The Body Under Continuous Partial Attention
The body knows before the mind does. Stone's framework insists that the cost of continuous partial attention is borne first and most accurately in physiological systems: the respiratory rhythm disrupted by screen apnea, the sympathetic activation maintained for hours past its evolutionary design window, the cortisol accumulating without the physical action that would metabolize it, the prefrontal cortex degraded by the very chronic stress its restoration would require. The mind constructs narratives of productivity and competence. The body responds without rhetorical interest in the narrative. And the response — measurable in breathing patterns, heart rate variability, muscle tension, and cortisol profiles — constitutes a report on attentional quality the mind's narrative cannot override.
The Maintenance Economy of Vigilance — Contrarian ^ Opus
There is a parallel reading that begins from the political economy of attention management. The physiological costs Stone documents are not accidental byproducts but necessary inputs to a specific mode of value extraction. The sustained cortisol, the degraded executive function, the chronic sympathetic activation — these are not bugs but features of a system that requires workers to maintain availability for intermittent, high-stakes decision points while managing the cognitive load of multiple AI systems. The body under continuous partial attention is performing exactly the labor the current configuration of AI-augmented work requires: a biological substrate maintaining readiness for unpredictable intervention points.
The recovery mechanisms the entry prescribes — genuine disengagement, natural settings, parasympathetic restoration — assume a worker with sufficient autonomy to structure their own time and sufficient economic security to prioritize physiological health over immediate productivity demands. But the gig worker monitoring multiple delivery algorithms, the content moderator scanning AI-flagged posts, the customer service representative managing chatbot escalations — these workers experience continuous partial attention not as a choice but as a job requirement. Their bodies bear the cost of maintaining the human oversight layer that makes AI systems commercially viable. The degradation of their prefrontal cortex is not an ironic side effect but a predictable outcome of a system that externalizes the physiological costs of maintaining AI's appearance of seamless operation. The body knows what the mind cannot afford to acknowledge: that someone must pay the biological price for keeping the machines supervised, and that price is systematically extracted from those with the least power to refuse the terms.
In the AI Story
The mechanism is well documented in stress physiology, though its connection to attentional states is less widely appreciated. When the body enters a state of vigilance, the sympathetic nervous system engages — the autonomic branch that prepares the organism for action. Heart rate elevates. Blood vessels constrict, redirecting flow to large muscle groups. Cortisol releases from the adrenal glands, mobilizing glucose. Breathing shallows in preparation for quick exertion. This activation is adaptive when the threat is real and brief: the organism mobilizes, responds, and recovers. The cortisol metabolizes through physical action. The sympathetic engagement gives way to parasympathetic recovery. The body returns to baseline. The sequence is designed by evolutionary logic to last minutes.
The person under continuous partial attention does not complete the sequence. The vigilance is sustained but the physical action that would resolve it never arrives. The body prepares for a response that does not come. The cortisol accumulates without exertion. The sympathetic engagement persists without parasympathetic recovery. The body maintains a state of low-grade readiness for hours — a state designed to last minutes. The chronically elevated cortisol impairs immune function, disrupts sleep architecture, and degrades the function of the prefrontal cortex — precisely the brain region most essential for the executive functions AI-augmented work demands.
The irony is precise. The builder who monitors the AI's output for eight hours is less capable of the judgment the monitoring requires than she would have been after a morning of sustained, single-channel engagement. She is scanning more and judging less well. Monitoring more and understanding less. The body is undermining the mind's capacity for the very work the body is being asked to support. In The Orange Pill, the bodies of the builders tell this story with consistency: engineers leaning toward screens, jaw tension, the grinding momentum of a person who has confused productivity with aliveness.
Recovery requires what attention-restoration research calls genuine disengagement — not the half-disengagement of checking the phone in the park but full withdrawal of attention from monitored channels for a duration sufficient to allow parasympathetic engagement, breathing to deepen, cortisol to metabolize. Rachel and Stephen Kaplan's research identifies natural settings as reliably restorative because they produce all four conditions of attention restoration simultaneously: being away, extent, fascination, and compatibility. The AI-augmented ecology makes such genuine disengagement structurally difficult because the tool is always available, always valuable, and the gap between impulse and prompt has shrunk to the width of a text message.
Origin
Stone documented these physiological signatures over years of observation in technology workplace settings, beginning informally in 1990s Microsoft corridors and later through systematic data collection on screen apnea. The connection to broader stress physiology research — particularly Hans Selye's general adaptation syndrome framework — provides the empirical foundation for the cumulative-cost claims.
The cortisol research establishing the link between chronic vigilance and prefrontal degradation has been developed across decades by researchers including Bruce McEwen, Robert Sapolsky, and the broader stress neuroscience community. Stone's distinctive contribution is the application of this framework to attentional states produced specifically by digital monitoring environments.
Key Ideas
Sympathetic activation without resolution. The vigilance posture mobilizes the stress response without the physical action evolution designed to metabolize it.
Cortisol degrades the executive brain. The chronic stress of sustained scanning impairs the prefrontal cortex — the region most essential for the judgment AI-augmented work requires.
The dashboard cannot see the body. Productivity metrics capture output but not the physiological cost being paid in shallow breathing, accumulated cortisol, and degraded executive function.
Recovery requires genuine disengagement. Half-disengagement (the phone in the park) does not allow parasympathetic recovery; only full withdrawal from monitored channels permits restoration.
The cycle is self-reinforcing. Shallow breathing reduces oxygen, which degrades cognition, which increases the felt need for more scanning, which sustains the shallow breathing — a feedback loop that intensifies until structural intervention breaks it.
Appears in the Orange Pill Cycle
The Distribution of Attentional Burden — Arbitrator ^ Opus
The physiological mechanisms Stone documents are empirically robust — the cortisol cascades, the sympathetic lockup, the prefrontal degradation all occur as described (100% to Stone). Where the perspectives diverge is not on what happens to bodies under continuous partial attention but on why these conditions persist and who bears the greatest burden. For knowledge workers with autonomy, Stone's framework of recovery through deliberate disengagement holds — they can structure restoration into their days (70% to Stone). For platform workers and others in the attention-maintenance economy, the contrarian reading dominates — physiological degradation is not a bug but the substrate on which their labor value rests (80% to contrarian).
The question of intentionality reveals another split. Stone's account treats these costs as unintended consequences of a poorly designed attentional ecology — problems to be solved through better practices and tool design (60% to Stone for describing individual experience). The political economy reading sees these costs as structurally necessary for the current configuration of AI-augmented work — problems that cannot be solved without reorganizing the value system itself (75% to contrarian for systemic accuracy). Both are simultaneously true: individuals experience these states as dysfunctional while systems depend on their continuation.
The synthesis requires holding both the biological reality and its economic function. The body under continuous partial attention is simultaneously a physiological system responding to unprecedented cognitive load (Stone's empirical contribution) and a laboring substrate maintaining the attention economy's infrastructure (the contrarian's structural insight). The path forward cannot be purely individual (restoration practices) or purely systemic (reorganizing labor relations) but must address both the immediate physiological needs of bodies under strain and the economic structures that make such strain profitable. The body knows; the question is whether the system can afford to listen.
Further reading
- Hans Selye, The Stress of Life (McGraw-Hill, 1956)
- Robert Sapolsky, Why Zebras Don't Get Ulcers (Holt, 2004)
- Bruce McEwen, The End of Stress as We Know It (Joseph Henry Press, 2002)
- Rachel Kaplan and Stephen Kaplan, The Experience of Nature (Cambridge, 1989)
- Linda Stone, essays at lindastone.net