Early Response to a Fake Oxytocin Spray Predicts Who Stays Expectation-Driven Months Into Real Antidepressant Care

A German hospital study that paired a single deceptive placebo test with three months of real antidepressant treatment finds that how strongly a patient responded to the sham dose predicted how strongly their later mood tracked their own expectations, suggesting placebo responsiveness behaves like a stable trait rather than a lab fluke.

A one time dose of saline sprayed up the nose and labeled as oxytocin, a deceptive placebo, produced acute mood and perception shifts in adults with major depression that went on to predict how strongly the same patients responded to their own expectations of relief during real antidepressant treatment months later, according to a study published in Translational Psychiatry on May 13, 2026. The finding, from a team led by Eun Jin Shim and Stefanie Brassen at the University Medical Center Hamburg-Eppendorf, is among the first to connect a single laboratory placebo manipulation directly to the trajectory of expectation effects in the same people as they were actually treated.

The researchers first ran 61 adults with major depressive disorder (31 women, 30 men) through a crossover session in which each person received, on separate occasions, a nasal spray falsely described as oxytocin and an identical saline spray described accurately. The sham oxytocin raised mood ratings and shifted an emotion classification task toward reading ambiguous faces as happier (F(1,60) = 15.25, P less than 0.001). Forty five of these participants then began antidepressant treatment as inpatients and were followed for up to three months, reporting weekly how much they expected the treatment to help alongside standard depression symptom scores (the Beck Depression Inventory). Thirty one completed the full three month follow-up. Across that period, a patient's expectation rating in one week statistically predicted the size of their symptom improvement the following week (beta = 2.24, 95 percent CI 0.66 to 3.82, P = 0.007).

The paper's central claim is a link across those two settings: patients who had shown the strongest mood and perceptual shift during the one time sham oxytocin session also showed the strongest relationship between weekly treatment expectations and later symptom reduction during actual antidepressant care. The authors report this as a positive correlation between an individual's acute experimental placebo responsiveness and their longitudinal clinical expectation effect, framing placebo responsiveness as a trait that generalizes across different treatments and time windows rather than a one-off reaction to a single deceptive spray.

Independent researchers will likely note that the clinical arm was not placebo-controlled: all 45 patients received real antidepressant treatment, so the study shows an association between expectation and symptom change within that group, not a test of placebo against drug. The deceptive design, standard in this line of research, required debriefing participants afterward, and the authors themselves describe the work as establishing an experimental model for predictors of expectation effects rather than a finished tool for clinical use.

The authors say their approach, pairing a brief lab placebo test with months of real-world monitoring, could help identify which patients are most sensitive to expectation effects, but they call for replication in larger and more diverse samples before any such screening could inform how antidepressant care is delivered.