Placebo Pain Relief Is Not a Fixed Trait, Five Session Study Finds

A trial that tested the same sham analgesic on 30 healthy adults five times over two months found that how much relief each person felt shifted significantly session to session, rather than holding steady, raising questions for any protocol that leans on a placebo effect lasting weeks.

A placebo cream produced measurably less pain in volunteers than an identical-looking control cream overall, but the size of that relief changed significantly from one test session to the next, according to a study of 30 pain-free adults led by Lewis S. Crawford and colleagues, published in Communications Biology on May 28, 2026. Each participant underwent five conditioning and test sessions roughly two weeks apart, a span of about eight weeks that matches the length of many clinical drug trials.

Using repeated thermal pain stimuli and brain imaging, the researchers reported a significant interaction between cream type and session number (F[4,116] = 2.68), meaning placebo responses differed between sessions, even though a main effect of time alone, independent of the placebo, was not significant (F[4,116] = 2.19, p = 0.07). When they sorted individuals by how often they showed a measurable placebo response across the five visits, 14 of the 30 participants (46.7 percent) responded consistently in four or five sessions, 9 (30 percent) never showed a significant placebo effect, and 7 (23.3 percent) swung between responding and not responding from one session to another.

Functional MRI scans linked repeated placebo exposure to changing activity in brain regions tied to cognitive evaluation and pain modulation, including the dorsolateral prefrontal cortex, but these changes were confined to the group that responded consistently. The researchers also found that how a person responded in their very first session predicted their later pattern better than any psychological questionnaire did, a finding that argues against treating placebo responsiveness as a fixed trait that could be screened for in advance.

The authors frame the work as a caution for trial design generally: anywhere a treatment's measured benefit might rest partly on expectation, from conventional drug trials to open-label approaches that tell patients outright they are receiving an inactive treatment, the assumption that a placebo response captured once will hold steady over weeks of repeated use is not supported by this data.

The researchers call for longer, more diverse cohorts before the pattern they describe in healthy volunteers can be assumed to apply to people managing chronic pain or other symptoms day to day.