← back to proto-thoughtsEstrogen, Depression, and the Hormonal Angle
January 2026
Two proto-notes merged: estrogen depression effects + the trans-specific angle.
Standard narratives for trans depression rates (~41% attempted suicide):
• Rightists: degeneracy
• Leftists: discrimination
• Transmedicalists: the disorder called Gender Dysphoria
Missing hypothesis: it's largely hormones.
—
The neurochemical picture of exogenous estrogen:
Prolactin ↑ → depression-associated
Prolactin elevation is a transdiagnostic hormonal phenotype in depression (see Wieck et al.)
High prolactin → low dopamine (they're inversely related via tuberoinfundibular pathway)
This creates a depression-compatible neurochemical state independent of social factors
Serotonin receptor changes:
5-HT(2A) receptors ↑ → depression-associated (this is NOT protective — the 1A is the good one)
Estrogen decreases 5-HT(1A) function
1A agonism is anxiolytic (buspirone works here)
So estrogen shifts serotonergic balance toward the depression-associated phenotype
CRF pathway:
Estrogen modulates corticotropin-releasing factor
CRF strongly implicated in depression and anxiety
Estrogen can increase HPA axis reactivity
—
The active inference angle (connecting to "Depression as High Learning Rate"):
Depression = high precision on ascending prediction errors (ACh↑) + low precision on priors/policies (DA↓)
Does estrogen shift precision weighting in predictive processing?
Hypothesis: Estrogen may increase ACh-mediated gain on sensory prediction errors while decreasing catecholaminergic precision on high-level priors
Mechanisms:
• Prolactin elevation → DA suppression → lower prior precision
• Estrogen's effects on cholinergic system (complex but generally facilitatory)
• Net effect: bias toward the "high learning rate" depressive phenotype
This would mean HRT doesn't just "add stress via social factors" — it directly shifts the computational parameters of the brain toward a depression-susceptible regime.
—
Trans-specific considerations:
If cis women have different depression profiles than cis men (they do — ~2x prevalence, different symptom clusters), why would we expect someone on HRT to sit in neither category?
They're in a third hormonal state:
• Exogenous estrogen without endogenous cycling
• Different receptor sensitivities (natal male brain exposed to estrogen late)
• Different feedback loops (no ovaries to regulate)
This third state has its own neurochemical signature that we haven't properly characterized.
The question isn't being seriously investigated. The biological channel is politically inconvenient:
• Right doesn't want to admit transition could be medically valid at all
• Left doesn't want to admit transition has biological costs beyond "discrimination stress"
• Medicine follows the funding
—
Possible interventions (speculative, based on mechanism):
Buspirone
5-HT(1A) partial agonist
May partially compensate for estrogen's 5-HT(1A) downregulation
Already used for anxiety; underexplored for HRT-associated depression
Dopamine support
Bupropion (DA/NE reuptake inhibitor)
Bromantane (non-depleting dopaminergic)
Targeting the prolactin→DA suppression axis
Progesterone considerations
Progesterone has its own neuroactive effects (GABA-A modulation)
Some trans women report mood improvement with progesterone
Mechanism unclear but may provide counterbalance
Thyroid optimization
Estrogen increases thyroid-binding globulin
Can effectively reduce free T3/T4
Thyroid intimately linked to mood
May need to dose thyroid higher on HRT
—
Integration with Depression as High Learning Rate framework:
The depressive state is characterized by:
• High ACh → high gain on prediction errors → treating noise as signal
• Low DA → low precision on policies/priors → fragile high-level beliefs
• Excessive REM → nightly erosion of stabilizing self-models
Estrogen potentially pushes all three dials in the depression direction.
This doesn't mean transition is wrong. It means:
1. The depression isn't purely social/psychological
2. It may be pharmacologically addressable
3. We need actual research instead of pretending the question doesn't exist
Direction for future thinking:
• Does estrogen shift precision weighting? (testable in neuroimaging)
• Does prolactin mediate a significant portion of HRT-depression? (testable with prolactin-lowering drugs)
• Is there an optimal HRT protocol that minimizes depression-susceptibility? (requires actually looking)
The 41% is not destiny. It might be partly iatrogenic. And iatrogenic problems are solvable.