HPA axis

The hypothalamic-pituitary-adrenal axis: the neuroendocrine arm of the stress response, terminating in cortisol release. Its relevance here is mechanistic — it is the most concrete pathway offered for how a social variable (attachment-style) could produce a physiological difference in interoception.

The architectural argument

Oldroyd et al. (2019) note that stress and interoception traverse the same brain-body pathways in opposite directions: the stress response is the descending brain-body connection, interoception the ascending one (citing Seth 2013). Anything that chronically dysregulates the descending arm should therefore be expected to leave marks on the ascending one. And individual differences in attachment are reliably characterized by differential HPA reactivity to stress across the lifespan (Allen & Miga 2010; Diamond & Fagundes 2010; Lovallo 2013).

Two proposed routes to altered interoception

1. Signal strength. HPA activation releases epinephrine, which increases cardiac contractility, heart rate and depolarization — and thus stroke volume. Schandry et al. (1993) found stroke volume predicts heartbeat-detection performance: the more blood pumped per beat, the better people estimate their own heart rate. Higher stroke volume is therefore a literally stronger interoceptive signal, and chronically increased sympathetic outflow has been proposed as one route to high interoceptive accuracy (Paulus & Stein 2010).

This is a notable point for the heartbeat-detection-task: it implies heartbeat-detection scores partly index cardiodynamics, not only perceptual skill. A person with a more forceful heartbeat has an easier signal to detect. Any interpretation of heartbeat-detection differences as differences in perceptual ability inherits this confound.

2. Signal processing. Cortisol appears to lower the threshold for interoceptive signal processing centrally — the brain becomes more attuned to interoceptive signals in its presence (Rief et al. 1998). Intravenous cortisol (4 mg) improves interoceptive accuracy (Schulz et al. 2013), and the regions responsible for attentional processing of interoceptive signals — ACC and OFC — show greater activation under cortisol (Cameron 2002; Critchley et al. 2004; Pollatos et al. 2007). Chronic HPA dysregulation may therefore permanently alter perception of bodily cues (Schulz & Vögele 2015).

Interpretive tension

Both routes predict that stress increases interoceptive accuracy — stronger signal, lower processing threshold. But the developmental thesis of oldroyd-2019-attachment-interoception is that insecure attachment (and hence HPA dysregulation) yields worse or distorted interoception. The paper does not reconcile these, and the mechanism section is never tested against its own data: no cortisol, stroke volume, or HPA measure was collected in either study.

A charitable reading is that the anxious profile is where the mechanism fits — heightened noticing plus heightened worry (attachment-style) is what “stronger signal + lower threshold” should produce, and the term the authors reach for is distorted rather than diminished interoception. It does not obviously explain the avoidant profile, which the paper attributes to attentional strategy and neural architecture instead. Worth holding loosely: this is a plausible mechanism sketch, not a result.

What drives the axis, and why the timing matters

LeDoux (2012) supplies the upstream half this page previously left implicit. Central amygdala outputs target neurons activating the sympathetic division of the ANS (releasing adrenergic hormones from the adrenal medulla) and the HPA axis (releasing cortisol from the adrenal cortex) — so the axis is an output limb of a survival circuit, triggered by threat detection rather than free-standing.

His timing observation bears on the wiki’s mechanism discussion above: central neuromodulator effects are rapid, whereas peripheral hormone effects are “considerably slower, allowing the prolongation of the survival state for extended periods of time.” Cortisol crosses the blood-brain barrier and binds receptors widely; adrenergic hormones affect the CNS indirectly (McGaugh 2000). This is a concrete route by which a brief trigger yields a sustained state — and it matters for the chronic-dysregulation claims Oldroyd et al. rest on, since it is the slow limb that would plausibly leave lasting marks.

Also relevant: arousal signals feed back to facilitate the very circuit that triggered them, and facilitate sensory and memory areas — a loop, not a one-way output. See global-organismic-state.

The PTSD profile: the axis runs backwards from what you’d expect

The van der Kolk (1994) ingest adds the axis’s behaviour under chronic trauma, and it is the opposite of the acute-stress picture this page otherwise carries — and worth keeping straight because it is easy to assume more stress means more cortisol.

Acute stress mobilizes the axis and raises glucocorticoids. But organisms adapt to chronic, inescapable stress by a negative-feedback down-regulation, and van der Kolk reports the resulting PTSD profile as:

• Low resting cortisol (Mason, Yehuda) — even with comorbid depression, which normally raises cortisol. This is the finding that most sharply separates PTSD from major depression neuroendocrinologically.
• Enhanced negative feedback: glucocorticoid-receptor up-regulation (proportional to symptom severity, Yehuda) and hyperresponsiveness to low-dose dexamethasone — a more sensitive, faster-recovering axis, not a blunted one.
• Chronically elevated catecholamines alongside the low cortisol (elevated urinary norepinephrine/epinephrine; exaggerated MHPG response to yohimbine, which also precipitated panic and flashbacks). So the sympathetic and glucocorticoid arms dissociate.
• Prior trauma blunts the cortisol response to new trauma (Resnick’s rape-victim data, cited unpublished) — a sensitization of the feedback loop.

Van der Kolk reads this as chronic stress permanently altering how the organism handles subsequent stress — the same “chronic dysregulation leaves lasting marks” logic Oldroyd et al. rest their developmental thesis on, here with a concrete (if 1994-vintage, and internally inconsistent — Pitman & Orr failed to replicate) endocrine signature. The literature it summarizes is contested; the wiki records the PTSD low-cortisol/high-feedback pattern as van der Kolk’s reading of it, not as settled fact.

The same tension, confirmed from a second source (Bonaz et al. 2021)

The interpretive tension recorded above — both of Oldroyd et al.’s mechanisms predict stress raising interoceptive accuracy, while their developmental thesis needs it lowered — recurs in Bonaz et al. (2021), from an entirely different literature and without either source noticing.

Their early-adversity evidence runs the other way from this page’s mechanisms: childhood trauma is negatively correlated with interoceptive ability (heartbeat detection after cold pressor, Schaan et al. 2019), and childhood maltreatment is associated with lower cardiovascular and HPA responses alongside higher subjective emotional distress (Gooding et al. 2016) — blunted physiology, amplified report. Early adversity also dampens functional reactivity in circuits overlapping interoceptive networks (Ansell et al. 2012; Seo et al. 2014).

Two independent reviews now record a chronic-stress-lowers-interoceptive-accuracy pattern while the wiki’s mechanism section holds two routes predicting the opposite. That makes it a field-level gap rather than one paper’s oversight, and it is worth stating what would resolve it: the mechanisms here (stroke volume, cortisol lowering the central threshold) are acute effects of an intact axis, whereas the adversity findings are chronic outcomes of a down-regulated one. That reconciliation is available and neither source makes it. Recorded as an open seam.

Bonaz et al. also extend the axis’s reach in two directions this page did not have. Stress hormones act as conditioned interoceptive cues motivating alcohol consumption, with alcohol upregulating glucocorticoid receptors across limbic forebrain and mPFC — the axis as a learned drug cue rather than only a stress output (see craving). And stress and sex hormones are proposed to act on the biomechanical functioning of connective tissue itself, contributing to organ-level dysfunction in joint-hypermobility — the axis acting on the tissue that generates the interoceptive signal, not only on the brain that reads it. That second claim is speculative and carries one citation.

Related

Sits alongside homeostasis and allostasis as the endocrine face of the body-regulation story that interoception senses. The chronic-dysregulation story here is the hyperarousal arm of trauma’s bimodal response; the numbing arm runs through stress-induced-analgesia.