement.

The reallocation is demand-driven, not intention-driven. An engineer who intends to use her freed time for architectural thinking but whose workflow does not present architectural challenges will experience drift regardless of her intentions. The prefrontal cortex recruits resources in response to demands it encounters, not demands the individual believes she should encounter. This transforms the ascending friction thesis from an optimistic assertion about AI's future into a design specification: the cognitive environment must present ascending challenges at the moment lower-order friction is removed.

The neural specificity of ascending operations is instructive. Lower-order friction engages primarily the dorsolateral prefrontal cortex in monitoring mode — working memory maintenance, rule application, narrow pattern-matching within single domains. Higher-order challenges engage a broader prefrontal coalition: dorsolateral contributes working memory and flexibility, orbitofrontal contributes reward and value evaluation, ventrolateral contributes controlled semantic retrieval across domains, medial prefrontal contributes self-referential integration with long-term goals. The transition from lower-order to higher-order is a qualitative shift in activation pattern — from narrow single-subregion monitoring to broad multi-subregion evaluation — and the multi-subregion engagement exercises circuits that lower-order work left dormant.

Organizations deploying AI tools and measuring results exclusively through productivity metrics cannot distinguish ascending friction from lateral expansion — more work at the same cognitive level. Both produce measurable output. Both consume freed resources. The measurement required to detect genuine ascent is itself a higher-order cognitive operation: assessing the cognitive complexity of the work produced, the number of integrated domains, the depth of evaluative reasoning brought to strategic choices. This is a self-referential problem: the capacity to assess ascending friction is one of the capacities that ascending friction develops.

Key Ideas

Metabolic, not metaphorical. The freed resources are a real finite budget that gets reallocated or dissipates.

Growth is conditional. Ascent requires environmental demand; without it, freed resources produce drift, not growth.

Different prefrontal coalitions. Lower-order engages one subregion narrowly; higher-order engages a broader coalition evaluatively.

Standard metrics cannot distinguish ascent from expansion. Output volume looks identical whether freed resources ascend or dissipate laterally.

Manager as environment designer. In AI-augmented organizations, the manager's primary function becomes designing demands that recruit freed resources to higher-order engagement.

Debates & Critiques

Critics have questioned whether the thesis's conditional structure — ascent-if-the-environment-presents-challenges — rescues it from falsifiability by placing all failures on the environment rather than on the thesis itself. Defenders respond that the conditionality is the thesis's empirical content: it predicts specific patterns (ascent in challenge-rich environments, drift in challenge-poor environments) and the predictions can be tested.