THEORETICAL ARTICLE The Triple-Invisibility Model of Childhood Depression:Somatic Routing, Circuit Immaturity, and Rapid Recovery Across Typical Development and Neurodevelopmental Conditions

THEORETICAL ARTICLE

The Triple-Invisibility Model of Childhood Depression:

Somatic Routing, Circuit Immaturity, and Rapid Recovery Across Typical Development and Neurodevelopmental Conditions

――――――――――

Running head: TRIPLE-INVISIBILITY MODEL

Keywords: childhood depression, developmental psychopathology, somatic expression, neuroplasticity, evolutionary psychiatry, autism spectrum disorder, attachment disorder, developmental trauma, BDNF, HPA axis, default mode network, kindling

Word count: ~15,000 | Tables: 2 | Figures: 1 (conceptual)

Abstract

Depression in childhood has historically been under-recognised, misdiagnosed, and theorised as either absent or phenomenologically distinct from adult depression. We propose the Triple-Invisibility Model (TIM), a novel integrative framework that explains the low clinical visibility of childhood depression through three partially independent, developmentally regulated mechanisms: H1 — Somatic Routing, whereby immature interoceptive-cortical translation channels depressive neurobiological states into bodily rather than psychological expression; H2 — Circuit Immaturity, whereby the prefrontal-cortical, ruminative, and self-schema circuits required for adult-form depression are structurally incomplete in early childhood; and H3 — Rapid Recovery, whereby elevated neuroplasticity, abundant slow-wave sleep, and high brain-derived neurotrophic factor (BDNF) allow depressive states to resolve within days, preventing accumulation into clinically-threshold episodes. We map the progressive weakening of all three mechanisms across five developmental stages — early childhood, middle childhood, early adolescence, late adolescence, and adulthood — demonstrating that the well-documented epidemiological rise in depression prevalence at puberty reflects the near-simultaneous collapse of all three protective mechanisms rather than a simple increase in stressor exposure. We then apply the model as a comparative framework to autism spectrum disorder (ASD), attachment disorder, and childhood trauma, demonstrating that each condition produces a qualitatively distinct depression architecture by disrupting, locking, or inverting the three mechanisms through constitutional, relational, and experiential routes respectively. The model generates testable predictions — including the existence of clinically undetected sub-threshold depressive micro-episodes in early childhood, a pharmacological parallel between natural childhood recovery mechanisms and antidepressant action, and the critical importance of sleep architecture in both the aetiology and treatment of depression in neurodevelopmental populations. Implications for assessment, diagnosis, and intervention are discussed.

1. Introduction

The proposition that depression does not occur in children once dominated clinical thinking. Rooted in psychoanalytic theory — specifically in the premise that a fully developed superego is a prerequisite for self-directed hostility — this view persisted in various forms until the introduction of explicit diagnostic criteria for childhood depression in DSM-III (1980). Yet even after formal diagnostic recognition, rates of detected childhood depression remained strikingly low, clinicians continued to misattribute its presentations to other conditions, and the theoretical literature struggled to explain why depression, so prevalent in adults, was so rarely identified before puberty.

Three explanatory positions have historically competed. The first holds that children are genuinely incapable of depression by virtue of cognitive and neural immaturity: without abstract self-representations, temporal reasoning about a hopeless future, or the ruminative cognitive apparatus that characterises adult melancholia, the relevant psychological conditions for depression are simply not met. The second holds that children do become depressed but express it differently — through somatic complaints, behavioural disturbance, and irritability rather than classical dysphoria — the position enshrined in the DSM-5 note that irritability may substitute for depressed mood in children. The third, most clinically disruptive position, is that childhood depression exists, is expressed in recognisable ways when properly assessed, and is simply missed by clinical systems trained on adult phenomenology.

The present paper argues that all three positions capture partial truths and that their apparent conflict can be resolved within a single integrative framework. We propose that childhood depression is neither simply absent nor simply misrecognised, but is regulated by three overlapping biological and developmental mechanisms that suppress its clinical visibility, duration, and cognitive elaboration simultaneously. As these mechanisms weaken across development, depression becomes progressively more visible, more persistent, and more cognitively organised — producing the observed epidemiological curves that show sharp increases in prevalence at adolescence with no discontinuous change in underlying vulnerability.

The framework we develop — the Triple-Invisibility Model (TIM) — draws on evolutionary psychiatry, developmental neuroscience, interoceptive neuroscience, sleep science, and comparative psychopathology. We first describe each mechanism and its developmental trajectory in typical development. We then map the framework onto four populations — typical development, autism spectrum disorder, attachment disorder, and trauma — demonstrating that each produces a phenomenologically and mechanistically distinct form of childhood depression with its own clinical signature and its own reasons for remaining undetected. We conclude with theoretical propositions, testable predictions, and clinical implications.

2. Theoretical Background

2.1 Evolutionary Foundations

To understand why depression can occur in childhood brains that lack full cortical maturation, it is necessary to situate depression within its evolutionary context. Contemporary evolutionary psychiatry recognises that the neurobiological substrates of depression are phylogenetically ancient — predating the mammalian neocortex by hundreds of millions of years. Engel and Schmale’s (1972) conservation-withdrawal hypothesis identified the depressive response — inactivity, reduced appetite, social withdrawal, metabolic conservation — as a fundamental vertebrate survival strategy activated when escape from threat is impossible. This response is observable in reptiles, amphibians, and fish, and involves the phylogenetically oldest component of the autonomic nervous system: the dorsal vagal complex, later theorised by Porges (1995, 2011) in his Polyvagal framework as the neural substrate of dissociative shutdown in humans.

The next evolutionary layer — present in all mammals — is the PANIC/GRIEF system described by Panksepp (2011), organised around the periaqueductal gray, anterior cingulate, and bed nucleus of the stria terminalis. This system generates separation distress — the protest-despair sequence documented by Harlow and Zimmermann (1959) in primate maternal-separation paradigms — and constitutes what may be the most direct evolutionary antecedent of human depression. Crucially, this system is functional from birth in all mammals, including human infants. Panksepp and Watt (2011) argued persuasively that the PANIC/GRIEF circuit generates genuine affective suffering independent of cortical elaboration, which has direct implications for our framework: subcortical depressive suffering is not a developmental achievement but a mammalian birthright.

The social competition and subordination hypothesis (Price et al., 1994; Gilbert & Allan, 1998) adds a further layer: depression in hierarchical social vertebrates may have been selected as an involuntary subordinate strategy — a set of appeasement behaviours (reduced activity, averted gaze, withdrawal, diminished libido) that signal submission to dominant conspecifics, reducing escalating conflict. The monoaminergic correlates of this system — reduced serotonergic tone and elevated HPA reactivity in subordinate animals — are precisely those observed in adult human depression. These systems are present and functional in childhood, making children constitutionally capable of at least the subcortical, evolutionarily ancient core of depression.

The implication is critical: the relevant question is not whether children can experience depression at a neurobiological level — they clearly can — but rather what developmental mechanisms buffer the expression, duration, and cognitive elaboration of that experience, and how those buffers change over time.

2.2 The Developmental Epidemiology of Depression

Epidemiological data provide the scaffolding against which any developmental model must be assessed. Lifetime prevalence estimates for major depressive disorder (MDD) in early childhood (ages 2–7) range from approximately 1–2%, rising to 2–3% in middle childhood (ages 7–11), to 5–8% in early adolescence (ages 11–14), and reaching 8–15% by late adolescence (Costello et al., 2006). The sharp inflection at puberty is among the most robust findings in developmental psychopathology. Of comparable importance is the emergence of a gender gap at this inflection point: pre-pubertal rates are approximately equal between boys and girls; post-puberty, girls show prevalence roughly twice that of boys, a difference robustly attributed to differential ruminative response styles by Nolen-Hoeksema and Girgus (1994).

These data patterns are consistent with a model in which biological protections against depression weaken progressively, with an especially rapid collapse at puberty, rather than with models that posit a categorical qualitative change in the nature of depression at adolescence. Critically, the low rates in early childhood may substantially underestimate true occurrence, since standard diagnostic instruments were designed for adult presentations and fail to capture the somatic, brief, and behaviourally expressed episodes we argue are the primary childhood form.

2.3 Prior Explanatory Frameworks

Several prior frameworks have anticipated elements of our synthesis. The masked depression concept (Glaser, 1967; Cytryn & McKnew, 1972) proposed that children experience depression expressed through behavioural ‘masks’ — aggression, school refusal, somatic complaints — rather than classical dysphoria. Though this concept has fallen out of favour due to imprecision, it correctly identified that childhood depression does not present as a miniaturised version of adult depression. Kovacs and Beck (1977) demonstrated empirically that depressive cognitions exist in children as young as 8–9, challenging the view that cognitive prerequisites for depression are entirely absent in childhood. More recently, Luby et al. (2003, 2009) provided the strongest evidence that MDD can occur in preschool children aged 3–5 years, albeit with age-modified symptom profiles dominated by anhedonia, psychomotor changes, and somatic features rather than verbal expression of hopelessness.

Our framework integrates these observations and provides the mechanistic explanation they lack: why does childhood depression express somatically? Why do even genuine depressive states in young children tend to be brief? And why does the transition to adolescence produce such a rapid change in the form, prevalence, and persistence of depression? The TIM answers each question by mapping specific developmental mechanisms.

3. The Triple-Invisibility Model

The Triple-Invisibility Model proposes that three partially independent, developmentally regulated mechanisms operate simultaneously in early childhood, each suppressing a different dimension of depression’s clinical visibility. The mechanisms interact such that the weakening of one accelerates the weakening of the others, producing the sharp epidemiological inflection observed at puberty. Each mechanism has a distinct neurobiological substrate, a distinct developmental trajectory, and distinct clinical implications.

3.1 H1: Somatic Routing

The first mechanism concerns the channel through which depressive neurobiological states are expressed. In adults, depression involves both neurobiological dysregulation (HPA hyperactivation, monoamine depletion, reward circuit suppression) and conscious psychological suffering — the subjective experience of sadness, worthlessness, and hopelessness that constitutes the phenomenological core of the disorder. The translation of neurobiological depressive states into conscious emotional experience depends critically on the anterior insula and its connections to the prefrontal cortex: the interoceptive-cortical pathway that maps bodily physiological states onto conscious feeling-states, described by Craig (2009) as the substrate of human emotional awareness.

This pathway undergoes protracted postnatal development, with the anterior insular-prefrontal connectivity continuing to mature through adolescence. In early childhood, this pathway is functionally immature: children cannot reliably translate internal bodily states into labelled emotional experiences. This is not mere emotional immaturity in the colloquial sense — it is a specific developmental deficit in what Barrett and colleagues have called emotional granularity: the capacity to generate distinct, differentiated conscious emotional representations from undifferentiated interoceptive signals.

The practical consequence is that children function, by developmental default, in a state of partial alexithymia — not because emotions are absent but because the cortical machinery for converting them into conscious, labelled experience is not yet operational. Depressive neurobiological states — elevated cortisol, suppressed mesolimbic dopamine signalling, activation of the PANIC/GRIEF circuit — cannot be fully translated into the psychological channel and are therefore expressed primarily through their somatic correlates: gastrointestinal distress (mediated by CRH receptors densely expressed in the gut), headache, fatigue, sleep disruption, appetite change, and psychomotor agitation or retardation. The child presents as physically unwell, behaviourally dysregulated, or irritable — not as subjectively depressed.

A striking cultural parallel supports the developmental account: somatic expression of depression is the dominant presentation in cultural contexts where psychological introspection is not culturally privileged — notably in Chinese and several sub-Saharan African contexts (Kleinman, 1982; Ryder et al., 2008). The convergence between cultural and developmental routes to somatic expression suggests that the psychological channel is not a biological given but a developmental and cultural achievement requiring specific scaffolding. Children, like members of less psychologised cultural environments, use the somatic channel as the default — requiring the development of interoceptive awareness, emotional vocabulary, and reflective capacity to override it.

Somatic routing produces clinical invisibility through a dual mechanism: the expression channel is wrong for adult-trained clinicians (who expect dysphoric mood as the primary complaint), and the attributional channel is wrong for caregivers (who bring sick children to paediatricians, not psychiatrists). The result is that the depressive state is expressed, but expressed in a language that clinical systems were not designed to read as depression.

3.2 H2: Circuit Immaturity

The second mechanism concerns the architecture of the depressive state itself. Adult MDD is characterised not only by subcortical neurobiological features (HPA dysregulation, monoamine depletion) but by a specific pattern of cortical circuit engagement that gives the disorder its distinctive phenomenological character: the self-sustaining ruminative loops, the stable negative self-schema, the prospective hopelessness about the future, and the chronic activation of the subgenual anterior cingulate cortex (sgACC, Brodmann Area 25) that is so reliably implicated in adult MDD that it has become a target for deep brain stimulation in treatment-resistant cases.

These cortical features depend on neural circuits that are structurally immature in early childhood. The default mode network (DMN) — the large-scale brain network mediating self-referential thought and the primary substrate of rumination — requires extensive myelination of long-range prefrontal-posterior connections for its adult-typical synchronisation. This myelination continues into the mid-20s. The DMN’s capacity for the sustained, self-amplifying negative self-referential processing that characterises adult rumination — the cognitive engine of depressive chronicity — is therefore not available in early childhood in any functional sense.

The negative self-schema — a stable, abstract, generalised cognitive structure representing the self as fundamentally defective, unlovable, or worthless — is similarly developmental. Piagetian research establishes that stable abstract trait-based self-descriptions do not emerge reliably until middle childhood (around age 7–8), when children transition from situational self-descriptions (‘I was bad today’) to trait-level attributions (‘I am a bad person’). Before this transition, the stable, globally negative self-representation central to Beck’s cognitive model of depression simply cannot form, because the cognitive architecture to support it — abstract, cross-situational, temporally extended self-representation — is not yet available.

The pharmacological evidence provides indirect but compelling support. Bridge et al.’s (2007) meta-analysis of antidepressant trials demonstrated that SSRI efficacy increases with age — very low in young children, modest in adolescents, robust in adults. If childhood depression were simply adult depression running on a smaller brain, SSRIs should work equivalently across ages. The age gradient in efficacy instead suggests that the serotonergic architectural substrate underlying adult MDD is genuinely less complete in younger children, and that the full adult depressive circuit being targeted by SSRIs is not yet instantiated in the same form.

An important qualification is required: circuit immaturity does not render children incapable of any depressive state — it renders them incapable of the distinctively adult, cortically-elaborated form of depression. Subcortical depressive states (separation distress, behavioural despair, basic anhedonia) are fully available from early infancy, as both primate separation studies and Luby’s preschool MDD data demonstrate. The H2 mechanism protects against the chronic, ruminatively-maintained, cognitively-elaborated form of depression while leaving the more primitive subcortical form accessible.

3.3 H3: Rapid Recovery

The third mechanism concerns the duration of depressive states. Even if a depressive episode instantiates in a child — through the subcortical circuits that are functional from birth — its persistence depends on the balance between the neurobiological forces maintaining it and the neuroplastic and sleep-based forces resolving it. In early childhood, this balance is heavily tilted toward resolution.

Two neurobiological systems are central. First, BDNF and neuroplasticity: brain-derived neurotrophic factor, the primary mediator of synaptic plasticity and the molecule whose depletion is among the most replicated findings in depression neuroscience (and whose restoration is increasingly understood as a primary mechanism of antidepressant action), is naturally elevated in early childhood — reflecting the period of maximum synaptic density and neural remodelling. The capacity to restore and reorganise synaptic connectivity after stress is therefore at its developmental peak, providing a natural antidepressant action.

Second, and more specifically, sleep architecture: children in early childhood sleep longer (10–12 hours per night) and spend a substantially greater proportion of this time in slow-wave sleep (SWS) than at any later developmental stage. Tononi and Cirelli’s (2006) synaptic homeostasis hypothesis proposes that SWS serves the function of downscaling synaptic potentiation accumulated during waking — essentially resetting the neural state toward baseline by pruning non-essential connections. Walker and van der Helm (2009) further demonstrated that REM sleep serves an emotional memory processing function, ‘stripping’ the affective charge from distressing memories by re-encoding them under conditions of suppressed norepinephrine, reducing their emotional valence. The combination of abundant SWS and REM sleep in early childhood provides a powerful overnight emotional reset mechanism that adults — with shorter sleep, reduced SWS proportion, and more disrupted architecture — simply do not have access to at equivalent strength.

The consequence for depression is that even genuine depressive neurobiological states — a day of separation distress, an episode of behavioural despair following a frightening experience — are regularly cleared by the overnight reset mechanism. The child wakes neurochemically reorganised. Post (1992) has shown, in his kindling hypothesis, that depressive episodes sensitise the circuit, making subsequent episodes more easily triggered and more self-sustaining. If childhood episodes are consistently brief — extinguishing before the kindling mechanism can accumulate load — each individual episode also fails to leave the structural trace (altered glucocorticoid receptor sensitivity, hippocampal volume changes, epigenetic modifications) that increases vulnerability to subsequent episodes. The protective effect of H3 is thus both immediate (clearing the current episode) and longitudinal (preventing the sensitisation that makes future episodes more severe).

Together, the three mechanisms create a nested protective architecture: H1 ensures that depressive states are expressed in channels that clinicians and caregivers do not interpret as depression; H2 ensures that the full adult form of depression cannot be constructed by an immature circuit; and H3 ensures that even the subcortical form that can occur resolves rapidly enough to remain below clinical detection thresholds. The model predicts that childhood depression is not rare but clinically invisible — and that it becomes visible precisely and progressively as each of these mechanisms weakens.

4. Developmental Trajectory of the Three Mechanisms

4.1 Early Childhood (Ages 2–7): Triple Protection at Maximum Strength

In early childhood, all three mechanisms operate at maximum strength simultaneously. H1 is maximally engaged: the anterior insular-prefrontal connectivity required for interoceptive-to-emotional translation is at its most immature, and caregiver affect-attunement — the primary developmental input that gradually builds the psychological channel — has had minimal time to operate. Depressive neurobiological states are almost exclusively expressed through somatic channels. The child presents with recurrent stomachaches, fatigue, appetite disruption, sleep resistance, irritability, and behavioural regression rather than articulated sadness or hopelessness.

H2 is maximally protective: without a functional DMN capable of adult-typical synchronisation, without stable abstract self-representations, and without the sgACC connectivity central to adult MDD, the full adult depressive circuit cannot instantiate. Depression at this stage is structurally subcortical — closer in architecture to the protest-despair response of an infant rhesus macaque than to an adult human MDD episode. H3 is maximally operational: BDNF is at peak developmental levels, SWS comprises a large proportion of abundant total sleep time, and caregiver physical co-regulation provides external HPA down-regulation after stress through what Coan and Sbarra’s (2015) social baseline theory frames as a neurobiological resource evolution designed the human nervous system to depend upon.

The epidemiological prevalence of approximately 1–2% at this stage almost certainly underestimates true occurrence; it represents only the most severe and persistent cases where chronic environmental adversity — typically severe neglect or abuse — overwhelms even the maximal protective capacity. The vast majority of depressive states in early childhood are invisible not because they do not occur, but because they are expressed somatically, resolve within hours to days through H3, and are interpreted by caregivers and clinicians as physical illness or temperamental difficulty.

4.2 Middle Childhood (Ages 7–11): Partial and Differential Weakening

Middle childhood represents a transitional period in which the three mechanisms begin to weaken at different rates, producing the first qualitative shift in the form of depression without yet generating dramatic prevalence increases. H1 begins to weaken as the concrete operational stage (in Piagetian terms) enables children to think logically about concrete present-state information, including information about their own emotional states. Expanding social environments — classrooms, peer groups, organised activities — demand and scaffold emotional literacy. An 8-year-old can say ‘I feel sad because my friend excluded me’ in a way that a 4-year-old cannot organise, representing a partial opening of the psychological channel. Somatic expression remains predominant but is increasingly accompanied by expressible emotional content.

H2 weakens as stable trait-based self-descriptions emerge around age 7–8. Abela and colleagues have demonstrated that ruminative response styles — a trait-level tendency to focus on and amplify negative emotional states rather than distract from them — are measurable and predictive of depression from approximately age 8–9 onward, suggesting that the DMN and associated prefrontal circuits are beginning to support rudimentary rumination. The first negative self-schemas form around academic failure, peer rejection, and family conflict — but these are still relatively concrete and situation-specific compared to the abstract globalised self-representations of adolescent and adult depression. H3 weakens modestly: total sleep decreases (from 11 to 9–10 hours), and the emergence of early rumination creates a counterforce to the sleep-based reset — depressive states cleared overnight can be partially re-activated by ruminative processing during the following day.

The net result is a clinical picture that is more psychologically recognisable than early childhood presentations but still substantially mixed with somatic features, and episodes that last days to approximately one week rather than the hours of early childhood. Prevalence rises modestly to approximately 2–3%, and the first kindsomething begins to accumulate: early depressive episodes at this stage may leave small but measurable traces that modestly increase future vulnerability.

4.3 Early Adolescence (Ages 11–14): The Critical Collapse

This is the developmental window the model identifies as the point of most rapid change, and it is here that the epidemiological inflection — from approximately 3–4% to 5–8% prevalence, with the emergence of the gender gap — is concentrated. All three mechanisms collapse in a compressed window, driven largely by a single biological event: puberty.

H2 collapses most dramatically. Gonadal hormone surges at puberty act directly on monoaminergic neurotransmission, altering the gain and homeostatic set-points of serotonergic and dopaminergic systems — the precise systems implicated in MDD. The well-documented prefrontal-limbic imbalance of early adolescence — where pubertal acceleration of subcortical limbic development outpaces the more gradual trajectory of prefrontal cortical maturation (Casey et al., 2008) — creates a period of enhanced emotional reactivity combined with reduced regulatory capacity. Simultaneously, DMN long-range connectivity achieves a functional threshold that supports adult-typical self-referential processing: the full ruminative machinery becomes available for the first time. The sgACC achieves sufficient connectivity to serve as the hub of the adult depressive network. What was previously architecturally impossible is now possible: the full cortical form of adult MDD can instantiate.

H1 weakens substantially as prefrontal-insular connectivity crosses a functional threshold allowing consistent translation of bodily states into conscious emotional experience. The adolescent can now experience and articulate — at least to themselves — conscious emotional pain: loneliness, shame, worthlessness. For the first time, the internal phenomenology of depression begins to approximate the adult form, with subjective suffering becoming the primary channel rather than an accompaniment to somatic expression. This shift is paradoxical: depression simultaneously becomes more visible (adolescents can report emotional states) and more subjectively painful (the psychological channel adds conscious suffering to the somatic substrate).

H3 is substantially compromised by puberty through a specific mechanism: the well-documented circadian phase shift that delays sleep onset by 1–2 hours while social and academic schedules prevent compensatory later rising, producing chronic sleep restriction of 1–3 hours per night relative to biological need in the majority of adolescents. SWS — the stage most critical to the synaptic homeostasis function of H3 — is the stage most vulnerable to restriction-related loss. The overnight emotional reset mechanism, already weakened by growing rumination, is further compromised by insufficient and architecturally disrupted sleep at precisely the developmental moment when the circuit it must reset has become adult-complex.

4.4 Late Adolescence and Adulthood: Stable Adult Configuration

By late adolescence, all three protective mechanisms have reached their adult baseline. H1 is minimal in typical development — depression is primarily expressed through the psychological channel, with somatic features present but secondary. H2 is effectively absent — the full adult depressive circuit is operational, though prefrontal regulatory capacity continues strengthening into the mid-20s, explaining the peak incidence of first-onset MDD in the 18–25 age range as the moment when the depressogenic circuit is fully mature while regulatory maturation is still completing. H3 operates at adult baseline — BDNF is no longer at developmental-peak levels, sleep is shortened and constrained, rumination is an established habitual response style, and kindling has had time to accumulate from earlier episodes, sensitising the circuit and progressively lowering the stress threshold required to trigger subsequent episodes.

The SSRI efficacy gradient that maps across these developmental stages — from very low in early childhood to robust in adulthood — is consistent with the TIM’s account: antidepressants may achieve their effects partly by pharmacologically restoring what natural mechanisms once provided — BDNF upregulation (mimicking the naturally elevated BDNF of childhood), synaptic plasticity enhancement, and HPA normalisation. The weeks-long delay in antidepressant efficacy, notoriously inconsistent with their immediate neurochemical actions, corresponds approximately to the time required to pharmacologically rebuild what the developing brain once reset nightly.

Table 1.

Table 1. Developmental Trajectory of the Triple-Invisibility Model Mechanisms Across Five Developmental Stages

Dimension	Early Childhood 2–7 yrs	Middle Childhood 7–11 yrs	Early Adolescence 11–14 yrs	Late Adolescence 16–25 yrs	Adulthood 25+ yrs
H1: Somatic Routing (protection level)	Maximum	High	Moderate	Low	Minimal
H2: Circuit Immaturity (protection level)	Maximum	High	Low — Puberty Collapse	Minimal	Absent
H3: Rapid Recovery (protection level)	Maximum	High	Substantially Impaired	Low	Baseline
Primary Presentation	Somatic; behavioural; irritable	Predominantly somatic; emerging psychological	Psychological + somatic; anhedonic	Full adult phenomenology emerging	Full adult phenomenology; vegetative features
Typical Episode Duration	Hours–days	Days–~1 week	Days–weeks	Weeks	Weeks–months
MDD Prevalence (approximate)	1–2%	2–3%	5–8%	8–15%	7–12% (current)
SSRI Efficacy	Very low	Low–moderate	Moderate	Moderate–high	Full
Kindling Accumulation	None	Minimal	Beginning	Moderate	Significant

5. Application to Neurodevelopmental and Experiential Conditions

Autism spectrum disorder, attachment disorder, and developmental trauma each function as a natural experiment testing the TIM by selectively disrupting, locking, or inverting its mechanisms. None produces simple ‘more depression earlier’ — each produces depression of a qualitatively distinct architecture, with its own clinical signature and its own mechanisms of invisibility. The key distinction is between three routes of scaffold failure: constitutional (autism), relational (attachment disorder), and acute experiential (trauma).

5.1 Autism Spectrum Disorder: Constitutional Reorganisation

Autism modifies the TIM from within — through constitutional differences in social-emotional neurobiology that alter the default settings of H1 and H3 without producing the same gradual developmental weakening that characterises typical development. Approximately 50% of autistic individuals meet criteria for alexithymia (Garfinkel et al., 2016), and this alexithymia is not merely a social communication artefact: it reflects atypical anterior insular organisation and genuine differences in interoceptive accuracy — the ability to detect and represent internal bodily signals. For these individuals, the somatic routing that H1 represents is not a developmental default awaiting override but a constitutional feature of nervous system organisation. The psychological channel that typically opens through caregiver attunement and accumulating emotional vocabulary may never open to the same degree. Somatic routing persists across the lifespan as the primary expression channel for depression, rather than being gradually replaced.

H2 in autism is not simply delayed — it is differently organised. Social pain circuits may process rejection through atypical pathways; the monotropic reward system (Murray, Lesser, & Lawson, 2005) creates domain-specific anhedonia when engagement with special interests is curtailed; and negative self-schemas may form early around autistic identity. Most significantly, the phenomenon of camouflaging or masking — the effortful suppression of autistic traits to appear neurotypical — creates a form of depressive vulnerability that has no equivalent in the TIM’s typical development trajectory. Masking requires continuous, unrelenting performance of an identity that is neurologically incongruent with the person’s actual processing style. In terms of the social competition hypothesis, this constitutes an inescapable, lifelong involuntary subordinate strategy — precisely the condition Gilbert and Allan (1998) identified as generating the most treatment-resistant forms of depression. Cassidy et al. (2019) and Cage and Troxell-Whitman (2019) have both demonstrated that masking is among the strongest predictors of depression and suicidality in autistic individuals, and that it operates through a mechanism — chronic social defeat — that is qualitatively distinct from the stressor-reactive depression of typical development.

H3 in autism is structurally compromised from birth. Sleep onset insomnia affects 50–80% of autistic children, melatonin synthesis and circadian synchronisation are atypical, slow-wave sleep is reduced in polysomnographic studies, and sensory sensitivities create sleep-hostile environments. The consequence is that autistic children do not have the full protective benefit of H3 even in early childhood — the period when it is at maximum strength in typical development. The overnight emotional reset mechanism is attenuated from the very start of life. Depression in autism therefore tends to be chronic rather than episodic, somatically expressed across the lifespan, clinically invisible behind the autism diagnosis (diagnostic overshadowing), and treatment-resistant through CBT approaches that assume the psychological channel is open. Lifetime depression prevalence in autistic individuals is estimated at 40–70% (Hollocks et al., 2019) — almost certainly an underestimate given systematic under-detection.

5.2 Attachment Disorder: Relational Erosion of the Scaffold

All three TIM mechanisms depend in part on the external relational scaffold provided by the caregiving relationship: H1 is built through caregiver affect attunement and emotional labelling (Fonagy et al.’s mentalization); H2 maturation is buffered by the caregiver’s HPA-regulatory function keeping cortisol within non-sensitising ranges; and H3 depends directly on caregiver physical co-regulation as the primary mechanism for restoring the parasympathetic baseline that enables SWS. Attachment disorder — whether reactive (RAD) or disinhibited (DSED), or the broader spectrum of insecure and disorganised attachment — represents the failure or distortion of this scaffold, producing consequences for all three mechanisms simultaneously.

H1 is locked in the high-somatic state not constitutionally (as in autism) but relationally — by the absence of the developmental input that was supposed to weaken it. Fonagy et al.’s (2002) mentalization framework is directly relevant: children who develop in environments where caregivers do not treat them as minded beings develop impaired reflective function — exactly the capacity required to translate somatic states into psychological awareness. The psychological channel never receives the developmental scaffolding needed to open. H2 is not merely immature but epigenetically sensitised: Meaney’s (2001) foundational work on maternal care and glucocorticoid receptor methylation in rats, extended to human populations by McGowan et al. (2009), demonstrates that the quality of early caregiving permanently alters HPA axis reactivity through epigenetic modification of the NR3C1 gene in the hippocampus. The circuit matures not into a normal adult configuration but into a stress-sensitised configuration resembling the post-kindling state of a repeatedly depressed adult — achieved through epigenetic programming rather than through the accumulation of depressive episodes.

H3 is chronically eroded by the absence of the primary mechanism evolution designed for its maintenance: caregiver co-regulation. Coan and Sbarra’s (2015) social baseline theory proposes that human nervous systems evolved to treat the proximity of trusted others as a neurobiological resource that reduces the metabolic cost of stress regulation. Children with attachment disorder must regulate their HPA axis without this resource, generating chronically elevated cortisol that suppresses BDNF, prevents SWS, and steadily depletes the neuroplastic substrate of H3. The depression that emerges from this configuration is chronologically displaced — resembling an adolescent or adult depression profile in children as young as toddlerhood in severe cases. Van der Kolk’s (2005) proposed category of developmental trauma disorder captures many of these presentations, but the TIM explains the mechanism: it is not simply that these children have experienced trauma, but that the relational deprivation has dismantled the protective architecture from the earliest weeks and months of life, leaving a nervous system without the tools to modulate its own distress.

5.3 Trauma: Acute Assault and the Inversion of H3

Trauma — particularly complex or Type II trauma involving repeated, inescapable events in early childhood — represents the most complex modification of the TIM because it simultaneously disrupts all three mechanisms and, in the case of H3, actively inverts it from a protective into a destructive force. H1 is altered in a paradoxical direction: somatic hyperexpression and dissociative hypoexpression coexist simultaneously. Traumatic memory is encoded not in the verbal-declarative hippocampal systems that require cortical consolidation but in implicit, procedural, and visceral memory systems — the body ‘keeps the score’ (van der Kolk, 2014) in the form of muscle tension, autonomic dysregulation, visceral sensations, and sensorimotor fragments. The body is saturated with the somatic residue of depression. Simultaneously, severe trauma activates dissociative mechanisms — mediated by Porges’ dorsal vagal shutdown system — that sever the connection between experience and consciousness entirely. The person is flooded with somatic signal and disconnected from it at the same time: depression expressed at maximum intensity through a channel the person has been neurobiologically cut off from accessing.

H2 is bypassed not through epigenetic sensitisation (as in attachment disorder) but through threat-driven precocious circuit activation. Perry et al.’s (1995) sequential trauma model documents that repeated early trauma accelerates the maturation of brainstem and limbic threat-response circuits while higher cortical development proceeds at its typical rate — producing a developmentally inverted architecture in which the accelerator is adult-strength while the regulatory brake remains immature. A single severe traumatic event can produce kindling-equivalent circuit sensitisation in compressed time (Post, 1992), meaning a traumatised 5-year-old may have a depressive circuit with the sensitisation profile of a repeatedly depressed adult. Shame-based negative self-schemas — ‘I caused this’, ‘I deserved this’, ‘I am permanently damaged’ — form precociously through traumatic meaning-making (Feiring & Taska, 2005), bypassing the developmental timeline that normally delays abstract negative self-representation until middle childhood.

H3 is not merely weakened in trauma — it is actively inverted into a re-traumatisation mechanism. Walker and van der Helm’s (2009) model of REM sleep as emotional memory processing specifies that affect-stripping requires conditions of suppressed norepinephrine during REM — a neurochemical state that allows traumatic memories to be re-encoded with reduced emotional charge. In post-traumatic states, norepinephrine remains elevated during REM sleep, preventing the affect-stripping mechanism from operating. Traumatic memories do not lose their emotional charge during sleep — they re-activate as nightmares at full intensity. Sleep, which was the primary recovery mechanism, becomes a site of nightly re-traumatisation. The loss of H3 function generates further HPA dysregulation, BDNF suppression, and sleep disruption — creating a positive feedback loop in which the destruction of the recovery mechanism accelerates the consolidation of the depressive state it was supposed to clear. Treatment approaches that target sleep architecture directly — including prazosin for nightmare reduction and EMDR, which may work partly by simulating the emotional processing normally performed during healthy REM — can be understood as directly addressing the inverted H3 mechanism.

Table 2.

Table 2. Comparative Profile of TIM Mechanisms Across Four Populations

Mechanism	Typical Development	Autism Spectrum	Attachment Disorder	Trauma
H1: Somatic Routing	Default in childhood; progressively weakened by caregiver attunement and emotional vocabulary development	Constitutionally maintained across lifespan; alexithymia is structural, not transitional	Relationally locked; psychological channel never receives scaffolding from absent or harmful caregiver	Paradoxical: somatic hyperexpression (body memory) + dissociative disconnection from body simultaneously
H2: Circuit Immaturity	Matures on normal developmental timeline; full adult circuit available by late adolescence	Differently organised, not simply delayed; atypical social circuits; masking creates subordination-based depression	Epigenetically sensitised; glucocorticoid receptor hypermethylation programs HPA hyperreactivity from earliest weeks	Forcibly accelerated; amygdala sensitised; PFC-limbic imbalance in young children; shame schemas form precociously
H3: Rapid Recovery	Maximum in early childhood; gradually weakened by rumination onset and puberty-driven sleep disruption	Structurally compromised from birth; atypical sleep architecture intrinsic to autism, not secondary to depression	Chronically eroded; absent co-regulation → elevated cortisol suppresses BDNF and prevents SWS from infancy	Actively inverted: sleep becomes re-traumatisation via NE-elevated REM; positive feedback loop entrenches depression
Scaffold Broken By	Intact; progressively dismantled by normal maturation alone	Constitutional neurobiological differences in interoception and sleep architecture	Absence of relational scaffolding; external regulatory resource never established	Acute assault on all three mechanisms simultaneously; H3 inverted
Depression Onset	Late childhood → adolescence; driven by normal maturation	Early but undetected; chronic across lifespan; masked by autism diagnosis	Potentially from infancy in severe cases; embedded in relational context	Immediately post-trauma at any developmental stage; rapidly chronic
Primary Invisibility Mechanism	Age-appropriate: clinical systems lack schema for childhood depression	Diagnostic overshadowing: symptoms absorbed into autism diagnosis	Behavioural overshadowing: relational dysregulation absorbs clinical attention	Diagnostic capture: PTSD absorbs depression; somatic signals misread; dissociation misattributed
Treatment Implication	Standard approaches effective once developmental transition completed	Body-based; CBT requires adaptation; sleep intervention critical; masking reduction essential	Relationship is the treatment; therapeutic attachment compensates for absent relational scaffolding	Somatic + trauma processing first; sleep restoration essential (prazosin, EMDR); antidepressants insufficient alone

6. Discussion

6.1 Theoretical Propositions

From the foregoing analysis, we derive four formal theoretical propositions that constitute the core of the Triple-Invisibility Model and generate its testable predictions.

Proposition 1: Childhood depression is not rare but recurrently invisible. The low clinical prevalence of depression in early childhood reflects the operation of three simultaneous suppression mechanisms — somatic routing, circuit immaturity, and rapid recovery — rather than a genuine absence of depressive neurobiological states. The true rate of depressive episodes in early childhood may substantially exceed detected clinical rates, with the majority of episodes brief, somatically expressed, and resolved below clinical detection thresholds.

Proposition 2: The adolescent prevalence spike is a visibility transition, not purely a vulnerability increase. While adolescence brings new stressors, the sharp epidemiological inflection at puberty is primarily attributable to the near-simultaneous collapse of all three protective mechanisms — driven by a single biological event — rather than to a commensurate increase in environmental adversity. Depression does not suddenly begin at adolescence; it becomes visible.

Proposition 3: The three mechanisms are interdependent and their developmental trajectories are linked. H2 maturation enables rumination, which counteracts H3 sleep-based recovery. H1 weakening depends on the same caregiver attunement that externally supports H3 through co-regulation. And H2 epigenetic sensitisation, as seen in attachment disorder, bypasses the immaturity protection without maturation, demonstrating that the protection of H2 depends on environmental conditions not intrinsic to the circuit itself.

Proposition 4: Antidepressant pharmacotherapy partly substitutes for the natural recovery mechanisms of childhood. The developmental curve of SSRI efficacy mirrors the developmental curve of H3 strength. This convergence suggests that antidepressants may achieve therapeutic benefit partly by pharmacologically restoring BDNF-mediated neuroplasticity, synaptic remodelling capacity, and HPA normalisation that the developing brain once performed rapidly through sleep and neuroplastic recovery. Understanding the childhood recovery mechanism may therefore illuminate antidepressant action and point toward novel therapeutic targets.

6.2 Testable Predictions

The TIM generates several specific empirical predictions that distinguish it from existing frameworks and are, in principle, tractable with available methodology.

Prediction 1 — Ecological Momentary Assessment: Mood sampling studies using EMA methods measuring affect multiple times daily in children aged 4–9 should reveal brief, intense, somatically-accompanied episodes of negative affect with the neural correlates of depression (elevated cortisol, suppressed positive affect) that resolve within 24–72 hours. These micro-episodes should be invisible to weekly or monthly measures, consistent with the invisibility thesis, but should predict later adolescent depression as embryonic kindling events — providing the first direct empirical test of H3.

Prediction 2 — Sleep intervention and depression prevention: Protecting slow-wave sleep architecture in early-to-mid adolescence — particularly through addressing circadian misalignment caused by school start times — should reduce first-episode depression incidence, since this intervention directly targets the H3 collapse that the model identifies as a primary driver of the adolescent inflection. Sleep extension studies in adolescents provide a partial test; longitudinal studies examining the relationship between SWS quality specifically and first-episode MDD would provide a more direct test.

Prediction 3 — Somatic assessment sensitivity: Paediatric somatic complaint patterns (recurrent stomachache, headache, fatigue) in children aged 4–10, assessed in the context of life stressors or family history, should prospectively predict later depressive episodes at rates comparable to established cognitive risk factors — if somatic complaints are partly expressions of subclinical depressive states rather than purely medical presentations. A longitudinal study correlating paediatric somatic patterns with later depression diagnosis would test this directly.

Prediction 4 — BDNF trajectories and episode duration: If H3 operates partly through BDNF-mediated neuroplasticity, serum or CSF BDNF levels should be inversely associated with depressive episode duration across the developmental span, with the association strongest in younger age groups where the BDNF advantage is greatest. This prediction is consistent with existing literature on BDNF depletion in adult MDD but has not been tested developmentally with episode duration as the primary outcome.

6.3 Clinical Implications

The TIM has several practical implications for clinical assessment and intervention with children and adolescents.

Somatic complaints in paediatric settings should be routinely screened for depressive aetiology, particularly when they are recurrent, correlate with social or academic stressors, and are accompanied by reduced engagement in previously enjoyed activities. Current paediatric assessment protocols treat somatic complaints primarily as medical presentations; the TIM argues they should trigger systematic mental health assessment, particularly in children with family history of depression or early adversity.

Assessment instruments for childhood depression should be designed for the developmental presentation rather than downward-extending adult instruments. The capacity to ask about sadness and hopelessness presupposes the psychological channel is open — H1 implies it may not be in early childhood. Instruments that assess somatic, behavioural, and functional dimensions, and that use caregiver and teacher informants rather than relying solely on child self-report, are more ecologically valid for younger children.

For neurodevelopmental and trauma populations, the treatment implications derived from the TIM suggest departures from standard approaches. For autistic individuals, body-based interventions and sleep-targeted treatments address the constitutional H1 and H3 disruptions that standard CBT cannot reach. For attachment disorder, the therapeutic relationship itself functions as the reparative scaffold, providing the mentalization-building and co-regulatory function that was absent in early development — consistent with relationship-based models of treatment such as dyadic developmental psychotherapy. For trauma, somatic and trauma-focused processing must precede standard cognitive-behavioural approaches, and sleep restoration (pharmacological and behavioural) is not adjunctive but primary.

6.4 Limitations and Future Directions

The TIM is a theoretical framework rather than a tested model, and several limitations require acknowledgement. First, the three mechanisms are described in terms of their developmental trajectories but are not precisely operationalised for measurement — future work must develop specific biomarkers and behavioural indices for each mechanism to allow empirical testing. Second, the model treats typical development, autism, attachment disorder, and trauma as relatively distinct categories, whereas clinical reality involves substantial overlap — many autistic children also experience early adversity, many traumatised children have insecure attachment — and future extensions must address comorbid configurations.

Third, the model is built primarily on data from Western, developed-world samples. The cultural evidence that somatic depression presentation varies across cultural contexts (supporting H1’s cultural analogue) also implies that the developmental trajectories described here may not be universal — cultures with different caregiving practices, different sleep norms, or different patterns of emotional socialisation may show different developmental trajectories for each mechanism.

Finally, the evolutionary foundations of the model — while providing important context — carry the risk of naturalistic fallacy: that what is evolutionarily conserved must be adaptive or acceptable. The model explicitly does not endorse this inference. The invisibility of childhood depression is not good simply because it has evolutionary roots — it is a clinical and ethical problem that demands better assessment, better recognition, and ultimately better intervention.

7. Conclusion

The Triple-Invisibility Model proposes a unified account of why childhood depression is clinically rare, phenomenologically distinctive, and epidemiologically late-appearing — not because children are incapable of depression, but because they are equipped from birth with three overlapping mechanisms that suppress its visibility, duration, and cortical elaboration. Somatic routing channels depressive states into the body rather than consciousness. Circuit immaturity prevents the adult form of depression from being architecturally instantiated. And rapid recovery, mediated by neuroplastic mechanisms and abundant sleep, clears subcortical depressive states before they can accumulate into detectable episodes or sensitise the circuit through kindling.

Development, puberty, and accumulated stress progressively dismantle these protections — producing the depression increase of adolescence not as a new phenomenon but as the emergence into visibility of something that was always there. The conditions that break the protective scaffold early — autism, through constitutional neurobiological differences; attachment disorder, through relational deprivation; trauma, through acute assault — produce depression that is earlier, more chronic, more treatment-resistant, and more systematically invisible to diagnostic systems built around the typical developmental trajectory.

Childhood is not a period during which depression cannot occur. It is a period during which the brain is naturally equipped with mechanisms that make depression hard to see. Understanding those mechanisms — and how they fail — is among the most clinically important tasks in developmental psychopathology.

The model does not resolve all questions about childhood depression. It does not replace the detailed empirical work needed in each condition. But it offers what the field has lacked: a unified mechanistic account that explains the epidemiological patterns, the phenomenological differences, the pharmacological gradient, and the distinctive clinical invisibility of depression across the developmental span — and that is specific enough to generate the testable predictions that will determine its validity.

References

Abela, J. R. Z., & Hankin, B. L. (2011). Rumination as a vulnerability factor to depression during the transition from early to middle adolescence: A multiwave longitudinal study. Journal of Abnormal Psychology, 120(2), 259–271.

Andersen, S. L., & Teicher, M. H. (2008). Stress, sensitive periods and maturational events in adolescent depression. Trends in Neurosciences, 31(4), 183–191.

Bridge, J. A., Iyengar, S., Salary, C. B., et al. (2007). Clinical response and risk for reported suicidal ideation and suicide attempts in pediatric antidepressant treatment. JAMA, 297(15), 1683–1696.

Cage, E., & Troxell-Whitman, Z. (2019). Understanding the reasons, contexts and costs of camouflaging for autistic adults. Journal of Autism and Developmental Disorders, 49(5), 1899–1911.

Casey, B. J., Getz, S., & Galvan, A. (2008). The adolescent brain. Developmental Review, 28(1), 62–77.

Cassidy, S., Chez Bourke, L., Coghill, D., & Hartley, M. (2019). Camouflaging autistic traits and its relationship to mental health in autistic adults. Autism in Adulthood, 1(3), 219–228.

Coan, J. A., & Sbarra, D. A. (2015). Social baseline theory: The social regulation of risk and effort. Current Opinion in Psychology, 1, 87–91.

Costello, E. J., Egger, H., & Angold, A. (2006). 10-year research update review: The epidemiology of child and adolescent psychiatric disorders: II. Developmental epidemiology. Journal of the American Academy of Child and Adolescent Psychiatry, 44(10), 972–986.

Craig, A. D. (2009). How do you feel — now? The anterior insula and human awareness. Nature Reviews Neuroscience, 10(1), 59–70.

Cytryn, L., & McKnew, D. H. (1972). Proposed classification of childhood depression. American Journal of Psychiatry, 129(2), 149–155.

Dantzer, R., O’Connor, J. C., Freund, G. G., Johnson, R. W., & Kelley, K. W. (2008). From inflammation to sickness and depression: When the immune system subjugates the brain. Nature Reviews Neuroscience, 9(1), 46–56.

Engel, G. L., & Schmale, A. H. (1972). Conservation-withdrawal: A primary regulatory process for organismic homeostasis. CIBA Foundation Symposium, 8, 57–75.

Feiring, C., & Taska, L. S. (2005). The persistence of shame following sexual abuse: A longitudinal look at risk and recovery. Child Maltreatment, 10(4), 337–349.

Fonagy, P., Gergely, G., Jurist, E., & Target, M. (2002). Affect Regulation, Mentalization and the Development of the Self. Other Press.

Garfinkel, S. N., Tiley, C., O’Keeffe, S., Harrison, N. A., Seth, A. K., & Critchley, H. D. (2016). Discrepancies between dimensions of interoception in autism: Implications for emotional and social functioning. Autism, 20(6), 655–665.

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(39), 15638–15643.

Gilbert, P., & Allan, S. (1998). The role of defeat and entrapment (arrested flight) in depression: An exploration of an evolutionary view. Psychological Medicine, 28(3), 585–598.

Glaser, K. (1967). Masked depression in children and adolescents. American Journal of Psychotherapy, 21(3), 565–574.

Harlow, H. F., & Zimmermann, R. R. (1959). Affectional responses in the infant monkey. Science, 130(3373), 421–432.

Hollocks, M. J., Lerh, J. W., Magiati, I., Meiser-Stedman, R., & Brugha, T. S. (2019). Anxiety and depression in adults with autism spectrum disorder: A systematic review and meta-analysis. Psychological Medicine, 49(4), 559–572.

Kaufman, I. C., & Rosenblum, L. A. (1967). The reaction to separation in infant monkeys: Anaclitic depression and conservation-withdrawal. Psychosomatic Medicine, 29(6), 648–675.

Kleinman, A. (1982). Neurasthenia and depression: A study of somatization and culture in China. Culture, Medicine and Psychiatry, 6(2), 117–190.

Kovacs, M., & Beck, A. T. (1977). An empirical-clinical approach toward a definition of childhood depression. In J. G. Schulterbrandt & A. Raskin (Eds.), Depression in Childhood: Diagnosis, Treatment, and Conceptual Models (pp. 1–25). Raven Press.

Krishnan, V., Han, M. H., Graham, D. L., et al. (2007). Molecular adaptations underlying susceptibility and resistance to social defeat in brain reward regions. Cell, 131(2), 391–404.

Luby, J. L., Heffelfinger, A. K., Mrakotsky, C., et al. (2003). The clinical picture of depression in preschool children. Journal of the American Academy of Child and Adolescent Psychiatry, 42(3), 340–348.

Luby, J. L., Si, X., Belden, A. C., Tandon, M., & Spitznagel, E. (2009). Preschool depression: Homotypic continuity and course over 24 months. Archives of General Psychiatry, 66(8), 897–905.

McGowan, P. O., Sasaki, A., D’Alessio, A. C., et al. (2009). Epigenetic regulation of the glucocorticoid receptor in human brain associates with childhood abuse. Nature Neuroscience, 12(3), 342–348.

Meaney, M. J. (2001). Maternal care, gene expression, and the transmission of individual differences in stress reactivity across generations. Annual Review of Neuroscience, 24, 1161–1192.

Murray, D., Lesser, M., & Lawson, W. (2005). Attention, monotropism and the diagnostic criteria for autism. Autism, 9(2), 139–156.

Nesse, R. M. (2000). Is depression an adaptation? Archives of General Psychiatry, 57(1), 14–20.

Nolen-Hoeksema, S., & Girgus, J. S. (1994). The emergence of gender differences in depression during adolescence. Psychological Bulletin, 115(3), 424–443.

Panksepp, J. (2011). Cross-species affective neuroscience decoding of the primal affective experiences of humans and related animals. PLOS ONE, 6(9), e21236.

Panksepp, J., & Watt, D. (2011). Why does depression hurt? Ancestral primary-process separation-distress (PANIC/GRIEF) and diminished brain reward (SEEKING) processes in the genesis of depressive affect. Psychiatry: Interpersonal and Biological Processes, 74(1), 5–13.

Perry, B. D., Pollard, R. A., Blakley, T. L., Baker, W. L., & Vigilante, D. (1995). Childhood trauma, the neurobiology of adaptation, and use-dependent development of the brain: How states become traits. Infant Mental Health Journal, 16(4), 271–291.

Porges, S. W. (1995). Orienting in a defensive world: Mammalian modifications of our evolutionary heritage. A Polyvagal Theory. Psychophysiology, 32(4), 301–318.

Porges, S. W. (2011). The Polyvagal Theory: Neurophysiological Foundations of Emotions, Attachment, Communication, and Self-Regulation. W. W. Norton.

Post, R. M. (1992). Transduction of psychosocial stress into the neurobiology of recurrent affective disorder. American Journal of Psychiatry, 149(8), 999–1010.

Price, J., Sloman, L., Gardner, R., Gilbert, P., & Rohde, P. (1994). The social competition hypothesis of depression. British Journal of Psychiatry, 164(3), 309–315.

Ryder, A. G., Yang, J., Zhu, X., et al. (2008). The cultural shaping of depression: Somatic symptoms in China, psychological symptoms in North America. Journal of Abnormal Psychology, 117(2), 300–313.

Suomi, S. J. (2006). Risk, resilience, and gene x environment interactions in rhesus monkeys. Annals of the New York Academy of Sciences, 1094, 52–62.

Tarullo, A. R., & Gunnar, M. R. (2006). Child maltreatment and the developing HPA axis. Hormones and Behavior, 50(4), 632–639.

Teicher, M. H., Samson, J. A., Anderson, C. M., & Bhatt, M. (2016). The effects of childhood maltreatment on brain structure, function and connectivity. Nature Reviews Neuroscience, 17(10), 652–666.

Tononi, G., & Cirelli, C. (2006). Sleep function and synaptic homeostasis. Sleep Medicine Reviews, 10(1), 49–62.

van der Kolk, B. A. (2005). Developmental trauma disorder: Toward a rational diagnosis for children with complex trauma histories. Psychiatric Annals, 35(5), 401–408.

van der Kolk, B. A. (2014). The Body Keeps the Score: Brain, Mind, and Body in the Healing of Trauma. Viking.

Walker, M. P., & van der Helm, E. (2009). Overnight therapy? The role of sleep in emotional brain processing. Psychological Bulletin, 135(5), 731–748.

Willner, P. (1997). Validity, reliability and utility of the chronic mild stress model of depression: A 10-year review and evaluation. Psychopharmacology, 134(4), 319–329.

Zeanah, C. H., & Gleason, M. M. (2015). Annual research review: Attachment disorders in early childhood — clinical presentation, causes, correlates, and treatment. Journal of Child Psychology and Psychiatry, 56(3), 207–222.

(CL)