"People with social anxiety have too much serotonin, not too little." This sentence is going around social media again, often with the idea that the medicines prescribed for decades were built on a mistake. The studies behind it are real and serious. But they measured something narrower than the headline suggests, and the same research team has since published trials that tell a more interesting story.
- What is true: two brain imaging studies from Uppsala University (Sweden) found a more active serotonin system in some brain regions of people with social anxiety disorder.
- What is overstated: the studies were small, measured production capacity rather than "amounts" of serotonin, and cannot say whether this is a cause or a consequence of anxiety.
- What it does not mean: it does not show that serotonergic antidepressants are useless or harmful. Never stop or change a treatment without your doctor.
Where the claim comes from
In June 2015, Uppsala University announced a study published in JAMA Psychiatry under the headline "Individuals with social phobia have too much serotonin, not too little". The opening line of the viral post is a quote from the first author, Andreas Frick, in that press release: "Serotonin can increase anxiety and not decrease it as was previously often assumed."
The idea that anxiety and depression come from a "lack of serotonin" was always a simplification. It became popular because the most prescribed antidepressants, the selective serotonin reuptake inhibitors (SSRIs), increase the serotonin available between nerve cells. So a finding in the opposite direction made headlines, and still does, eleven years later.
What the 2015 study actually did
The researchers used positron emission tomography (PET), a scan that follows a slightly radioactive tracer in the brain. Two tracers were used:
- one measures how fast the brain can make serotonin (18 people with social anxiety disorder, 17 people without);
- the other measures the serotonin transporter, the pump that brings serotonin back into the nerve cell (26 other patients, compared with the same group of controls).
People with social anxiety showed a higher synthesis capacity in several regions, including the amygdala, a structure involved in fear, as well as the hippocampus and the anterior cingulate cortex. They also had more transporters in several regions. According to the press release, the higher the production, the stronger the social anxiety. The authors concluded that the serotonin system "upstream" of the synapse is overactive in this disorder.
A second article by the same team, in 2016, tried to reproduce the finding in a separate group of 18 patients, this time with only 6 controls, and looked at a variant of a gene involved in making serotonin (TPH2). The rise was clear in the hippocampus and the basal ganglia; in the amygdala and the cingulate cortex it only appeared with a less strict statistical threshold. The viral post quotes both articles, but swaps their references.
What "too much serotonin" does not mean
These are careful studies, but the headline goes further than the data:
- Capacity is not quantity. PET measured how fast serotonin can be produced and how many transporters there are, at rest, not how much serotonin is present at a given moment.
- Small samples. 18 against 17 in the main study, 18 against 6 in the second. Results from groups this size need to be confirmed.
- A single snapshot. The scans were taken once. They cannot tell whether this activity causes the anxiety, results from years of anxiety, or both.
- One disorder only. The studies concern social anxiety disorder, not anxiety in general, and even less depression.
The picture also became more complex. In 2021, the same group compared 27 patients with 43 controls using two tracers. With the strictest statistics, the serotonin transporter was higher in only one region (the nucleus accumbens), and only the dopamine transporter was linked to the severity of symptoms; with a less strict threshold, both transporters were higher in several regions and linked to severity. The authors conclude that serotonin and dopamine are both involved. Serotonin is one piece of the puzzle, not the whole explanation.
So why do serotonergic antidepressants help?
This is where the story gets interesting, because the Uppsala team itself went on to test treatments.
The overall evidence. The efficacy of SSRIs in social anxiety disorder was not deduced from the "low serotonin" theory: it was measured in clinical trials. A Cochrane review (2017) gathered 66 trials and 11,597 participants, all medicines combined. For SSRIs (24 trials, 4,984 participants), people on treatment were about 1.65 times as likely to respond as people taking a placebo, although the reviewers rated the certainty of this evidence as very low. SSRIs were also the only class of medicine shown to reduce relapse (moderate certainty).
Serotonin went down, including with placebo. In a small randomised trial (2016), 18 patients received an SSRI, an experimental molecule or a placebo for six weeks. In all three groups, serotonin synthesis in the amygdala decreased, and this decrease went hand in hand with fewer symptoms. The authors see this as support for the idea that serotonin in the amygdala feeds anxiety. But the placebo group showed the same change, so the drug is not the only route.
What you expect matters. In a 2017 trial, 46 patients took the same SSRI for nine weeks. Half were told the truth; the others were told they were receiving an "active placebo". Response rates were 50% in the first group and 14% in the second, with the same medicine. In a 2021 trial with the same design (27 patients), the blockade of the serotonin transporter was high and similar in both groups, and it was not linked to the improvement. The authors conclude that this blockade alone is not enough to explain the clinical response.
In a 2025 perspective article, the team sums up: the effect of SSRIs in social anxiety cannot be reduced to serotonin reuptake; expectations and brain circuits shared with the placebo effect also play a part. That is very different from saying the medicines were "based on a false assumption" and should be dropped.
What this means for you
- If you take an antidepressant, do not stop it on your own after reading a post. The NHS advises talking to your doctor first and reducing the dose slowly, usually over several weeks or months, because stopping suddenly can cause withdrawal symptoms.
- Medication is not the only option. In the UK, the NICE guideline recommends, for adults, an individual cognitive behavioural therapy designed specifically for social anxiety as the first choice.
- The science is moving. These studies change how researchers explain anxiety. They do not yet change what works for patients: that is judged on trials, not on brain scans.
If you recognise yourself in social anxiety (fear of being judged, avoiding speaking in public, dreading social situations), the social anxiety test (SPIN) takes about 6 minutes for its 17 questions. For more general worry, the anxiety test (GAD-7) takes about 3 minutes. These are self-assessment questionnaires, not diagnoses: the result is a starting point for a conversation with a doctor or a psychologist. To go further on brain chemistry, read our articles on serotonin and the gut-brain axis and on the GABA-glutamate balance.
References: Frick A, Åhs F, Engman J, et al. (2015). Serotonin Synthesis and Reuptake in Social Anxiety Disorder: A Positron Emission Tomography Study. JAMA Psychiatry 72(8):794-802. Cross-sectional PET study, 18 patients (synthesis) and 26 patients (transporter), 17 controls analysed for each | Uppsala University (17 June 2015). Individuals with social phobia have too much serotonin, not too little. Press release | Furmark T, Marteinsdottir I, Frick A, et al. (2016). Serotonin synthesis rate and the tryptophan hydroxylase-2: G-703T polymorphism in social anxiety disorder. Journal of Psychopharmacology 30(10):1028-1035. Separate group, 18 patients, 6 controls | Frick A, Åhs F, Appel L, et al. (2016). Reduced serotonin synthesis and regional cerebral blood flow after anxiolytic treatment of social anxiety disorder. European Neuropsychopharmacology 26(11):1775-1783. Randomised double-blind trial, 18 patients | Hjorth OR, Frick A, Gingnell M, et al. (2021). Expression and co-expression of serotonin and dopamine transporters in social anxiety disorder: a multitracer positron emission tomography study. Molecular Psychiatry 26(8):3970-3979. 27 patients, 43 controls | Faria V, Gingnell M, Hoppe JM, et al. (2017). Do You Believe It? Verbal Suggestions Influence the Clinical and Neural Effects of Escitalopram in Social Anxiety Disorder: A Randomized Trial. EBioMedicine 24:179-188. 46 patients | Hjorth OR, Frick A, Gingnell M, et al. (2021). Expectancy effects on serotonin and dopamine transporters during SSRI treatment of social anxiety disorder: a randomized clinical trial. Translational Psychiatry 11:559. 27 patients | Furmark T, Wahlstedt K, Faria V (2025). Revisiting the SSRI vs. placebo debate in the treatment of social anxiety disorder: the role of expectancy effects, neural responsivity, and monoamine transporters. Frontiers in Psychology 16:1531725. Perspective article | Williams T, Hattingh CJ, Kariuki CM, et al. (2017). Pharmacotherapy for social anxiety disorder (SAnD). Cochrane Database of Systematic Reviews 10:CD001206. 66 trials, 11,597 participants | NHS. Antidepressants: stopping or coming off antidepressants | NICE (2013). Social anxiety disorder: recognition, assessment and treatment (CG159)
