A new study finds that depression in older adults correlates with shrinkage in the hippocampus, the brain's memory center. Researchers are treating this correlation as if it were a revelation about how depression works—but it is not.
What it is instead is a finding we have been making and remaking for two decades, each time publishing it as fresh evidence of mechanism while systematically ignoring the one study that actually answered the question. In 2003, Yvonne Sheline at Washington University did something most depression neuroscience never does. She followed depressed patients over time as they received treatment.
She found that antidepressants—specifically SSRIs—restored hippocampal volume in people whose brains had shrunk. This was the moment causality resolved. The shrinkage was depression's effect, not its cause.
A researcher finds brain structure changes correlating with a psychiatric symptom. The correlation gets framed as a breakthrough biomarker—a potential mechanism, a new diagnostic window. Grants follow. Citations follow. The finding replicates across labs, each lab publishing the replication.
Sheline's work should have closed this chapter. Instead, the field has published essentially the same finding at least fifteen times since—same hippocampus, same depression, same shrinkage, each study framed as if causality were still uncertain. This is not carelessness, it is structural. A paper titled "Depression Causes Hippocampal Atrophy (Reversible With Treatment)" generates fewer grant cycles than "Hippocampal Shrinkage as Potential Biomarker for Depression Severity"—one closes a loop, the other opens a funding apparatus. Watch how your own field treats solved problems that have become economically useful to keep unsolved. The pattern is not unique to neuroscience.