Hypothesis

Central amygdala corticotropin‑releasing factor (CRF) neurons undergo an age‑dependent decline in intrinsic excitability that flips the sign of amygdala‑vPFC structural connectivity’s influence on trait anxiety. In youth, heightened CRF‑driven amygdala output amplifies threat signaling; reduced ventral prefrontal‑amygdala white‑matter integrity therefore fails to restrain this drive, yielding a positive correlation between connectivity strength and anxiety. With advancing age, CRF neuron hyperpolarization diminishes their feed‑forward excitation of basal‑amygdala circuits, so the same loss of connectivity now removes a maladaptive excitatory tone, revealing a negative association between connectivity and anxiety. Consequently, threat‑evoked amygdala reactivity continues to predict fear‑extinction failure (r = 0.56) [3], but the protective effect of reduced CRF signaling lowers overall anxiety despite persistent extinction deficits.

Mechanistic rationale

1. Developmental trajectory of CRF excitability – Rodent and post‑mortem human data show CRF expression peaks in early adulthood and declines with age via epigenetic silencing of the Crh promoter and increased expression of Kv7 channels. This predicts an inverted U‑shape of CRF‑mediated amygdala output across the lifespan.
2. Interaction with vPFC‑amygdala white matter – Reduced fractional anisotropy and increased radial diffusivity reflect weakened top‑down inhibitory control [1]. When CRF drive is high (young), weak control → net ↑ anxiety; when CRF drive is low (old), weak control → net ↓ anxiety because the excitatory drive that would otherwise be unchecked is already muted.
3. Extinction link – Amygdala reactivity to threatening faces indexes CRF‑dependent potentiation of basal‑amygdala output that interferes with safety learning, accounting for the observed correlation with fear‑potentiated startle during late extinction [3]. Age‑related CRF dampening blunts this interference, yet extinction retention remains impaired because downstream plasticity in intercalated cells is unaffected.

Testable predictions

• In vivo chemogenetics: Selective inhibition of CRF neurons in the central amygdala of young adult mice will replicate the aged phenotype—reduced anxiety despite preserved low vPFC‑amygdala FA—whereas activation of these neurons in aged mice will reinstate a positive connectivity‑anxiety correlation.
• Human PET‑MR: Using a CRF‑R1 antagonist radiotracer (e.g., ^11C‑MVB) combined with diffusion tensor imaging, we predict a negative correlation between CRF‑R1 binding potential and anxiety scores in participants < 40 y, but a positive correlation in those > 55 y, mirroring the connectivity‑anxiety reversal.
• Pharmacological challenge: Acute administration of a CRF‑R1 antagonist should decrease anxiety in young adults but increase it in older adults, producing an age‑by‑drug interaction on state‑anxiety measures.
• Extinction assay: Fear‑extinction retention will correlate with amygdala reactivity only when CRF signaling is pharmacologically enhanced; blocking CRF‑R1 will abolish the reactivity‑extinction link across ages.

Falsifiability

If manipulating CRF neuron activity does not invert the anxiety‑connectivity relationship as predicted, or if CRF‑R1 binding shows no age‑dependent shift in correlation with anxiety or connectivity, the hypothesis would be refuted. Likewise, failure to observe an age‑by‑CRF‑antagonist interaction on state anxiety would challenge the proposed mechanism.

Implications

Establishing a bidirectional switch in CRF neuron excitability would explain why structural disconnection can be protective in aging, reconcile extinction deficits with lower trait anxiety, and suggest CRF‑R1 modulation as an age‑specific therapeutic avenue for anxiety disorders.