SUMMARYPlacebo analgesia is a widely observed clinical phenomenon. Establishing a robust mouse model of placebo analgesia is needed for careful dissection of the underpinning circuit mechanisms. However, previous studies failed to observe consistent placebo effects in rodent models of chronic pain. We wondered whether strong placebo analgesia can be reverse engineered using general anesthesia-activated neurons in the central amygdala (CeAGA) that can potently suppress pain. Indeed, in both acute and chronic pain models, pairing a context with CeAGA-mediated pain relief produced robust context-dependent analgesia, exceeding that induced by morphine in the same paradigm. We reasoned that if the analgesic effect was dependent on reactivation of CeAGAneurons by conditioned contextual cues, the analgesia would still be an active treatment, rather than a placebo effect. CeAGAneurons indeed receive monosynaptic inputs from temporal lobe areas that could potentially relay contextual cues directly to CeAGA. However, in vivo imaging showed that CeAGAneurons werenotre-activated in the conditioned context, despite mice displaying a strong analgesic phenotype, supporting the notion that the cue-induced pain relief is true placebo analgesia. Our results show that conditioning with activation of a central pain-suppressing circuit is sufficient to engineer placebo analgesia, and that purposefully linking a context with an active treatment could be a means to harness the power of placebo for pain relief.