For decades, major depressive disorder was attributed to a deficit in monoamine neurotransmitters. Clinical latency of tricyclic and selective serotonin reuptake inhibitors, high non-response rates, and inconsistent genetic findings challenged this view and redirected research toward downstream biology. Preclinical work revealed that chronic stress triggers dendritic and spine loss in hippocampus and prefrontal cortex, whereas all effective treatments–including slow-acting monoaminergic drugs, rapid-acting ketamine, electroconvulsive therapy, and aerobic exercise–restore synapse number and function through brain-derived neurotrophic factor, TrkB, and mTOR signaling. Human connectomic studies then reframed depression as a disorder of mis-timed large-scale networks; targeted neuromodulation of nodes intrinsically anticorrelated with the subgenual cingulate provides proof of concept. Parallel findings in immunology and gut–brain science show that psychosocial stress, peripheral cytokines, and metabolic cues converge on the same plasticity pathways, dissolving the historical boundary between “reactive” and “endogenous” depression. Ketamine crystallizes this multiscale model: within minutes it induces dendritic spine formation, normalizes default-mode and limbic connectivity, and relieves symptoms within hours. We synthesize these lines of evidence into a framework of precision synaptic psychiatry, in which pharmacological, neuromodulatory, and lifestyle interventions are selected according to biomarkers that index glutamatergic tone, inflammatory load, or network dynamics. Future therapeutics will be judged less by the neurotransmitters they influence and more by their capacity to restore flexible, resilient brain circuitry.