A Regulation-First Model for Neurodivergence, Stress, and Support
Updated: July 17, 2026
Note: The model and ESF offer a conceptual lens for understanding nervous-system variation. It is not medical advice, diagnosis, or a treatment protocol. Within the Evolutionary-Stress Framework (ESF), the High-Gain Nervous System model offers a regulation-first way to understand neurodivergent stress, sensory salience, recovery cost, and support needs.
A diagnosis can name a pattern. It can open doors, guide services, and help families, clinicians, and educators recognize that a person’s needs are real. But a diagnosis is not the same thing as a roadmap.
A regulation-first roadmap begins with a different question: What kind of nervous system is this person living in?
Some children and adults appear to move through the world with what we might call a high-gain nervous system.
In engineering, gain is amplification — but it lives in two different places, and the distinction matters. At the front end, gain is how strongly a system detects a signal: turn it up and you pick up weaker inputs, finer patterns, smaller changes. Inside the control loop, gain is how hard the system corrects once it detects a mismatch: turn it up and the system responds faster and tracks more tightly — but it also overshoots, rings, and takes longer to settle. High loop gain buys responsiveness at the cost of stability margin.
A high-gain nervous system is not a broken one. It is a high-information, high-responsivity regulatory system. It may detect more at the front end and correct harder in the loop — which is why the same architecture can both notice more and recover more slowly. It may protect faster, need more precise conditions to stay stable, and take longer to return to baseline after it is pushed.
One more thing this framework insists on: gain is not a single dial for the whole person. It is channel-specific. The same nervous system may run high gain on sound, light, or social threat and low gain on, say, proprioception or internal pain signals — hyper-responsive in some channels, hypo-responsive in others. “High-gain” is therefore not a synonym for any one diagnosis, and not a claim that such a person is “turned up” everywhere. It is a claim about a profile of gains across channels. That profile is what the rest of this framework tries to read.
That changes the question from “Why won’t this person act normal?” to “What is this nervous system detecting, correcting, avoiding, seeking, or recovering from — and in which channels?”
That shift is the beginning of the Evolutionary-Stress Framework.
High gain is not the same thing as “overreacting”
The language of “overreaction” assumes the environment is neutral and the response is excessive. A regulation-first view starts elsewhere: the environment is never just background. Light, sound, heat, hunger, pain, social expectation, uncertainty, transitions, clothing texture, classroom structure, sleep debt, and adult tone all set the regulatory load placed on the system.
For a high-gain nervous system, the same environment can carry a different cost — and a system correcting hard against a strongly detected mismatch is not overreacting. It is doing exactly what a high-gain regulator does. What looks like an outsized response from outside may be a system with a narrow stability margin meeting a load that would be trivial for a lower-gain one.
One person may walk into a classroom, grocery store, workplace, church service, clinic waiting room, or family gathering and process only the most relevant signals. Another may be tracking the lights, overlapping conversations, movement in peripheral vision, social rules, bodily sensations, temperature, smell, uncertainty, emotional tone, and the effort to suppress visible distress — all at once. From the outside, both are “in the same room.” Within the nervous system, they do not do the same amount of work.
A useful distinction: gain, throughput, and cost
To make this more precise, it helps to separate three related ideas.
Gain is how strongly a nervous system detects, weights, or corrects — at the sensory front end and within the regulatory loop.
Sensory or regulatory throughput is how much information continues to be processed rather than fading into the background.
Processing cost is the cognitive, metabolic, emotional, autonomic, and behavioral work required to compare, predict, inhibit, integrate, mask, respond, or recover.
The two senses of gain map cleanly onto these. Front-end gain feeds into throughput: the more strongly signals are detected and weighted, the more of them remain salient rather than settling into the background. Loop gain feeds cost and recovery: the harder the system corrects, the more work each correction takes, and the longer it takes to settle afterward. “Sensitivity” is often treated as the whole story, but the more useful question may be: how much information is this nervous system having to process, how hard is it correcting, and what does that cost?
In contemporary neuroscience, perception is not usually understood as passive reception of the world. The brain is constantly predicting, comparing, correcting, and regulating. Allostasis describes regulation as anticipatory: the body does not simply react after imbalance occurs; it prepares for expected needs in advance. Interoceptive models extend this idea by describing bodily feeling as partly shaped by brain-based predictions about the body’s internal state, constrained by incoming visceral signals.
This means a high-gain system may not simply “notice more.” It may assign greater salience or precision to more signals, update predictions more often, and spend more capacity determining what matters. Meta-analytic work also indicates that sensory modulation differences are common in autism, consistent with the idea that front-end weighting can differ substantially across individuals (Ben-Sasson et al., 2009).
Predictive-processing accounts of autism have made related arguments. Pellicano and Burr proposed that autistic perception may involve reduced reliance on prior expectations, making the world feel more immediate or “too real.” Lawson, Rees, and Friston described autism in terms of altered precision weighting, where prediction errors may be experienced as more salient or harder to attenuate. Van de Cruys and colleagues similarly argued that autistic perception may involve unusually precise prediction errors in uncertain contexts.
These models are not identical to the High-Gain Nervous System framing. But they support the broader point: differences in perception, prediction, attention, and salience can change the cost of being in the world.
Processing is not free
Information processing has biological cost. The brain is metabolically expensive, and neural signaling consumes energy. Attwell and Laughlin’s classic energy-budget analysis showed that action potentials and postsynaptic glutamatergic activity account for major components of gray-matter signaling energy, and Harris, Jolivet, and Attwell later emphasized that much of the brain’s energy use is tied to synaptic transmission — making energy supply and information transmission tightly connected.
This does not mean we can calculate a person’s “bandwidth” by counting sensory inputs. The nervous system is not a simple battery. But it does mean the metaphor of processing cost is not merely psychological. Attention, working memory, inhibition, prediction, emotional regulation, social monitoring, masking, and recovery all require biological coordination. When more signal remains salient and the system corrects harder, more regulatory work is required.
A child who melts down after school may not have “held it together and then chosen to fall apart.” They may have spent the day suppressing sensory distress, monitoring social cues, tolerating uncertainty, managing transitions, tracking body needs, and performing expected behavior until the system no longer had capacity. An adult who seems “fine” in a meeting but collapses afterward may not be fragile. They may be paying a delayed regulatory cost.
High gain can be adaptive
High gain should not be framed only as vulnerability. A high-gain nervous system may support careful perception, pattern recognition, sensory richness, emotional depth, precision, creativity, justice-sensitivity, rapid mismatch detection, and intense focus. Many neurodivergent strengths may emerge from the same architecture that creates support needs.
This is why the goal is not to turn the system down until it looks typical. The goal is to understand the conditions under which the system becomes adaptive rather than overloaded.
Monotropism offers one useful bridge. Murray, Lesser, and Lawson described autistic attention as involving intense allocation of attentional resources along narrower channels of interest. This moves the conversation away from “poor attention” toward a more precise question: how is attention being allocated, protected, interrupted, or overloaded?
Double-empathy research extends this point in a way central to this framework: regulation and understanding are not located solely within the individual. Crompton and colleagues found that autistic peer-to-peer information transfer can be highly effective, and related work suggests that rapport is shaped by neurotype matching rather than by autism alone. If communication breakdown is a property of the mismatch between two systems rather than a deficit inside one of them, then the same logic applies to regulation: what a nervous system can sustain depends on the coupling between brain, body, environment, relationships, expectations, and available supports — not on the person in isolation.
When high gain becomes high load
High gain becomes costly when the environment repeatedly exceeds regulatory capacity. The chain may look like this:
High gain + high noise + low predictability + pressure to mask
→ increased sensory, interoceptive, social, or emotional throughput
→ harder and more frequent regulatory correction, with a narrower stability margin
→ higher processing and recovery cost
→ less available regulatory capacity
→ fatigue, shutdown, meltdown, pain, anxiety, avoidance, or burnout
→ increased allostatic burden when sustained over time
(We use “capacity” and “bandwidth” informally here; the construct is developed formally elsewhere as regulatory bandwidth, a bounded quantity rather than a limitless reserve.)
Allostatic load refers to the cumulative biological cost of adaptation. McEwen and Stellar introduced the concept to describe the hidden cost of chronic stress and repeated physiological adjustment. McEwen later emphasized that stress mediators can be protective in the short run but damaging when repeatedly or chronically activated.
This is a crucial distinction. A single busy room does not equal allostatic load. A single hard transition does not equal trauma. A single sensory demand does not cause burnout. But repeated, unsupported, poorly understood regulatory demand can accumulate. Over time, the system becomes more reactive, less flexible, slower to recover — and, past a point, more likely to be pushed into a different regulatory state altogether.
Autistic burnout research makes this visible. Raymaker and colleagues defined autistic burnout as a syndrome associated with chronic life stress, mismatch between expectations and abilities, and inadequate support, with autistic adults describing chronic exhaustion, loss of skills, and reduced tolerance to stimulus. That description is highly consistent with a regulation-first model: burnout is not weakness. But “running out of capacity” isn’t quite the right picture either — and the difference matters.
Burnout is not breakage — the system relocates
It is tempting to picture burnout as falling off a cliff: fine one day, broken the next. That picture is wrong, and the error matters, because it turns a working nervous system into a damaged one.
Here is the reframe. Under sustained danger signaling — inflammation, unrelenting mismatch, a world the system keeps reading as unsafe — a high-gain nervous system does not run out of fuel and break. It moves. It shifts toward a new configuration that is more stable given what the body currently believes about the world. Every one of those shifts is the correct response to the signal it is receiving. A system that detected extreme danger and did not move would be the broken one.
What varies is the step size — and the step size tracks the urgency of the signal, not the amount of damage. When the danger signal is extreme, the system moves a long way, fast: many steps at once. From the outside, and from the inside, that looks like sudden collapse. When the signal is a low chronic hum, the system moves one step at a time, and you barely notice you are moving at all. Same process. Different pace. This is what the previous section was pointing at: the accumulating load of an overtaxed system is not filling a tank toward a breaking point — it is the pressure that keeps nudging the system to move.
And it is not a slide in one direction. The system walks a landscape — five steps down and two across, one step up and three back — searching, continuously, every day, for a configuration it can hold. This is emergent allostasis doing exactly its job: not a controller running low, but a system hunting for stability against the signal it is being given. Under sustained load, that daily walk carries a net drift. “Burnout,” in these terms, is a large, fast, mostly-correct relocation to a new stable state — arrived at because the system was told, over and over, that the old configuration was no longer safe to hold.
This reframe explains something the cliff picture cannot: the way back is not the way in. If burnout were simple depletion, restoring the old conditions would restore the person. It usually does not — because the system did not empty, it relocated, based on what it came to believe. Returning the environment to how it was is not the same as convincing the system the danger is over. The path out is longer than the path in, and that is a feature of how the system moves, not a failure of willpower.
PTSD is the clearest case. It is not a broken regulator. It is a regulator holding a configuration that was correct for the danger it was given — and that never received the update that the danger has passed. The body keeps signaling you are still in it, so the system keeps holding, and keeps shifting, hunting for stability against a signal that will not resolve. Nothing in the machinery is broken. The reprogramming simply never arrived.
Which is why recovery is re-learning, not refilling. You cannot rest a nervous system back to its old configuration, because the thing holding it in place is not fatigue — it is a belief: demand is coming, the environment is not safe. Beliefs like that do not move because you slept. They move because predictions of threat are made and disconfirmed, repeatedly, over time. This is why the things we call “good for us” work when they work: lowering inflammation quiets a threat signal at its source, and a predictable, low-demand environment supplies the one thing that actually updates the system — evidence, delivered reliably enough that its own predictions finally start coming true. The active ingredient is not safety. It is predictability. You cannot tell a nervous system it is safe. You can only let it repeatedly find out.
And this dissolves the language of “good” and “bad” states. There is no broken configuration and no correct one — only configurations, each appropriate to the signal that produced it. The work of recovery is not repair. It is giving the system enough steady, disconfirming evidence to relocate, one step at a time, back toward fit.
Stress changes capacity
One reason behavior-based interpretations fail is that they treat capacity as fixed. Capacity is not fixed. Sleep, pain, hunger, illness, inflammation, hormones, trauma history, sensory load, social threat, uncertainty, and recovery time all change what the nervous system can do. A person may be able to speak, transition, tolerate noise, make decisions, or cooperate on one day and not another. That is not inconsistency in character. It is state-dependent capacity.
Stress physiology helps explain why. The prefrontal cortex supports working memory, flexible thinking, inhibition, planning, and top-down regulation. Arnsten’s review of stress signaling pathways shows that stress can impair prefrontal cortical structure and function, weakening the very systems needed for executive control. This is why “try harder” often fails: when regulatory load is already high, adding demand may reduce the capacities required to meet it. A regulation-first approach therefore asks: What has to become easier before this person can do what we are asking?
A safety example: elopement, water, and regulation
The high-gain lens becomes especially important in safety planning. Autistic elopement and drowning risk are real and urgent. Anderson and colleagues reported that elopement was common in a large sample of autistic children, with drowning and traffic injury among the major dangers when children went missing. CDC guidance similarly identifies wandering as a major safety issue for autistic children and youth. A 2017 analysis of unintentional drowning deaths in autistic children found that many incidents occurred after wandering and close to home.
A regulation-first framework does not minimize the danger. It strengthens prevention. A child who runs toward water may be moving toward something the nervous system experiences as regulating: cooling, pressure, rhythm, buoyancy, sound modulation, visual pattern, escape from demand, or a whole-body state shift. In the channel-specific terms above, water-seeking need not be an anomaly bolted onto a “sensitive” system — it can be a seeking channel pulling hard toward regulation, in the same nervous system that runs high gain elsewhere. (Water is also physiologically active in its own right; cold-water facial immersion can trigger reflex bradycardia and autonomic change.)
Safety intervention and mechanistic understanding belong together. Swimming instruction, door alarms, pool fencing, visual safety plans, GPS or ID supports, school elopement plans, first-responder awareness, and community preparedness can save lives. A regulation-first formulation adds another layer: identify what the child is moving away from, what they are moving toward, and how to meet the regulatory need safely before danger escalates.
What changes when we use a regulation-first roadmap?
A regulation-first roadmap does not replace diagnosis, therapy, education, medical care, or safety planning. It changes how we interpret the person’s needs.
For families, it means tracking stress before behavior: sleep, pain, hunger, sensory load, transitions, masking, uncertainty, heat, illness, and recovery.
For clinicians, it means formulating beyond symptoms: migraine, gastrointestinal distress, autonomic symptoms, sensory load, sleep disruption, pain, burnout, and environmental mismatch should not be dismissed as “just autism” or “just anxiety.”
For educators, it means modifying the environment before escalating demands: lighting, sound, transition timing, communication load, seating, predictable routines, recovery space, and adult tone can change what becomes possible.
For community programs, it means designing for nervous-system diversity: adaptive swim instruction, sensory-aware environments, visual supports, reduced uncertainty, trust-building, and regulation-informed safety protocols.
For neurodivergent people themselves, it means replacing shame with pattern recognition: What does my system detect? What costs me more than other people realize? What helps me recover? Where am I thriving because the environment fits? Where am I surviving because it does not?
The bottom line
A high-gain nervous system is not a defective nervous system. It is a sensitive, high-information regulatory architecture that may detect more signal, assign more salience, require more prediction, correct harder, and spend more capacity managing the world before visible behavior even begins.
The same task can have different cost. The same room can create different loads. The same demand can be possible in one state and impossible in another.
This is why support needs should not be reduced to motivation, compliance, or character. They are often signs of brain-body-environment mismatch. The goal is not to normalize the person.
The goal is to reduce preventable load, increase predictability, protect recovery, build capacity, and create environments where different nervous-system architectures can thrive.
References
Anderson, C., Law, J. K., Daniels, A., Rice, C., Mandell, D. S., Hagopian, L., & Law, P. A. (2012). Occurrence and family impact of elopement in children with autism spectrum disorders. Pediatrics, 130(5), 870–877. https://doi.org/10.1542/peds.2012-0762
Arnsten, A. Stress signalling pathways that impair prefrontal cortex structure and function. Nat Rev Neurosci 10, 410–422 (2009). https://doi.org/10.1038/nrn2648
Attwell, D., & Laughlin, S. B. (2001). An energy budget for signaling in the grey matter of the brain. Journal of Cerebral Blood Flow & Metabolism, 21(10), 1133–1145.
Barrett, L. F., & Simmons, W. K. (2015). Interoceptive predictions in the brain. Nature Reviews Neuroscience, 16, 419–429.
Ben-Sasson, A., Hen, L., Fluss, R., Cermak, S. A., Engel-Yeger, B., & Gal, E. (2009). A meta-analysis of sensory modulation symptoms in individuals with autism spectrum disorders. Journal of Autism and Developmental Disorders, 39, 1–11.
Crompton, Catherine J et al. “Autistic peer-to-peer information transfer is highly effective.” Autism : the international journal of research and practice vol. 24,7 (2020): 1704-1712. doi:10.1177/1362361320919286
Foster, G. E., & Sheel, A. W. (2005). The human diving response, its function, and its control. Scandinavian journal of medicine & science in sports, 15(1), 3–12. https://doi.org/10.1111/j.1600-0838.2005.00440.x
Guan, J., Li, G. Characteristics of unintentional drowning deaths in children with autism spectrum disorder. Inj. Epidemiol. 4, 32 (2017). https://doi.org/10.1186/s40621-017-0129-4
Harris, J. J., Jolivet, R., & Attwell, D. (2012). Synaptic energy use and supply. Neuron, 75(5), 762–777.
Lawson, R. P., Rees, G., & Friston, K. J. (2014). An aberrant precision account of autism. Frontiers in Human Neuroscience, 8, 302.
McEwen B. S. (1998). Protective and damaging effects of stress mediators. The New England journal of medicine, 338(3), 171–179. https://doi.org/10.1056/NEJM199801153380307
McEwen, B. S., & Stellar, E. (1993). Stress and the individual: Mechanisms leading to disease. Archives of Internal Medicine, 153(18), 2093–2101.
Murray, D., Lesser, M., & Lawson, W. (2005). Attention, monotropism and the diagnostic criteria for autism. Autism : the international journal of research and practice, 9(2), 139–156. https://doi.org/10.1177/1362361305051398
Pellicano E, Burr D. When the world becomes ‘too real’: a Bayesian explanation of autistic perception. Trends Cogn Sci. 2012 Oct;16(10):504-10. doi: 10.1016/j.tics.2012.08.009. Epub 2012 Sep 7. PMID: 22959875.
Raymaker, Dora M et al. “”Having All of Your Internal Resources Exhausted Beyond Measure and Being Left with No Clean-Up Crew”: Defining Autistic Burnout.” Autism in adulthood vol. 2,2 (2020): 132-143. doi:10.1089/aut.2019.0079
Sterling, P. (2012). Allostasis: A model of predictive regulation. Physiology & Behavior, 106(1), 5–15.
Van de Cruys, S., et al. (2014). Precise minds in uncertain worlds: Predictive coding in autism. Psychological Review, 121(4), 649–675.
CDC “Wandering (Elopement),” Child Development, cdc.gov. https://www.cdc.gov/child-development/disability-safety/wandering.html
Organized by Topic
Allostasis, stress, and regulation
McEwen, B. S., & Stellar, E. (1993). Stress and the individual: Mechanisms leading to disease. Archives of Internal Medicine, 153(18), 2093–2101. doi:10.1001/archinte.1993.00410180039004.
McEwen, B. S. (1998). Protective and damaging effects of stress mediators. New England Journal of Medicine, 338(3), 171–179. doi:10.1056/NEJM199801153380307.
Sterling, P. (2012). Allostasis: A model of predictive regulation. Physiology & Behavior, 106(1), 5–15. doi:10.1016/j.physbeh.2011.06.004.
Arnsten, A. F. T. (2009). Stress signalling pathways that impair prefrontal cortex structure and function. Nature Reviews Neuroscience, 10(6), 410–422. doi:10.1038/nrn2648.
Prediction, interoception, and autism perception models
Barrett, L. F., & Simmons, W. K. (2015). Interoceptive predictions in the brain. Nature Reviews Neuroscience, 16(7), 419–429. doi:10.1038/nrn3950.
Pellicano, E., & Burr, D. (2012). When the world becomes “too real”: A Bayesian explanation of autistic perception. Trends in Cognitive Sciences, 16(10), 504–510. doi:10.1016/j.tics.2012.08.009.
Lawson, R. P., Rees, G., & Friston, K. J. (2014). An aberrant precision account of autism. Frontiers in Human Neuroscience, 8, 302. doi:10.3389/fnhum.2014.00302.
Van de Cruys, S., Evers, K., Van der Hallen, R., Van Eylen, L., Boets, B., de-Wit, L., & Wagemans, J. (2014). Precise minds in uncertain worlds: Predictive coding in autism. Psychological Review, 121(4), 649–675. doi:10.1037/a0037665.
Sensory modulation, monotropism, and double empathy
Ben-Sasson, A., Hen, L., Fluss, R., Cermak, S. A., Engel-Yeger, B., & Gal, E. (2009). A meta-analysis of sensory modulation symptoms in individuals with autism spectrum disorders. Journal of Autism and Developmental Disorders, 39(1), 1–11. doi:10.1007/s10803-008-0593-3.
Murray, D., Lesser, M., & Lawson, W. (2005). Attention, monotropism and the diagnostic criteria for autism. Autism, 9(2), 139–156. doi:10.1177/1362361305051398.
Milton, D. E. M. (2012). On the ontological status of autism: The “double empathy problem.” Disability & Society, 27(6), 883–887. doi:10.1080/09687599.2012.710008.
Crompton, C. J., Ropar, D., Evans-Williams, C. V. M., Flynn, E. G., & Fletcher-Watson, S. (2020). Autistic peer-to-peer information transfer is highly effective. Autism, 24(7), 1704–1712. doi:10.1177/1362361320919286.
Processing cost and neural energy
Attwell, D., & Laughlin, S. B. (2001). An energy budget for signaling in the grey matter of the brain. Journal of Cerebral Blood Flow & Metabolism, 21(10), 1133–1145. doi:10.1097/00004647-200110000-00001.
Harris, J. J., Jolivet, R., & Attwell, D. (2012). Synaptic energy use and supply. Neuron, 75(5), 762–777. doi:10.1016/j.neuron.2012.08.019.
Autistic burnout and safety
Raymaker, Dora M et al. “”Having All of Your Internal Resources Exhausted Beyond Measure and Being Left with No Clean-Up Crew”: Defining Autistic Burnout.” Autism in adulthood vol. 2,2 (2020): 132-143. doi:10.1089/aut.2019.0079
Anderson, C., Law, J. K., Daniels, A., Rice, C., Mandell, D. S., Hagopian, L., & Law, P. A. (2012). Occurrence and family impact of elopement in children with autism spectrum disorders. Pediatrics, 130(5), 870–877. doi:10.1542/peds.2012-0762.
Guan, J., & Li, G. (2017). Characteristics of unintentional drowning deaths in children with autism spectrum disorder. Injury Epidemiology, 4, 32. doi:10.1186/s40621-017-0129-4.
Centers for Disease Control and Prevention. (2026). Wandering (Elopement). CDC Child Development. https://www.cdc.gov/child-development/disability-safety/wandering.html
Foster, G. E., & Sheel, A. W. (2005). The human diving response, its function, and its control. Scandinavian Journal of Medicine & Science in Sports, 15(1), 3–12. doi:10.1111/j.1600-0838.2005.00440.x.
Leave a Reply