Testosterone generally acts as a brake on anxiety and fear, dampening the body’s stress hormone output and shifting how the brain processes threats. Population studies consistently find that men with the lowest testosterone levels report more anxiety symptoms, and single-dose experiments show the hormone can reduce automatic fear responses within hours. But the relationship is not as straightforward as “more testosterone, less anxiety.” The story involves at least two chemical conversions inside the brain, a critical programming window during puberty, and situations where elevated testosterone actually accompanies worse anxiety.
How Testosterone Turns Down the Cortisol Response
One of the clearest ways testosterone affects anxiety is by dialing back cortisol, the body’s primary stress hormone. When researchers experimentally suppressed testosterone in healthy men and then challenged their stress systems with corticotropin-releasing hormone (CRH), cortisol responses shot up. Restoring testosterone brought those cortisol responses back down significantly. Interestingly, the pituitary hormone that tells the adrenal glands to release cortisol (ACTH) actually went up during testosterone replacement, suggesting testosterone’s suppressive effect happens at the adrenal gland itself rather than higher up in the chain.1PubMed Central. Testosterone suppression of CRH-stimulated cortisol in men In practical terms, this means testosterone doesn’t just make you feel less anxious subjectively; it physically limits how much stress hormone your body pumps out when provoked.
This cortisol-suppressing effect also feeds into sex differences in stress reactivity. Female rats exposed to stress secrete significantly more ACTH than males, a gap that disappears when their ovaries are removed, pointing to a hormonal basis for the difference. The implication is that the balance between androgens and estrogens shapes how intensely the stress system fires, not just whether it fires at all.
What Testosterone Does Inside the Amygdala
The amygdala is the brain region most associated with detecting threats and generating fear responses, and testosterone reshapes how it operates. The picture here is more nuanced than “testosterone calms the amygdala down.” In fact, giving women a single dose of testosterone increased amygdala responses to both fearful and happy faces equally, suggesting the hormone makes the amygdala more reactive across the board rather than selectively quieting fear signals.2PubMed. Testosterone administration in women increases amygdala responses to fearful and happy faces
Where testosterone does seem to reduce anxiety is in shifting what the amygdala does with threat information. Brain imaging shows that testosterone increases amygdala activity during threat approach and decreases it during threat avoidance, essentially rewiring the motivational direction from “get away” to “deal with it.”3PubMed Central. Testosterone biases the amygdala toward social threat approach This aligns with a related finding that testosterone rapidly reduces the communication between the amygdala and the orbitofrontal cortex, a brain area involved in cautious evaluation and behavioral regulation.4PubMed. Testosterone reduces amygdala-orbitofrontal cortex coupling When that coupling weakens, the brake that the prefrontal cortex normally applies to emotional impulses loosens. The result is less hesitation in the face of social threats.
This decoupling pattern also appears naturally during puberty. Adolescents whose testosterone rose the most over time showed the steepest drop in amygdala-orbitofrontal cortex connectivity, mirroring the effect seen in single-dose drug studies.5Social Cognitive and Affective Neuroscience. Pubertal testosterone influences threat-related amygdala–orbitofrontal cortex coupling So the brain rewiring testosterone produces in adulthood appears to be the same process that unfolds gradually during adolescent development.
Two Chemical Conversions That Explain the Anxiolytic Effect
Testosterone itself may not be the molecule doing most of the anxiety-reducing work. Instead, the brain converts testosterone into other compounds, and those derivatives are the active players.
The first pathway runs through GABA, the brain’s main inhibitory neurotransmitter and the same system targeted by benzodiazepines like Valium. In mice, testosterone is broken down into neurosteroids (androsterone and 3α-androstandione) that bind to GABA receptors, producing a calming effect. When researchers blocked GABA receptors with antagonists, testosterone’s anxiety-reducing effects vanished entirely.6PubMed. Testosterone rapidly reduces anxiety in male house mice (Mus musculus) The neurosteroid metabolites were anxiolytic at lower doses than testosterone itself, supporting the idea that testosterone is a precursor rather than the final active molecule.
The second pathway involves aromatase, an enzyme that converts testosterone into estradiol (a form of estrogen). This matters for fear extinction, the process by which your brain learns that something previously frightening is no longer dangerous. Blocking aromatase in male rats significantly impaired their ability to extinguish a conditioned fear, and giving them estradiol rescued the deficit completely.7PubMed Central. Inhibition of estradiol synthesis impairs fear extinction in male rats A separate experiment showed that testosterone and estradiol both kept fear memories precise and context-specific, preventing the overgeneralization of fear to neutral environments. But DHT, a potent androgen that cannot be converted to estradiol, failed to prevent fear generalization, as did blocking aromatase directly.8PubMed. Aromatized testosterone attenuates contextual generalization of fear in male rats The takeaway is that testosterone’s ability to keep fear responses appropriately targeted depends on its conversion to estradiol inside the brain. Without that conversion, fear bleeds into situations where it doesn’t belong.
Low Testosterone and Anxiety in Population Studies
The laboratory findings line up with what epidemiologists see in large populations. In the Tromsø study, which surveyed men from the general population of Norway, anxiety symptoms were negatively associated with both total and free testosterone. Men whose testosterone fell in the lowest tenth percentile had significantly elevated anxiety scores, and the link persisted after adjusting for age, waist circumference, education, and other confounders.9PubMed. Testosterone levels and psychological health status in men from a general population: the Tromsø study
Panic disorder offers a sharper clinical picture. Male patients experiencing acute exacerbations of panic disorder with agoraphobia had significantly lower plasma testosterone than healthy controls. A lower testosterone-to-cortisol ratio was associated with more severe panic symptoms in these men, though the same pattern did not hold in female patients.10Psychoneuroendocrinology. Associations of plasma testosterone with clinical manifestations in acute panic disorder The relationship between testosterone and anxiety also appears to change with age. In men with sexual dysfunction, an inverse association between body-focused anxiety and testosterone levels showed up only in the oldest group studied, suggesting the connection strengthens as testosterone naturally declines.11PubMed. The age-related decline of testosterone is associated with different specific symptoms and signs in patients with sexual dysfunction
A review of clinical presentations noted that testosterone deficiency can manifest as a wide range of anxiety-related states, from unfocused worry and phobic avoidance to full-blown panic syndromes.12PubMed. Testosterone, mood, behaviour and quality of life None of these associations prove that low testosterone causes the anxiety rather than vice versa, but the consistency across study designs is hard to dismiss.
Puberty as a Programming Window
Perhaps the most striking finding in this field is that testosterone’s influence on anxiety may be largely set during adolescence. In California mice, a species where both sexes can be socially defeated, adult females show strong anxiety-like responses to social stress (reduced social approach, heightened vigilance) while adult males are relatively resilient. But juvenile males and females respond to defeat equally. The sex difference only emerges after puberty.13PubMed Central. Sexual differentiation of neural mechanisms of stress sensitivity during puberty
Castrating male mice before puberty erased their adult resilience entirely, making them as anxiety-prone after social defeat as females. Crucially, castrating males after puberty did not have this effect; their stress resilience had already been programmed in. When prepubertally castrated males received testosterone or DHT replacement during the pubertal window, their adult resilience was restored.14bioRxiv. Pubertal Androgens Reduce the Effects of Social Stress on Anxiety-related Behaviors in California Mice This suggests pubertal androgens permanently organize the neural circuits that govern anxiety responses to social stress. The hormone doesn’t just modulate anxiety moment-to-moment; it shapes the hardware during a sensitive period.
These findings resonate with the human observation that anxiety disorders become more common in women than men after puberty but not before. While you can’t run the same experiments in teenagers, the animal data offer a plausible explanation for why that sex difference emerges when it does.
Single Doses and Social Threat
Some of the most intriguing experimental work in humans uses single sublingual doses of testosterone to see how quickly it changes behavior. In healthy women, a single dose significantly reduced automatic avoidance of angry faces without affecting responses to happy faces.15PubMed. Alleviating social avoidance: effects of single dose testosterone administration on approach-avoidance action In women diagnosed with social anxiety disorder, the same approach worked: testosterone administration increased their tendency to approach angry facial expressions rather than pull away, counteracting the persistent avoidance pattern that characterizes their condition.16Clinical Psychological Science. Dare to Approach
One study teased apart conscious and unconscious fear responses. Testosterone reduced the automatic vigilance response to masked (subconsciously presented) fearful faces, but it did not change participants’ self-reported anxiety.17Biological Psychiatry. Testosterone Reduces Unconscious Fear but Not Consciously Experienced Anxiety: Implications for the Disorders of Fear and Anxiety This distinction matters: testosterone appears to operate primarily on the automatic, reflexive fear system rather than the conscious worry system. You might still feel anxious while your body is behaving less fearfully, which could explain why people don’t always report feeling calmer even when objective measures of fear reactivity drop.
Why Testosterone Replacement Doesn’t Reliably Fix Anxiety
Given all the evidence above, you might expect that prescribing testosterone to anxious men with low levels would solve the problem. The clinical reality is much messier. A systematic review of testosterone therapy’s psychiatric effects in adult men found that anxiety outcomes were inconsistent across trials.18PubMed Central. Psychiatric and Cognitive Effects of Testosterone Therapy in Adult Men: A Systematic Review of Clinical Evidence and Mechanistic Insights In young men with congenital hypogonadism (meaning they produce very little testosterone from birth), replacement therapy significantly improved depression scores and sexual function, but the improvement in anxiety scores was not statistically significant.19Endocrine Journal. Increased frequency of anxiety, depression, quality of life and sexual life in young hypogonadotropic hypogonadal males and impacts of testosterone replacement therapy on these conditions
Why the disconnect? Several factors probably contribute. Anxiety is not a single entity; generalized worry, social avoidance, and panic responses involve overlapping but distinct brain circuits, and testosterone may affect some more than others. The timing and duration of treatment also matter. The pubertal programming data suggest that some of testosterone’s anti-anxiety effects are organizational and permanent, meaning you cannot replicate them with adult hormone replacement. And as the single-dose research shows, testosterone may reduce unconscious threat reactivity without touching the conscious experience of worry, which is what clinical questionnaires measure.
The Cortisol Complication
Testosterone doesn’t operate in a hormonal vacuum. The dual-hormone hypothesis holds that testosterone’s behavioral effects depend heavily on what cortisol is doing at the same time. Research on dominance behavior found that higher baseline testosterone was associated with more assertive, confident leadership and willingness to compete after a loss, but only when cortisol was low. When cortisol was high, those relationships weakened or even reversed.20PubMed Central. Beyond the Challenge Hypothesis: The Emergence of the Dual-Hormone Hypothesis and Recommendations for Future Research
For someone trying to understand their own anxiety, this has a practical implication. Having healthy testosterone levels may not do much if your cortisol is chronically elevated from poor sleep, ongoing stress, or other factors. The two hormones form a ratio, and it’s the ratio that predicts behavior, not either hormone in isolation. The panic disorder study mentioned earlier found the same thing: it wasn’t just low testosterone that tracked with worse symptoms in men, but specifically a low testosterone-to-cortisol ratio.
When High Testosterone Coincides with More Anxiety
If testosterone reliably reduced anxiety, you’d expect conditions involving excess androgens to come with less of it. Polycystic ovary syndrome (PCOS) tests that assumption and breaks it. Women with PCOS, who typically have elevated androgen levels, have dramatically increased odds of anxiety symptoms. A meta-analysis across nine studies found that women with PCOS had more than five times the odds of reporting anxiety symptoms compared to controls.21PubMed. Depression and Anxiety in Polycystic Ovary Syndrome: Etiology and Treatment When the analysis was restricted to moderate-to-severe anxiety, the odds were even higher.22Human Reproduction. High prevalence of moderate and severe depressive and anxiety symptoms in polycystic ovary syndrome: a systematic review and meta-analysis
But blaming the androgens themselves may be a mistake. In matched subgroups, testosterone levels were not correlated with mood scores; instead, the anxiety appeared attributable to PCOS symptoms like unwanted hair growth, acne, weight gain, and fertility problems, along with metabolic factors like insulin resistance.23PubMed. Testosterone and mood dysfunction in women with polycystic ovarian syndrome compared to subfertile controls PCOS is a reminder that the social and metabolic consequences of hormonal imbalances can dwarf the direct neurochemical effects of the hormones themselves.
Anabolic steroid abuse tells a different cautionary tale. In animal models, supraphysiological doses of androgens reduced anxiety during the exposure period but produced a rebound increase in anxiety during withdrawal.24PubMed Central. Adolescent Anabolic/Androgenic Steroids: Aggression and Anxiety During Exposure Predict Behavioral Responding During Withdrawal in Syrian Hamsters (Mesocricetus auratus) The exposure-then-withdrawal seesaw is consistent with human reports from steroid users, though a systematic review of studies in athletes concluded that the overall effects of anabolic steroids on anxiety are unclear and inconsistent across the literature.25PubMed Central. Anabolic-androgenic steroid use and psychopathology in athletes. A systematic review What is clearer is that starting anabolic steroid use during adolescence, when the brain is still being organized by pubertal hormones, may carry different and potentially worse psychiatric risks than adult use.26PubMed Central. Anabolic/androgenic steroid administration during adolescence and adulthood differentially modulates aggression and anxiety
Sleep, the Gut, and the Feedback Loop
Testosterone and anxiety don’t just influence each other through direct brain chemistry. Lifestyle factors create feedback loops that can make either one worse. Sleep fragmentation offers a clear example. In rats subjected to disrupted sleep, testosterone, estradiol, and progesterone all dropped significantly compared to controls, while anxiety-related behavior increased, even though corticosterone (the rodent equivalent of cortisol) did not change.27PLOS ONE. Short-term sleep fragmentation enhances anxiety-related behavior: The role of hormonal alterations The fact that anxiety rose without a cortisol increase suggests the hormonal drop itself was enough to shift behavior.
This creates a vicious cycle familiar to many people: anxiety disrupts sleep, disrupted sleep lowers testosterone and other hormones, and the hormonal decline feeds more anxiety. The gut microbiome may participate in this loop as well. Gut bacteria actively modulate sex hormone levels through enzymatic processes, and fluctuations in sex hormones in turn alter gut barrier function and microbial composition.28PubMed. Gut feeling: Exploring the intertwined trilateral nexus of gut microbiota, sex hormones, and mental health The practical relevance is that factors like diet, antibiotic use, and chronic stress could all indirectly affect testosterone-anxiety dynamics through routes that have nothing to do with the gonads.
Prenatal Testosterone and Anxiety Patterns in Adulthood
Testosterone’s influence on anxiety may begin before birth. Digit ratio (the length of the index finger relative to the ring finger) is used as a rough proxy for prenatal androgen exposure, with a lower ratio suggesting higher fetal testosterone. A study using this marker found that men with more female-typical digit ratios and sex-linked traits reported greater anxiety, consistent with the idea that lower prenatal testosterone exposure is associated with an anxiety-prone phenotype in men. In women, the pattern was different: anxiety was associated with both female-typical and male-typical traits, and digit ratio itself did not predict anxiety levels.29PubMed Central. Anxiety, sex-linked behaviors, and digit ratios (2D:4D)
Digit ratio research is inherently imprecise, and no single study of this kind should be taken too seriously. But the finding fits the broader pattern seen in the pubertal programming work: testosterone’s organizational effects during sensitive developmental windows may matter more for long-term anxiety vulnerability than your circulating hormone levels on any given Tuesday.
When Higher Testosterone Means More Caution
One of the more surprising findings in this field comes from wild gerbils. Free-living Baluchistan gerbils with higher testosterone levels actually took fewer risks in experimental settings, not more. This held for both males and females, and experimentally boosting testosterone with implants produced the same pattern.30PubMed. Free-living gerbils with higher testosterone take fewer risks The finding is a useful corrective to the popular notion that testosterone universally promotes boldness. In a wild animal balancing survival against reproduction, the relationship between hormones and risk-taking is shaped by the ecological context. Testosterone has many functions, and in some settings, the most adaptive response to high testosterone may be heightened caution rather than recklessness. The researchers’ own conclusion captures the state of the field well: testosterone’s effects are complex and often unpredictable.