Jared Edward Reser, Ph.D.

Abstract

Schizophrenia is generally studied as a human disorder whose characteristic features arise from genetic liability, abnormal neurodevelopment, and environmental adversity. This article approaches the problem from the opposite direction following up on my previous work (Reser, 2007). It begins with the comparative biology of stressed mammals and asks whether schizophrenia recruits mechanisms that ordinarily recalibrate the mammalian brain for dangerous, unstable, or socially adverse environments. Across rodents, guinea pigs, pigs, prairie voles, tree shrews, nonhuman primates, and humans, stress can alter sensorimotor gating, auditory filtering, hippocampal regulation of dopamine, action selection, social behavior, cortical maturation, inhibitory interneurons, extracellular nets, microglia, complement signaling, chromatin regulation, and oligodendrocyte–myelin biology. Many of these same systems are altered in schizophrenia.

The hypothesis proposes that schizophrenia recruits an extreme, persistent, developmentally retimed, or genetically amplified form of conserved stress-calibration mechanisms. These mechanisms can increase vigilance, broaden the admission of weak sensory signals, strengthen rapid threat learning, favor well-practiced actions over slow deliberation, reduce hazardous exploration, and reorganize social behavior after exclusion or defeat. In a prehistoric foraging ecology, these shifts could have improved survival. Their expression at schizophrenia-level intensity could represent the severe tail of a graded adaptation. A second possibility also deserves consideration: natural selection may sometimes have favored the integrated end-spectrum phenotype itself as a rare, condition-dependent human specialization. Present evidence leaves both evolutionary models open.

This comparative hypothesis produces testable predictions. It predicts coordinated cross-species profiles, strong effects of developmental timing, sex, social buffering, and environmental match, convergence between stress and schizophrenia within specific cell types and circuits, and incomplete recovery when a defensive state becomes biologically stabilized. If supported, the hypothesis would help divide schizophrenia into mechanistic subtypes, identify earlier markers of risk, discover capacities associated with the phenotype, and direct treatment toward context sensitivity, social regulation, goal-directed control, sleep, and neural plasticity. It gives evolutionary psychiatry a practical role in determining what the affected systems were built to do and how their original functions can inform humane care.

Keywords: schizophrenia; psychosis; comparative neuroscience; chronic stress; early-life adversity; phenotypic plasticity; sensorimotor gating; dopamine; habit; microglia; HDAC1; neurodevelopment

1. Introduction

Schizophrenia arises from a distributed combination of genetic liability, neurodevelopment, environmental experience, and current physiological state. Its heritability is spread across many genetic variants, its onset is shaped by development, and its expression is strongly influenced by stress, trauma, social adversity, migration, isolation, sleep disruption, and substance exposure. The syndrome is heterogeneous. Hallucinations, delusions, disorganization, withdrawal, motivational change, and cognitive impairment vary independently across people and over time.

This complexity is commonly described as a collection of interacting dysfunctions. Some processes are overactive, others underactive, and their convergence produces illness in the present environment. A deeper biological account also asks why the affected mechanisms exist, why stress regulates them, and why similar changes repeatedly appear in other mammals.

My claim is that comparative evidence offers a way forward. Mammals exposed to predation risk, unstable care, social defeat, isolation, resource uncertainty, or chronic restraint often change in organized, developmentally timed, and context-sensitive ways. Sensory filters are adjusted. Threat learning is strengthened. Exploration and social behavior change. Control can shift from flexible, outcome-based decisions toward faster habits. Hippocampal activity can alter dopamine recruitment. Critical periods can open or close at different times. Glia, myelin, inhibitory interneurons, and synaptic pruning are modified. These responses carry costs and have intelligible functions in adverse environments.

Many of the systems recalibrated by mammalian stress are the same systems implicated in schizophrenia. This structured overlap warrants a specific hypothesis: schizophrenia expresses a conserved stress-calibration phenotype whose intensity, persistence, developmental organization, and ecological fit vary among people. Mismatch and internal desynchronization may explain some cases, while ancestral selection for a rare integrated phenotype remains possible in others.

This formulation extends the phenotypic-plasticity account that I previously proposed for schizophrenia (Reser, 2007) and the broader neuroecological model of chronic stress (Reser, 2016). The present argument places the comparative evidence first and the evolutionary interpretation second. Measurable biological correspondences across mammals establish the foundation. They then allow us to ask which ecological problems the conserved responses normally solve and whether natural selection ever favored their full human expression.

2. The hypothesis

The stress-calibrated neurodevelopment hypothesis has four parts.

First, mammals possess conditional developmental programs that use experience as information. Early adversity, chronic danger, disrupted social support, and repeated uncontrollability can forecast the kind of environment in which an organism will have to survive. Hormones, immune signals, neural activity, and social cues help tune the developing brain to that forecast. This is phenotypic plasticity: the capacity of one genotype to produce different traits under different environmental conditions.

Second, this calibration is distributed across sensory gating, the hippocampus, prefrontal and striatal circuits, dopamine systems, inhibitory interneurons, glia, myelination, circadian organization, and social neuropeptides. Because the component systems mature at different times, adversity can produce a mosaic phenotype in which some circuits are accelerated, others delayed, and still others diverted onto a different developmental course.

Third, schizophrenia can recruit this same distributed machinery. Genetic liability may lower the activation threshold, increase the gain of individual pathways, weaken recovery, or cause ordinarily coordinated components to become desynchronized. Severe or repeated stress may drive the system farther. A phenotype that would have been temporary or moderate can become prolonged and self-reinforcing.

Fourth, the evolutionary target may exist at more than one level. Natural selection may have favored a graded conditional program whose ordinary settings improved vigilance, threat learning, rapid action, and social caution. Schizophrenia could occupy an intense tail of that program. Natural selection may also have favored the coordinated end-spectrum phenotype under rare ancestral conditions. Hallucinations, fixed convictions, radical salience shifts, withdrawal, unusual association, and reduced commitment to conventional goals could have functioned together as an emergency strategy or social-cognitive specialization. Adaptations can be painful and costly. Anxiety, depression, fever, and defensive aggression all illustrate how survival functions can impose suffering. Modern disability therefore provides limited information about ancestral fitness.

This model is compatible with genetic, neurodevelopmental, dopamine, glutamate, immune, and social accounts of schizophrenia. It asks how their findings fit within an ancient environmental-response system and whether that system sometimes produced a complete alternative human phenotype.

3. Terms and levels of comparison

Comparative psychiatry works best when behavior, assay, mechanism, and diagnosis are kept at their proper levels. A rat experiment can illuminate a human mechanism by showing the same stress-sensitive circuit, cellular process, or information-processing operation in both species. The full human phenotype can then be studied as a higher-order configuration of these conserved elements.

Term	Meaning in this article
Phenotypic plasticity	The ability of an organism to develop different traits in response to environmental information.
Stress calibration	A lasting adjustment of physiology or behavior based on cues of danger, scarcity, instability, or social adversity.
Positive symptoms	Experiences added to ordinary mental life, including hallucinations, delusions, and unusually intense salience. “Positive” means added, not beneficial.
Negative symptoms	Reductions in motivation, emotional expression, speech, pleasure, or social engagement. “Negative” means diminished, not morally bad.
Endophenotype	A measurable feature—such as reduced sensory gating—that can connect genes and mechanisms to one component of a complex clinical syndrome.
Sensorimotor gating	The brain’s ability to reduce its response to a strong event when a weaker warning signal occurs immediately before it. It is commonly measured with prepulse inhibition.
Prepulse inhibition (PPI)	Reduction of the startle reflex when a faint “prepulse” precedes the startling stimulus. Lower PPI indicates that the nervous system is filtering or integrating the warning cue differently.
Auditory gating	Suppression of the neural response to the second of two repeated sounds. Weak suppression suggests reduced filtering of redundant input.
Mismatch negativity (MMN)	An automatic electrical brain response to a sound that violates a regular pattern. It reflects rapid, largely preconscious prediction and updating.
NMDA receptor	A type of glutamate receptor that helps neurons change the strength of their connections during learning and development. NMDA-receptor function is also implicated in schizophrenia.
Aberrant salience	Excessive importance assigned to stimuli, thoughts, or coincidences that would ordinarily be neutral. Dopamine is one contributor to this assignment of significance.
Goal-directed or model-based control	Choosing an action by representing its likely outcome and whether that outcome is currently valuable. It is flexible but cognitively demanding.
Habitual or model-free control	Repeating actions that have been reinforced in the past without fully reevaluating their present outcome. It is fast and robust but less flexible.
Parvalbumin (PV) interneuron	A fast-spiking inhibitory neuron that helps synchronize cortical activity and regulate the timing and gain of information processing.
Perineuronal net (PNN)	A specialized extracellular matrix that forms around some neurons, especially PV interneurons, and stabilizes mature circuit organization.
Microglia and complement	Microglia are resident immune cells of the brain. Complement proteins can mark synapses for removal, allowing microglia to participate in activity-dependent circuit refinement.
Oligodendrocyte and myelin	Oligodendrocytes form myelin, the insulating material that increases the speed and reliability of signaling along axons.
HDAC1	Histone deacetylase 1, an enzyme that can change which genes remain accessible by modifying chromatin, the protein–DNA structure that regulates gene expression.
HPA axis and glucocorticoids	The hypothalamic–pituitary–adrenal axis is a hormonal stress-control system. Its main glucocorticoid is cortisol in humans and many mammals and corticosterone in rats and mice.
Critical period	A developmental interval when experience has an unusually strong and lasting effect on how a neural circuit is organized.
Heterochrony	A change in the timing or rate of development. A circuit may mature earlier, later, faster, or for a shorter period, producing an uneven developmental profile.
Social buffering	Reduction of a stress response through the presence or support of a familiar social partner.
Match–mismatch	The possibility that a phenotype prepared by early cues functions better in a similar later environment and worse when later conditions differ from the forecast.
Environment of evolutionary adaptedness (EEA)	The recurring ecological and social conditions that shaped a trait over evolutionary time. The EEA refers to a pattern of selection pressures across many generations and places.
Balancing selection	Evolutionary processes that preserve multiple forms of a trait in a population because each has advantages under particular conditions.
Antagonistic pleiotropy	A genetic effect that increases fitness in one period or context while producing costs in another.
Frequency-dependent selection	Selection in which a phenotype’s value changes with how common it is. A rare cognitive or social strategy can be useful partly because most group members use another strategy.
Inclusive fitness	Genetic success gained through one’s own survival and reproduction plus effects on the survival and reproduction of genetic relatives.
Neurodiversity	Naturally occurring variation in how human nervous systems perceive, learn, assign salience, regulate emotion, and organize behavior. Neurodiversity can include distinctive capacities, vulnerabilities, and support needs.

The strongest comparative case rests on five kinds of correspondence. The first is phenotypic correspondence: stress changes an assay in animals that is also altered in schizophrenia. The second is circuit correspondence: the same neural pathway produces the change. The third is molecular correspondence: the same cell type or signaling system is involved. The fourth is developmental correspondence: timing, sex, and sensitive periods influence both processes in similar ways. The fifth is functional correspondence: the change has an intelligible effect on survival-related behavior.

Each correspondence contributes a different form of evidence. A repeated pattern across all five levels forms the strongest comparative case.

4. Comparative evidence across mammals

4.1 Sensorimotor and auditory gating

The brain cannot process every stimulus at equal strength. It filters, suppresses repetition, and uses context to decide what deserves access to attention and action. Schizophrenia is associated with disruptions in several forms of this gating. Reduced PPI is one of the most widely used cross-species measures because startle modulation can be tested with closely related procedures in humans and other mammals.

Post-weaning social isolation in rats produces a developmentally specific reduction in PPI together with novelty-related hyperactivity and an enhanced dopamine response to amphetamine (Wilkinson et al., 1994). The timing matters. Isolation during a juvenile period produces a different outcome from the same duration of adult isolation. This pattern is evidence of developmental calibration.

Prenatal variable stress in rats can also reduce adult PPI, weaken suppression of the second response in a paired-click auditory-gating task, and increase later sensitivity to amphetamine (Koenig et al., 2005). Maternal experience has altered the offspring’s developmental trajectory by the time adult testing occurs. Socially isolated rats likewise show impaired paired-click gating, and nicotine can temporarily normalize the measure (Stevens et al., 1997). Heavy nicotine use is common in schizophrenia, and this experimental effect demonstrates that a stress-sensitive gating deficit can be pharmacologically modulated through receptor systems also studied in the disorder. The health consequences of smoking make nicotine delivery an unsuitable clinical inference from this finding.

The same relationship appears outside conventional laboratory rats. Prenatal stress in guinea pigs changes PPI in a sex- and age-dependent manner. Juvenile males can show increased PPI, whereas adult males show reduced PPI; testosterone replacement can alter the adult effect (Kapoor and Matthews, 2011). This reversal across age is an example of heterochrony: the direction of the result depends on when development is sampled and on the endocrine state of the animal.

Acute stress can reduce PPI through the neurosteroid allopregnanolone in the prefrontal cortex (Cadeddu et al., 2022). Allopregnanolone rapidly changes GABA-A receptor function, and GABA is the brain’s principal inhibitory transmitter. The study provides a short mechanistic route from a stressful event to altered gating. It also illustrates an important principle: acute neurochemistry, developmental history, strain, sex, and chronic experience can all influence the same assay. Reduced PPI is a final common measurement reached through several pathways.

Pigs also display measurable PPI, and amphetamine disrupts it (Lind et al., 2004). This matters methodologically. Pigs have large, highly folded brains and sensory and social capacities that complement rodent studies. A conserved assay across rats, guinea pigs, pigs, nonhuman primates, and humans allows us to test whether a stress effect generalizes across distinct mammalian life histories.

Why would reduced gating ever be useful? Gating is a tradeoff. Strong gating protects a stable cognitive stream from distraction. Weaker gating admits more faint, novel, or peripheral information. In a dangerous environment, the crack of a branch, a slight change in birdsong, an unfamiliar odor, or an unusual movement in the grass may be worth interrupting ongoing behavior. The advantage is sensitivity; the cost is overload. When gating is weakened too far, the nervous system can become flooded by stimuli that other people automatically suppress.

MMN provides a related bridge. It measures the automatic detection of a violation in an auditory pattern. MMN is reliably reduced in many people with schizophrenia and is associated with functional outcome (Harms et al., 2020). Comparable MMN-like responses can be measured in rats and macaques (Harms et al., 2014; Gil-da-Costa et al., 2020). This creates an excellent comparative assay. The decisive direct test—whether defined stress histories produce the same MMN alteration across species and through the same circuit—remains an open research opportunity.

4.2 Hippocampal regulation of dopamine and the assignment of salience

Dopamine contributes to learning, action, motivation, and the assignment of importance. A stimulus that predicts food, danger, novelty, or a change in strategy can recruit dopamine. In psychosis, dopamine can be released or interpreted in ways that make neutral events feel unusually significant. The person then has to explain a genuine feeling that something important is happening. This is the core of the aberrant-salience account.

The hippocampus provides context and memory to this system. It helps answer whether the current setting is familiar, safe, uncertain, or associated with prior danger. The ventral tegmental area is a midbrain source of dopamine, and the ventral striatum helps translate motivation and predicted value into action. In rats, aversive stimuli increase the population activity of ventral-tegmental dopamine neurons through a pathway involving the ventral hippocampus and ventral striatum (Valenti, Lodge, and Grace, 2011). Persistent hippocampal overactivity can therefore place the dopamine system in a high-gain state in which more events acquire motivational force.

This circuit is especially relevant to schizophrenia. Hippocampal hyperactivity and striatal dopamine dysregulation are both prominent findings in psychosis research, and animal models provide a mechanistic link between them. Developmental stress can alter hippocampal inhibition, dendritic structure, neurogenesis, and glucocorticoid regulation. Chronic psychosocial stress in tree shrews, for example, reduces hippocampal neurogenesis and remodels dendrites (Fuchs et al., 2001). Their chronic social-subordination paradigm models an ethologically meaningful form of psychosocial stress.

Human findings point in the same direction. Psychosocial stress can evoke greater dopamine release in people at clinical high risk for psychosis and in schizophrenia, and the interaction between trauma history and dopamine release predicts experimentally induced psychotic symptoms (Mizrahi et al., 2012; Tseng et al., 2018; Dahoun et al., 2019). Environmental history, dopamine reactivity, and psychotic experience are quantitatively connected in humans, as the comparative circuit predicts.

In ecological terms, raising the gain on novelty and uncertainty can be useful during displacement, predation, conflict, or rapid environmental change. Weak signals are learned faster and unusual events interrupt routine. The liability is false significance. A system tuned to avoid missed threats will necessarily produce more false alarms. At a clinical extreme, coincidences can feel directed, background events can seem personally relevant, and internally generated thoughts can acquire an external or urgent quality.

4.3 From flexible planning to habit

Under stable conditions, it pays to deliberate. An animal can compare options, estimate consequences, and update behavior when an outcome loses value. Under immediate or repeated threat, prolonged deliberation can be costly. Fast, practiced actions may be more reliable.

Neuroscience distinguishes goal-directed control from habitual control. Goal-directed actions depend on a representation of the relationship between an action and its current outcome. Habits depend more strongly on reinforcement history. A habitual action can continue even after its outcome becomes less valuable.

Chronic stress in rats causes structural reorganization of frontostriatal circuits and shifts behavior toward habit (Dias-Ferreira et al., 2009). Dendritic changes occur in regions associated with flexible action selection, while regions supporting habitual control are strengthened. The stressed animal becomes less sensitive to changes in outcome value. The result is a redistribution of behavioral control.

Recent mouse work has identified two pathways by which stress can produce this shift. Stress weakens a basolateral-amygdala pathway to the dorsomedial striatum that supports action–outcome knowledge, while recruiting a central-amygdala pathway that promotes habitual responding (Giovanniello et al., 2025). The amygdala can therefore change which decision system governs behavior.

A related shift occurs in stressed humans and can reverse after recovery from the stressor (Soares et al., 2012). People with schizophrenia show impairments in action–outcome learning and reduced model-based control (Morris et al., 2018). The comparison is functionally coherent: chronic stress and schizophrenia both reduce the influence of represented future consequences on present action.

Habit allows rapid, low-cost execution of skills and frees attention for other problems. In a dangerous ecology, familiar routes, shelter-building sequences, defensive responses, food-processing routines, and coalition rules may need to be executed under high arousal. When habitual control expands across too many situations, behavior becomes rigid, context updating weakens, and old responses continue after circumstances change. Some negative and cognitive features of schizophrenia may partly reflect a redistribution away from flexible agency and toward established action patterns.

4.4 Social withdrawal, defeat, and social buffering

Mammals regulate stress socially. Separation from caregivers, loss of a partner, low rank, and defeat can alter the hypothalamic–pituitary–adrenal axis, autonomic activity, immune function, dopamine, oxytocin, vasopressin, and behavior. Conversely, a familiar partner can reduce the response to a stressor. This is social buffering.

Prairie voles make the principle especially clear because they form selective pair bonds. Post-weaning isolation changes anxiety-like and social behaviors and modifies dopamine, oxytocin, vasopressin, and stress-related gene expression (Pan et al., 2009). A bonded partner can reduce behavioral and adrenal responses to chronic mild stress (McNeal et al., 2017). These findings show that conserved attachment systems regulate threat physiology, reward, and exploration.

Marmosets also depend heavily on family care. Repeated parental deprivation during early development alters later reward-related effort, behavioral inhibition, and physiological homeostasis (Pryce et al., 2004). Rhesus macaques exposed to adverse early rearing show long-lasting changes in monoamine responses, social behavior, stereotypy, and hypothalamic–pituitary–adrenal regulation (Feng et al., 2011; Wood et al., 2021). Maternal separation can also alter the developmental trajectory of immature amygdala neurons and produce persistent changes in amygdala microglia (McHale-Matthews et al., 2023; King et al., 2025). These findings extend the stress-calibration pattern into primates with prolonged development and complex social cognition.

Pigs provide support and an important boundary condition. Early social isolation can alter behavior, cortisol regulation, immune responses, and hippocampal interleukin-1 beta (Tuchscherer et al., 2004). Social support can reduce cortisol responses and modify glucocorticoid-related gene expression (Kanitz et al., 2016). A brief overnight isolation can leave spatial learning and decision-making intact (van der Staay et al., 2015). Duration, developmental stage, predictability, and social meaning determine the outcome.

Social withdrawal after defeat can have a short-term function. It can reduce renewed aggression, exposure to hostile group members, infection risk, or costly attempts to compete when the probability of success is low. Reduced exploration can keep an animal near known shelter. Increased observation from the social periphery may allow reevaluation of alliances. The cost is the loss of social buffering itself. Once an individual withdraws, the absence of corrective social information and affiliative regulation can intensify stress, sleep disruption, suspiciousness, and motivational loss. A protective short-term response can therefore become a self-amplifying chronic state.

This interpretation is particularly relevant to schizophrenia because social adversity is both a risk factor and a consequence. The model predicts feedback: adversity changes social behavior; altered behavior reduces support and opportunity; reduced support increases stress; and the resulting physiology stabilizes further withdrawal. Treatment can interrupt this loop by restoring safety, connection, and opportunities for successful engagement.

4.5 Developmental timing: acceleration, delay, and mosaic maturation

Early-life stress is often described as “damaging development.” That phrase is too coarse. Development can be accelerated in one system, delayed in another, and redirected in a third.

Fragmented maternal care in mice accelerates several indices of hippocampal maturation in males. Parvalbumin expression rises earlier, an NMDA-receptor subunit transition occurs sooner, myelin basic protein changes, and emotional learning becomes adult-like at an earlier age (Bath, Manzano-Nieves, and Goodwill, 2016). Early transport stress in mice can advance the timing of the visual-cortex critical period and alter perineuronal-net development (Poplawski et al., 2024). These findings show that adversity changes when circuits stabilize as well as their later structure.

Humans show a comparable principle. Children exposed to early maternal deprivation can display an earlier adult-like pattern of amygdala–medial-prefrontal connectivity. Cortisol partly mediates the relationship, and the altered connectivity can buffer anxiety in the short term (Gee et al., 2013). This finding is important because it demonstrates an interpretable developmental tradeoff. Earlier maturation of threat-regulation circuitry may provide earlier emotional independence when dependable care is unavailable. The cost may be premature stabilization and a shorter period for experience-dependent refinement.

Schizophrenia is also characterized by developmental unevenness. Some deviations are present long before psychosis, while others emerge during adolescence or early adulthood, when dopamine systems, prefrontal networks, myelination, social roles, and sleep patterns are being reorganized. The stress-calibration hypothesis predicts heterochrony: threat and salience systems may gain influence before contextual, inhibitory, and social-cognitive systems are ready to regulate them.

This helps explain why the same early stressor can produce different outcomes at different ages. A juvenile may initially look mature or unusually independent, while later tests reveal reduced flexibility. Sex hormones can reveal or reverse effects that were initially subtle. The phenotype unfolds as a trajectory.

5. Molecular and cellular convergence

Behavioral comparisons are most persuasive when they are supported by common cellular mechanisms. Several systems now provide that bridge.

5.1 HDAC1 and the biological memory of early adversity

Experience can persist by altering gene regulation while the DNA sequence remains stable. Epigenetic regulation refers to biochemical processes that make genes more or less accessible. Histone deacetylases remove acetyl groups from the proteins around which DNA is wrapped, generally making parts of chromatin less accessible to transcription.

HDAC1 is unusually relevant because one experimental program connects early-life stress, adult schizophrenia-related phenotypes, prefrontal gene regulation, and human tissue. In mice, early-life stress produced a delayed increase of HDAC1 in the adult medial prefrontal cortex, together with memory and PPI deficits. Increasing HDAC1 experimentally reproduced important parts of the phenotype, while an HDAC inhibitor rescued behavioral and molecular effects. In human postmortem prefrontal tissue, HDAC1 messenger RNA and protein were elevated in schizophrenia, and blood HDAC1 was associated with schizophrenia and early-life stress in a smaller clinical sample (Bahari-Javan et al., 2017).

HDAC1 proposes a mechanism by which adversity can become biologically embedded and remain latent until a later developmental stage. The effect’s delay resembles the clinical timing of schizophrenia, in which liability can accumulate before psychosis becomes apparent. The human samples were limited, making replication and refinement by cell type important next steps. Even at this stage, HDAC1 illustrates the kind of comparative causal chain the hypothesis predicts.

5.2 Complement, microglia, and synapse selection

The developing brain produces more synaptic connections than it keeps. Activity, experience, and molecular tags help determine which connections are stabilized and which are removed. Complement proteins, originally characterized in immunity, can mark synapses. Microglia can then engulf tagged material.

Genetic variation that increases expression of complement component 4A is associated with schizophrenia risk (Sekar et al., 2016). Patient-derived cellular models show increased microglial uptake of synaptic material, with complement contributing to the effect (Sellgren et al., 2019). Elevated C4A has also been detected in the cerebrospinal fluid of people with first-episode psychosis who subsequently developed schizophrenia (Gracias et al., 2022).

Stress reaches the same machinery. Chronic stress in male mice can activate complement and microglia and cause C3-dependent loss of specific thalamocortical synapses in superficial layers of the medial prefrontal cortex. Removing C3 protects both synapses and working memory (Tillmon et al., 2024). The effect shows layer-, input-, and pathway-sensitive remodeling.

From a calibration perspective, synapse selection could prioritize circuits that are repeatedly useful in an adverse environment. The pathological risk is excessive, mistimed, or poorly targeted removal, especially during adolescence. Genetic complement liability and environmental stress could converge on the same pruning machinery. This yields a direct gene-by-environment prediction: complement-associated risk should have its strongest cognitive and circuit effects in particular stress histories and developmental windows.

5.3 Parvalbumin interneurons and perineuronal nets

PV interneurons provide rapid inhibition and help coordinate cortical rhythms. Their perineuronal nets regulate ionic stability, protect against oxidative stress, and constrain plasticity after a circuit has matured. These two elements form a natural interface between excitation, inhibition, critical-period timing, and long-term stabilization.

Stress affects the PV–PNN system in a duration- and region-dependent manner. Short restraint stress can increase PV and PNN measures in medial prefrontal cortex, consistent with accelerated stabilization (Pesarico et al., 2019). Prolonged social isolation can progressively reduce prefrontal PV expression through oxidative mechanisms (Schiavone et al., 2012). In schizophrenia, PV-cell density can remain intact while PV protein labeling and perineuronal-net components become weaker in dorsolateral prefrontal cortex (Enwright et al., 2016).

The apparent difference between early strengthening and later weakening may reveal the process. Initial stress can drive rapid maturation or closure of plasticity; continued oxidative and excitatory load can then degrade the same inhibitory system. A cross-sectional adult sample records the endpoint and can obscure the sequence. Longitudinal comparative studies can test whether schizophrenia likewise involves an early stabilization phase followed by decompensation in a vulnerable subgroup.

5.4 Oligodendrocytes, myelin, and network economy

Myelin provides insulation and regulates timing. Coordinated thought depends on signals arriving at the right time across distant networks. Oligodendrocytes support axons metabolically and adjust conduction through myelin formation.

Chronic social defeat reduces myelination in mouse medial prefrontal cortex, and chronic stress can disrupt the progression of oligodendrocyte-lineage cells (Lehmann et al., 2017; Kokkosis et al., 2022). Schizophrenia is associated with altered white matter and impaired differentiation of oligodendrocyte precursors in prefrontal cortex (Mauney et al., 2015). These findings are consistent with shared stress-sensitive network-timing machinery.

It would be too simple to call reduced myelination an adaptation. Some of it may be a resource cost, a consequence of inflammation, or a failure of repair. The functional hypothesis applies at the system level: adversity reallocates developmental investment and changes which circuits are repeatedly engaged and stabilized. In a severe or genetically vulnerable case, the resulting myelin pattern may undermine long-range coordination and contribute to cognitive fragmentation. Comparative work should distinguish active, tract-specific remodeling from generalized loss.

5.5 Neurosteroids and rapid inhibitory control

Neurosteroids such as allopregnanolone are synthesized within the nervous system and can quickly alter GABA-A receptors. Because GABA controls neural inhibition, this provides a rapid route by which stress can change gating, arousal, and network stability. The finding that acute stress impairs PPI through prefrontal allopregnanolone links an endocrine response directly to a schizophrenia-relevant assay (Cadeddu et al., 2022).

This pathway may help explain state dependence. A person can move closer to or farther from a psychotic threshold over hours or days as sleep, stress hormones, neurosteroids, and inhibitory balance change. The long-term phenotype may be developmentally prepared, while acute physiology determines when it becomes clinically expressed.

6. The comparative pattern

The evidence reveals a specific cross-level pattern linking stress calibration to schizophrenia-relevant biology.

Comparative finding	Mammalian function affected	Correspondence in schizophrenia	Hypothesized tradeoff
Reduced PPI or auditory gating after prenatal or juvenile stress	Filtering of weak and redundant stimuli	Reduced gating in many patients and relatives	Greater admission of weak cues at the cost of sensory flooding
Hippocampal recruitment of dopamine during aversive states	Contextual salience, learning, mobilization	Hippocampal hyperactivity and striatal dopamine dysregulation	Faster response to uncertain events at the cost of false significance
Stress-induced habit bias	Rapid action selection under load	Reduced action–outcome and model-based control	Speed and robustness at the cost of flexibility
Isolation and defeat effects on social and endocrine systems	Affiliation, rank, exploration, social buffering	Withdrawal, reduced social reward, stress sensitivity	Short-term avoidance of conflict at the cost of chronic isolation
Accelerated threat-circuit or critical-period maturation	Earlier independence and stabilization	Developmental unevenness before psychosis	Earlier competence at the cost of prematurely constrained plasticity
HDAC1 elevation after early stress	Long-term regulation of gene expression	Elevated HDAC1 in prefrontal tissue in schizophrenia	Stable environmental memory at the cost of a difficult-to-reverse state
Complement-dependent stress synapse loss	Experience-dependent circuit refinement	C4-associated genetic risk and excess synaptic uptake	Efficient pruning at the cost of excessive or mistimed elimination
PV–PNN remodeling	Inhibitory timing and stabilization	Reduced PV/PNN integrity in prefrontal cortex	Rapid circuit stabilization at the cost of later fragility or rigidity
Oligodendrocyte and myelin changes	Long-range timing and metabolic support	White-matter and oligodendrocyte abnormalities	Reallocation and circuit specialization at the cost of integration

Three features of this pattern are especially important.

First, it crosses levels. The comparison extends from behavior to circuits, cell types, and gene regulation. Second, it is developmentally conditional. The same event can have different effects before birth, after weaning, during adolescence, or in adulthood. Third, it contains tradeoffs. The change can improve one operation while degrading another.

Boundary findings help reveal the calibration rules. Brief intermittent separation followed by reunion can improve later response inhibition and reduce stress reactivity in squirrel monkeys (Levine and Mody, 2003; Parker et al., 2005). Mild early stress can increase resilience. Early adversity can sometimes improve performance when the adult environment resembles the predicted environment. Maternal deprivation and later corticosterone exposure can interact in nonadditive ways (Choy et al., 2009), and short isolation can leave cognition intact in pigs (van der Staay et al., 2015).

These findings show that the mammalian stress response follows a calibrated, nonlinear course. Dose, controllability, predictability, recovery, social reunion, developmental timing, sex, genotype, and later environmental match shape the trajectory. The hypothesis therefore predicts a family of conditional outcomes that includes resilience, specialized performance, transient psychosis, persistent schizophrenia, and recovery.

7. Evolutionary possibilities in prehistoric foragers

The comparative evidence permits the evolutionary question to be asked more precisely. The environment of evolutionary adaptedness (EEA) refers to the recurring conditions that shaped a phenotype over many generations. It was a varied set of foraging ecologies rather than one time or place. Human ancestors repeatedly faced variable food availability, predation, intergroup violence, injury, infection, bereavement, status loss, forced movement, unstable alliances, and periods in which reliable social support disappeared. These conditions sometimes rewarded rapid detection, decisive action, persistent vigilance, unusual inference, solitary endurance, and resistance to prevailing group assumptions.

7.1 Functions of the component changes

The proposed phenotype reweights cognitive, behavioral, and developmental priorities.

Weaker gating could increase detection. A forager in a safe and familiar camp benefits from suppressing irrelevant sounds. A displaced, hunted, or isolated individual may benefit from admitting weak signals. The tradeoff is distractibility and sensory overload. At high intensity, internal and external signals can compete strongly for attention.

Higher salience gain could accelerate learning. When conditions change abruptly, treating novelty as important helps an individual revise predictions. A faint track, a change in animal movement, an unfamiliar human mark, or a pattern in the behavior of rivals may deserve immediate attention. High salience gain increases false pattern detection as it reduces missed detection. Psychotic ideas may emerge when a system built to find consequential relationships assigns great precision to weak coincidences.

Threat generalization could reduce fatal repetition. After one dangerous encounter, avoidance of similar places, people, smells, or circumstances can be life-saving. Broad generalization exchanges accuracy for caution. At the far end, ambiguous social behavior can be interpreted as hostile and physiologically safe environments can continue to feel dangerous.

Habitual control could preserve action under stress. Escape routes, defensive routines, food-processing sequences, tool use, shelter construction, and responses to known predators can be executed rapidly without simulating every outcome. Habit also reduces the cognitive burden of action when sleep deprived or frightened. Perseveration is the corresponding cost when the old rule loses its value.

Reduced exploration could conserve safety and energy. Novel places carry information and resources along with predation, conflict, and injury. After repeated danger, staying near known terrain and relying on familiar routines can be prudent. This setting conserves safety and energy while reducing access to new resources and corrective experience.

Social withdrawal could reduce immediate interpersonal danger. Following defeat, accusation, bereavement, or loss of rank, withdrawal can prevent further attack and allow observation of a changing coalition. Suspicion can protect against exploitation when trust has become costly. Prolonged withdrawal eventually reduces access to allies, shared information, food, care, and emotional regulation.

Earlier maturation could prepare a child for unreliable care. If protection is absent, earlier threat learning, emotional self-reliance, and stabilization of locally useful skills may improve short-term survival. Earlier stabilization can shorten the period available for later developmental flexibility.

Sleep fragmentation and vigilance could protect during acute danger. Shallow or interrupted sleep permits monitoring during a period of threat. Its value may be greatest across nights or weeks, while chronic continuation destabilizes cognition and increases psychosis risk.

7.2 Two evolutionary models for schizophrenia-level expression

The first model treats schizophrenia-level expression as the intense end of a graded reaction norm. A reaction norm is the rule by which environmental conditions map onto different phenotypes. In this account, most of the range was useful, while the severe clinical tail appears when the environmental input is extreme, genetic thresholds are low, developmental components become unsynchronized, or recovery mechanisms fail. Hallucinations and fixed delusions can arise as high-gain outputs of perceptual, predictive, and salience systems selected for their performance at lower intensity.

The second model proposes that natural selection sometimes favored the integrated end-spectrum phenotype itself. In its strongest form, full-spectrum schizophrenia was a complete adaptation whose combined positive, negative, and cognitive features increased direct or inclusive fitness under recurrent conditions in the EEA. Full schizophrenia-level organization may have been a rare, condition-dependent specialization within human groups. Its component features can form a coordinated state: weak gating admits remote cues; heightened salience makes them urgent; unusual association generates unconventional interpretations; strong conviction sustains action under ambiguity; withdrawal reduces conformity and interpersonal exposure; altered reward reduces investment in ordinary status pursuits; vigilance and fragmented sleep maintain monitoring; and habit preserves a narrow set of defensive routines. The whole configuration may have had properties that its components lacked in isolation.

Several ancestral roles could have rewarded such a configuration. A highly vigilant individual might detect environmental or social changes before others. A person capable of forming remote associations might identify patterns in weather, animal behavior, illness, conflict, or resource movement. Reduced conformity could permit rejection of a group consensus that had become dangerous. Solitary tolerance could support scouting, peripheral observation, night vigilance, or survival after exclusion. Unusual perceptual and visionary experiences could contribute to ritual, healing, divination, narrative innovation, or collective meaning in cultures that provided an intelligible role for them. Intense conviction could mobilize decisive action during a crisis when ordinary deliberation had stalled.

These possibilities describe potential ancestral performance, including its risks. A powerful specialization can produce both extraordinary success and catastrophic error. Evolution regularly preserves phenotypes with high variance in outcome when occasional large benefits compensate for frequent costs. The modern terms “full-blown schizophrenia” and “severe mental illness” are defined through distress, disability, social expectations, and contact with clinical systems. An ancestral expression of the same underlying organization may have unfolded episodically, received more social containment, occupied a recognized niche, or operated in a sensory and social environment better matched to its attentional style. It may also have appeared in a form closely resembling contemporary schizophrenia. In plain language, a phenotype that is disabling in many modern settings could have operated as an ancestral superpower under the conditions that elicited and supported it. Both possibilities remain open.

Several evolutionary mechanisms could preserve an end-spectrum phenotype. Balancing selection can maintain different cognitive strategies because their relative value changes across environments. Frequency-dependent selection can favor a rare strategy precisely because most group members perceive and behave conventionally. Antagonistic pleiotropy can preserve genes that increase early survival, creativity, vigilance, mating success, or kin protection while increasing later psychiatric vulnerability. Conditional developmental activation can restrict the phenotype to periods when adversity makes its benefits most likely. Benefits to close relatives can also contribute through inclusive fitness if vigilance, warning, innovation, healing, or group coordination improves the survival of genetic kin.

This expanded view is consistent with neurodiversity. Human groups may have benefited from maintaining a wide distribution of perceptual gain, associative distance, social motivation, threat sensitivity, and openness to anomalous experience. Schizophrenia may represent one end of that distribution. Present-day suffering establishes an urgent need for support; it leaves the phenotype’s value in the EEA as an empirical evolutionary question.

The two models make different predictions. The graded-tail model predicts ecological advantages at moderate expression and declining performance as the complete syndrome emerges. The full-phenotype model predicts circumstances in which people near the diagnostic end of the spectrum show an integrated advantage, especially in detection, unconventional inference, crisis behavior, solitary persistence, or socially recognized visionary roles. Research can test both models without deciding the answer in advance.

8. Predictions that distinguish the hypothesis

An evolutionary account becomes scientifically useful by changing what we measure. This hypothesis makes the following predictions.

8.1 Coordinated profiles will outperform single markers

Risk should be predicted better by a joint profile—stress history, cortisol dynamics, gating, MMN, model-based control, social buffering, sleep, hippocampal activity, and dopamine responsivity—than by any one measure. The profile should identify subtypes with different routes to psychosis.

8.2 Timing will matter as much as exposure

Prenatal stress, disrupted juvenile care, adolescent isolation, and adult defeat should be analyzed separately. Each should produce a different sequence across gating, PV–PNN maturation, myelination, social behavior, and dopamine. The strongest studies will repeatedly measure the same individuals from before exposure through recovery.

8.3 Stress and schizophrenia will converge within specific cell types

Bulk tissue can hide opposing changes in different cells. Single-cell and spatial transcriptomic studies should test whether stress-exposed animals and human schizophrenia tissue share regulatory programs within PV interneurons, oligodendrocyte-lineage cells, microglia, hippocampal subfields, and defined prefrontal layers. The prediction concerns directional convergence within matched pathways and cell types.

8.4 Genetic liability will alter calibration thresholds and recovery

C4A, dopamine-regulatory, glutamatergic, redox, circadian, and chromatin-related variants should influence which component responds, how strongly it responds, and whether it returns to baseline. Some risk variants may remain quiet in a neutral environment while changing the slope of response to adversity.

8.5 Social buffering will have measurable neural effects

Stable affiliative support should alter subjective distress and specific components of the stress-calibrated profile, including cortisol recovery, sleep, hippocampal reactivity, dopamine response, and flexible action selection. The effect should depend on whether the relationship is experienced as safe and predictable.

8.6 Environmental match will produce tradeoffs

Individuals calibrated by adversity may outperform nonexposed controls on selected tasks under danger, time pressure, uncertainty, or matched environmental conditions while performing worse under safety, ambiguity, or rapidly changing reward rules. Demonstrating this crossover interaction would provide stronger evidence of conditional adaptation than showing impairment alone.

8.7 Recovery failure will be a central transition

Many stress responses are reversible. Psychosis risk should increase when feedback mechanisms that normally terminate vigilance and reopen flexible control fail. Longitudinal work should distinguish activation, consolidation, and recovery. HDAC1, PNN stabilization, myelin remodeling, complement-mediated pruning, persistent sleep disruption, and social feedback are candidate mechanisms of consolidation.

8.8 Different symptom dimensions will map onto different components

Gating and salience changes should relate most directly to sensory flooding and unusual beliefs; habit bias and prefrontal dysfunction to disorganization and loss of agency; social-buffering failure to withdrawal and affective distress; and myelin or PV dysfunction to cognitive coordination. A diagnosis-wide average will conceal these relationships.

8.9 The graded-tail and full-phenotype models will separate under matched tests

The graded-tail model predicts peak ecological performance at moderate expression followed by declining performance near the diagnostic end of the spectrum. The full-phenotype model predicts at least some tasks, environments, historical roles, or crisis conditions in which end-spectrum participants equal or outperform moderate-expression groups. Relevant outcomes include weak-signal detection, remote association, inference under radical uncertainty, resistance to misleading consensus, night vigilance, solitary persistence, rapid threat action, and the production of socially useful narrative or ritual meaning. Longitudinal and cross-cultural studies should also test whether episodic schizophrenia-level states produce different fitness-relevant outcomes when communities provide recognized roles, protection, and interpretive support.

9. A comparative research program

The next step is to build the same measurement battery across species and connect each comparison to a defined mechanism.

A useful core battery would include PPI, auditory gating or MMN, reversal learning, outcome devaluation, probabilistic threat generalization, exploration, social preference, sleep architecture, cortisol reactivity and recovery, hippocampal activity, and dopamine release. Where possible, the battery should be paired with PV–PNN measures, oligodendrocyte lineage states, myelin mapping, microglial synaptic engulfment, complement expression, and HDAC1-related chromatin assays.

The design should compare at least four stress dimensions:

1. Timing: prenatal, juvenile, adolescent, and adult.
2. Pattern: acute, chronic, intermittent, and unpredictable.
3. Social context: isolation, defeat, caregiver disruption, and buffered stress.
4. Recovery: continued adversity, reunion, environmental enrichment, and restoration of control.

Rats and mice provide causal access to circuits and cell types. Guinea pigs add a different prenatal-developmental schedule and strong endocrine effects. Prairie voles test attachment and social buffering. Pigs contribute a large-brain social mammal with translationally useful physiology. Tree shrews provide a naturalistic psychosocial-stress model. Marmosets and macaques provide prolonged primate development, complex social organization, and close correspondence in prefrontal and sensory measures. Human cohorts provide subjective experience, clinical trajectories, imaging, electrophysiology, and direct tests of adversity by genetic liability.

The ideal program would use preregistered cross-species tasks and preserve both positive and null results. A null result after brief pig isolation helps identify the exposure necessary to engage the program. Likewise, improved response inhibition after manageable separation in squirrel monkeys demonstrates stress inoculation, in which a moderate, controllable challenge trains regulation and produces a different trajectory from chronic adversity.

The most informative experiments will test function as well as pathology. Animals should be evaluated in both safe and adverse test environments. If a stress-induced phenotype is a conditional adaptation, it should sometimes confer an advantage in the condition it predicts. For example, a habit-biased animal might perform worse when reward contingencies change but better when it must execute a well-learned escape response under distraction. An animal with lower gating might be more distractible in a benign arena yet faster to detect a weak predator cue. These crossover designs can identify the tradeoff directly.

The human arm should sample the complete schizophrenia spectrum, including people with current diagnoses and substantial symptom expression. It should measure capacities alongside difficulties and allow participants to identify the environments, roles, and forms of support in which they function best. This design is essential for testing the full-phenotype model. Restricting adaptive-performance tests to healthy people with mild schizotypal traits would predetermine the answer by excluding the phenotype under discussion.

10. Scientific and human implications if the hypothesis is true

If this hypothesis is even partly correct, it can help people and science by revealing the function, activation conditions, tradeoffs, and recovery rules of the affected systems. That knowledge changes what researchers measure, how clinicians divide the syndrome, which environments receive attention, and what counts as a successful outcome.

10.1 Schizophrenia would become biologically more intelligible

Gating, salience, threat learning, habit, social withdrawal, developmental timing, pruning, and myelination would form parts of an environmental-response architecture. Their co-occurrence would have a functional explanation. Researchers could study how the architecture activates, how its components coordinate, why coordination sometimes breaks down, and how the state resolves. This approach can join genetic, developmental, dopamine, glutamate, immune, endocrine, sleep, and social findings within a common causal model.

The comparative framework would also improve animal research. Each model could be evaluated by how well it reproduces a defined component and its ecological function. A rat isolation model could address gating and dopamine; prairie voles could address loss of social regulation; pigs could test developmental timing in a large social mammal; primates could address prolonged maturation and social interpretation. This division of labor would make cross-species findings cumulative.

10.2 Diagnosis could be divided by mechanism and trajectory

Two people can meet diagnostic criteria through different biological routes. One may have strong complement-related pruning and cognitive change, another marked hippocampal–dopamine reactivity after trauma, another severe sleep and inhibitory-circuit instability, and another a stable neurodivergent profile that becomes disabling mainly under social or sensory overload. Comparative profiles could identify these subgroups.

A useful clinical formulation would describe the person’s activation history, current state, recovery capacity, environmental fit, and strongest biological components. It could distinguish an acute salience crisis, a chronic threat-calibrated state, a developmentally stabilized cognitive style, and a progressive inflammatory or pruning process. These distinctions should lead to different research protocols and potentially different care.

10.3 Science would begin measuring capacities associated with the phenotype

Most schizophrenia research is designed to detect impairment. The full-phenotype hypothesis requires balanced measurement. Studies should examine weak-signal detection, remote association, unconventional problem framing, resistance to social conformity, sensitivity to interpersonal threat, persistence in solitude, crisis performance, narrative production, and the ability to construct meaning from ambiguous information. The expected result is a profile of tradeoffs, with performance varying by task and context.

This work could reveal useful capacities that standard neuropsychological batteries miss. It could also identify environments that convert a capacity into a disability. High sensory gain may assist detection in a quiet natural setting and overwhelm the same person in a crowded urban environment. Remote association may aid art, theory formation, tracking, or invention while complicating tightly constrained tasks. Reduced conformity may protect a group from collective error while increasing interpersonal conflict. Understanding these relationships would support better education, work design, community roles, and self-knowledge.

10.4 Risk detection could become earlier and more dynamic

The model directs attention to trajectories before a crisis: changing gating, rising stress-induced dopamine release, loss of model-based control, sleep fragmentation, social-buffering failure, and unusual changes in cortical maturation. A multimodal profile combining stress history, sleep, social context, electrophysiology, cognition, and imaging could detect a state transition earlier than symptom observation alone.

Dynamic markers could also indicate recovery. Cortisol normalization, restored sleep continuity, renewed goal-directed action, improved context discrimination, and reduced hippocampal hyperactivity may reveal movement away from an unstable threshold. Monitoring recovery biology could help determine when support should intensify, when ordinary demands can resume, and which intervention is actually changing the underlying state.

10.5 Care could target control, fit, and reversibility

If schizophrenia can involve a defensive or specialized state that has become rigid, treatment can aim to restore voluntary control over that state. Relevant targets include predictable daily structure, sleep stabilization, reduction of ongoing threat, safe social connection, cognitive work on context and uncertainty, rehabilitation of goal-directed action, and biological mechanisms that regulate plasticity.

A neurodiversity-informed plan would begin with the person’s own goals. One person may want relief from terror, voices, sleeplessness, or intrusive significance while valuing unusual associations and spiritual experience. Another may want the entire state quieted. Care can reduce danger and distress, improve choice, and preserve valued capacities. Medication can remain central for people who find it helpful, especially during acute positive symptoms. Environmental, social, cognitive, occupational, and sleep interventions address additional parts of the calibrated system.

Advance planning could also improve care. During stable periods, people could document which experiences they value, which signs indicate danger, who they trust, which environments calm or intensify the state, and what forms of intervention they prefer during a crisis. This turns individual experience into usable clinical information.

10.6 Social and environmental intervention would become mechanistic treatment

Housing instability, victimization, isolation, sensory overload, and chronic interpersonal threat are continuing inputs to the systems under study. Stable housing, protection from violence, supported employment, family education, access to nature, control over sensory intensity, and trustworthy clinical relationships can change the evidence the brain receives about its environment.

Social buffering deserves direct measurement in treatment trials. Researchers can test whether reliable relationships improve cortisol recovery, sleep, hippocampal activity, flexible action selection, and psychotic distress. Communities could also create respected roles for people with unusual perceptual and associative styles. Ecological fit may improve when a person has autonomy, protection, meaningful work, and an audience capable of interpreting unusual experience without immediate rejection.

10.7 Molecular targets could be organized by stage

HDAC1, complement, microglia, oxidative injury to PV interneurons, PNN regulation, oligodendrocyte maturation, hippocampal hyperactivity, and neurosteroid signaling may act at different stages. Complement modulation could matter most during an active pruning window; redox protection before prolonged PV deterioration; chromatin interventions while a stress program is consolidating; and plasticity-enhancing rehabilitation after safety has been restored.

This stage model could prevent a common translational error in which a biologically valid target is tested at the wrong time. Longitudinal biomarkers would help identify activation, consolidation, maintenance, and recovery. The molecular ideas discussed here remain research targets and require clinical trials before therapeutic use.

10.8 Prevention would acquire specific biological endpoints

Reducing childhood adversity is already justified on humanitarian grounds. The hypothesis adds measurable endpoints. Prevention studies could ask whether caregiver stability, reliable sleep, social buffering, control over exposure, and reduction of violence preserve gating, flexible control, inhibitory maturation, myelination, and stress recovery.

Prevention can also mean preventing mismatch and chronic escalation in a person who carries a schizophrenia-associated neurotype. Appropriate schooling, quieter housing, protection from bullying, flexible work, strong attachment relationships, and early sleep support may allow unusual cognition to develop constructively. This approach focuses on keeping a potentially valuable phenotype within a workable ecological and physiological range.

10.9 Neurodiversity could become experimentally informative

Neurodiversity supplies a research hypothesis as well as an ethical perspective. If schizophrenia-associated variation persists because it sometimes conferred advantages, then strengths should cluster systematically with vulnerabilities. The same variable that increases sensory flooding may improve weak-cue detection. The same associative looseness that increases false connections may increase remote or original connections. The same reduced social conformity that creates conflict may protect against collective error. The same high salience that generates fear may sustain extraordinary attention and commitment.

These paired predictions can be tested. They also encourage scientific participation by people with lived experience. Research priorities, task design, definitions of function, and outcome measures should include their observations. A person who has experienced the state may recognize capacities, triggers, recovery signals, and forms of environmental fit that an outside observer misses.

10.10 The hypothesis could reduce shame and improve social understanding

A calibrated-system account locates psychotic experience in understandable brain operations: detecting danger, assigning significance, building models, communicating meaning, and protecting the self. An evolutionary account adds the possibility that some people express a cognitive organization that was useful in ancestral groups and may retain value today.

This framing can replace a narrative of global defect with a more accurate account of capacities, tradeoffs, state changes, and support needs. It allows severe suffering to be taken seriously while treating the person as a bearer of a human neurotype with a history and possible functions. Families may gain a clearer explanation of why stress, sleep, safety, and social interaction matter. Clinicians may gain more precise targets. People with schizophrenia may gain language for understanding their experience and deciding which parts they want help changing.

10.11 Broader science would gain a model of conditional brain organization

Schizophrenia could become a leading case for studying how genes and environments jointly construct alternative brain states. The research would contribute to evolutionary biology, developmental plasticity, stress neuroscience, immunology, social neuroscience, and the study of consciousness. It could show how a conserved mammalian calibration system interacts with uniquely human language, culture, self-modeling, and collective meaning.

The larger scientific implication is that psychiatric syndromes may contain organized biological strategies whose contemporary costs obscure their original design. Testing that possibility can improve psychiatric classification across conditions, including depression, anxiety, trauma-related states, autism, and bipolar disorder. It directs science toward function, context, and tradeoff while retaining rigorous causal measurement.

11. Limits and safeguards

Precision gives the hypothesis its testable form.

Conservation supplies comparative support for adaptation. Predicted tradeoffs, environmental match, coordinated development, and improved performance in defined ecological conditions would provide stronger tests. Stress can also cause direct injury, metabolic cost, and system failure within the same organism. An evolved calibration process and downstream damage can coexist.

Animal models capture conserved components of the phenotype. PPI, social interaction, and amphetamine sensitivity should be named at that mechanistic level. The full human syndrome also involves language, autobiographical selfhood, culture, and the interpretation of unusual experience.

The term “maternal separation” also conceals very different procedures. A brief predictable separation followed by reunion is biologically different from fragmented care, unpredictable deprivation, or permanent social isolation. Comparative studies should describe the actual ecology of the manipulation.

Ethical use of the hypothesis requires protection from induced stress, treatment neglect, family blame, coercion, and romantic pressure to remain unwell. An evolved phenotype can be painful, disabling, dangerous, valuable, or several of these at once. The person’s own goals, safety, autonomy, and access to care remain primary.

Prehistoric function must be inferred from comparative biology, human behavioral ecology, formal tradeoff predictions, archaeological and ethnographic context, and convergent mechanisms. The graded-tail and full-phenotype models can be compared through the distinct predictions described above.

12. Conclusion

The comparative evidence supports a coherent evolutionary hypothesis. Across mammals, adversity can recalibrate sensory gating, salience, dopamine recruitment, decision systems, social behavior, developmental timing, inhibitory circuitry, synaptic selection, chromatin, and myelination. Schizophrenia involves many of the same systems. Their structured overlap makes it increasingly plausible that the syndrome is built from conserved stress-response mechanisms with an evolutionary history.

Mammalian brains contain evolved rules for becoming more vigilant, cautious, habitual, socially defensive, and developmentally stabilized when experience forecasts danger. In humans, those rules interact with language, culture, self-consciousness, prolonged development, and unusually complex social life. One evolutionary pathway could produce schizophrenia as the extreme tail of a graded calibration program. Another could produce schizophrenia-level organization as a rare integrated specialization favored under recurring conditions in the EEA. The available evidence justifies serious investigation of both possibilities.

This view expands the research question to include what the affected systems accomplish, what environmental information activates them, where their tradeoffs lie, which capacities appear at the diagnostic end of the spectrum, and what permits voluntary control or recovery. Those questions are experimentally tractable. They connect comparative neuroscience to prevention, early detection, mechanistic subtyping, environmental design, neurodiversity, and person-directed care. Schizophrenia may be a source of profound suffering in the present and may also preserve a powerful human specialization shaped by the deeper logic of mammalian adaptation.

References

Bahari-Javan, S., Varbanov, H., Halder, R., et al. (2017). HDAC1 links early life stress to schizophrenia-like phenotypes. Proceedings of the National Academy of Sciences, 114, E4686–E4694.

Bath, K. G., Manzano-Nieves, G., & Goodwill, H. (2016). Early life stress accelerates behavioral and neural maturation of the hippocampus in male mice. Hormones and Behavior, 82, 64–71.

Cadeddu, R., et al. (2022). Acute stress impairs sensorimotor gating via the neurosteroid allopregnanolone in the prefrontal cortex. Neurobiology of Stress, 21, 100489.

Choy, K. H. C., et al. (2009). The effect of “two-hit” neonatal and young-adult stress on dopaminergic modulation of prepulse inhibition and dopamine receptor density. British Journal of Pharmacology.

Dahoun, T., Nour, M. M., McCutcheon, R. A., Adams, R. A., Bloomfield, M. A. P., & Howes, O. D. (2019). The relationship between childhood trauma, dopamine release, and dexamphetamine-induced positive psychotic symptoms. Translational Psychiatry.

Dias-Ferreira, E., Sousa, J. C., Melo, I., et al. (2009). Chronic stress causes frontostriatal reorganization and affects decision-making. Science, 325, 621–625.

Enwright, J. F., Sanapala, S., Foglio, A., Berry, R., Fish, K. N., & Lewis, D. A. (2016). Reduced labeling of parvalbumin neurons and perineuronal nets in the dorsolateral prefrontal cortex of subjects with schizophrenia. Neuropsychopharmacology, 41, 2206–2214.

Feng, X., Wang, L., Yang, S., et al. (2011). Maternal separation produces lasting changes in cortisol and behavior in rhesus monkeys. Proceedings of the National Academy of Sciences, 108, 14312–14317.

Fuchs, E., Flügge, G., Ohl, F., Lucassen, P., Vollmann-Honsdorf, G. K., & Michaelis, T. (2001). Psychosocial stress, glucocorticoids, and structural alterations in the tree shrew hippocampus. Physiology & Behavior, 73, 285–291.

Gee, D. G., Gabard-Durnam, L. J., Flannery, J., et al. (2013). Early developmental emergence of human amygdala–prefrontal connectivity after maternal deprivation. Proceedings of the National Academy of Sciences, 110, 15638–15643.

Gil-da-Costa, R., et al. (2020). Nonhuman primate model of mismatch negativity at the scalp. Scientific Reports.

Giovanniello, J. R., et al. (2025). A dual-pathway architecture for stress to disrupt agency and promote habit. Nature, 640, 722–731.

Gracias, J., Orhan, F., Hörbeck, E., et al. (2022). Cerebrospinal fluid concentration of complement component 4A is increased in first-episode schizophrenia. Nature Communications, 13, 6427.

Harms, L., Fulham, W. R., Todd, J., et al. (2014). Mismatch negativity in freely moving rats with several experimental controls. PLOS ONE, 9, e110892.

Harms, L., et al. (2020). Mismatch negativity and deviance detection in schizophrenia. Schizophrenia Research.

Kanitz, E., et al. (2016). Social support modulates stress-related gene expression in various brain regions of piglets. Frontiers in Behavioral Neuroscience, 10, 227.

Kapoor, A., & Matthews, S. G. (2011). Testosterone is involved in mediating the effects of prenatal stress in male guinea pig offspring. Journal of Physiology, 589, 755–766.

King, D. P., Abdalaziz, M., Majewska, A. K., Cameron, J. L., & Fudge, J. L. (2025). Microglia morphology in the developing primate amygdala and effects of early-life stress. eNeuro, 12, ENEURO.0466-24.2024.

Kokkosis, A. G., Madeira, M. M., Mullahy, M. R., & Tsirka, S. E. (2022). Chronic stress disrupts the homeostasis and progeny progression of oligodendroglial lineage cells, associating immune oligodendrocytes with prefrontal cortex hypomyelination. Molecular Psychiatry, 27, 2833–2848.

Koenig, J. I., Elmer, G. I., Shepard, P. D., et al. (2005). Prenatal exposure to a repeated variable stress paradigm elicits behavioral and neuroendocrinological changes in the adult offspring: potential relevance to schizophrenia. Behavioural Brain Research, 156, 251–261.

Lehmann, M. L., et al. (2017). Chronic social defeat reduces myelination in the mouse medial prefrontal cortex. Scientific Reports.

Levine, S., & Mody, T. (2003). The long-term psychobiological consequences of intermittent postnatal separation in the squirrel monkey. Neuroscience & Biobehavioral Reviews, 27, 83–89.

Lind, N. M., et al. (2004). Prepulse inhibition of the acoustic startle reflex in pigs and disruption by amphetamine. Behavioural Brain Research.

Mauney, S. A., Pietersen, C. Y., Sonntag, K. C., & Woo, T. U. W. (2015). Differentiation of oligodendrocyte precursors is impaired in the prefrontal cortex in schizophrenia. Schizophrenia Research, 169, 374–380.

McHale-Matthews, A. C., DeCampo, D. M., Love, T., Cameron, J. L., & Fudge, J. L. (2023). Immature neurons in the primate amygdala: changes with early development and disrupted early environment. Developmental Cognitive Neuroscience, 61, 101248.

McNeal, N., Appleton, K. M., Johnson, A. K., et al. (2017). The protective effects of social bonding on behavioral and pituitary-adrenal axis reactivity to chronic mild stress in prairie voles. Stress, 20, 175–182.

Mizrahi, R., Addington, J., Rusjan, P. M., et al. (2012). Increased stress-induced dopamine release in psychosis. Biological Psychiatry, 71, 561–567.

Morris, R. W., Cyrzon, C., Green, M. J., Le Pelley, M. E., & Balleine, B. W. (2018). Impairments in action–outcome learning in schizophrenia. Translational Psychiatry, 8, 54.

Pan, Y., Liu, Y., Young, K. A., Zhang, Z., & Wang, Z. (2009). Post-weaning social isolation alters anxiety-related behavior and neurochemical gene expression in the brain of male prairie voles. Neuroscience Letters, 454, 67–71.

Parker, K. J., Buckmaster, C. L., Schatzberg, A. F., & Lyons, D. M. (2005). Mild early life stress enhances prefrontal-dependent response inhibition in monkeys. Biological Psychiatry, 57, 848–855.

Pesarico, A. P., et al. (2019). Chronic stress modulates interneuronal plasticity: effects on PSA-NCAM and perineuronal nets in cortical and extracortical regions. Frontiers in Cellular Neuroscience, 13, 197.

Poplawski, J., Montina, T., et al. (2024). Early life stress shifts critical periods and causes precocious visual cortex development. PLOS ONE, 19, e0316384.

Pryce, C. R., Dettling, A. C., Spengler, M., Schnell, C. R., & Feldon, J. (2004). Deprivation of parenting disrupts development of homeostatic and reward systems in marmoset monkey offspring. Biological Psychiatry, 56, 72–79.

Reser, J. E. (2007). Schizophrenia and phenotypic plasticity: schizophrenia may represent a predictive, adaptive response to severe environmental adversity that allows both bioenergetic thrift and a defensive behavioral strategy. Medical Hypotheses, 69, 383–394.

Reser, J. E. (2016). Chronic stress, cortical plasticity and neuroecology. Behavioural Processes, 129, 105–115.

Schiavone, S., et al. (2012). NADPH oxidase elevations in pyramidal neurons drive psychosocial stress-induced neuropathology. Translational Psychiatry, 2, e111.

Sekar, A., Bialas, A. R., de Rivera, H., et al. (2016). Schizophrenia risk from complex variation of complement component 4. Nature, 530, 177–183.

Sellgren, C. M., Gracias, J., Watmuff, B., et al. (2019). Increased synapse elimination by microglia in schizophrenia patient-derived models of synaptic pruning. Nature Neuroscience, 22, 374–385.

Soares, J. M., Sampaio, A., Ferreira, L. M., et al. (2012). Stress-induced changes in human decision-making are reversible. Translational Psychiatry, 2, e131.

Stevens, K. E., Johnson, R. G., & Rose, G. M. (1997). Rats reared in social isolation show schizophrenia-like changes in auditory gating. Pharmacology Biochemistry and Behavior, 58, 1031–1036.

Tillmon, H., et al. (2024). Complement and microglia activation mediate stress-induced synapse loss in layer 2/3 of the medial prefrontal cortex in male mice. Nature Communications, 15, 9803.

Tseng, H.-H., Watts, J. J., Kiang, M., et al. (2018). Nigral stress-induced dopamine release in clinical high risk and antipsychotic-naïve schizophrenia. Schizophrenia Bulletin, 44, 542–551.

Tuchscherer, M., et al. (2004). Effects of postnatal social isolation on hormonal and immune responses of pigs to an acute endotoxin challenge. Physiology & Behavior.

van der Staay, F. J., et al. (2015). Overnight social isolation in pigs decreases salivary cortisol but does not impair spatial learning and memory or performance in a decision-making task. Frontiers in Veterinary Science, 2, 81.

Valenti, O., Lodge, D. J., & Grace, A. A. (2011). Aversive stimuli alter ventral tegmental area dopamine neuron activity via a common action in the ventral hippocampus. Journal of Neuroscience, 31, 4280–4289.

Wilkinson, L. S., Killcross, S. S., Humby, T., Hall, F. S., Geyer, M. A., & Robbins, T. W. (1994). Social isolation in the rat produces developmentally specific deficits in prepulse inhibition of the acoustic startle response without disrupting latent inhibition. Neuropsychopharmacology, 10, 61–72.

Wood, E. K., et al. (2021). Early rearing conditions affect monoamine metabolite levels during baseline and periods of social separation stress: a nonhuman primate model. Frontiers in Human Neuroscience, 15, 624676.