Opposite brain sides gave higher consecutive hippocampal change covariance in 176 selected DLBS 1.3.0 adults with annual log rates

We tested whether consecutive hippocampal change covariance depends on brain side pairing. The test used 176 selected adults in OpenNeuro ds004856 version 1.3.0. Each score was an annual log volume rate. The same-side mean covariance was -2.378. The opposite-side mean covariance was +1.157. Their paired gap was +3.534, with a 95% person bootstrap interval of 1.453 to 6.463. The unit is (log percentage points/year)^2. All five fixed checks gave gap intervals above zero. These checks included ranks and removal of the two largest rates. An exact proof shows that the gap equals minus twice the covariance of changes in half the log left-to-right volume ratio. Synthetic examples show that measurement error and true change can give identical observations. Thus the gap does not identify measurement error or a neural cause. The local plan followed review of a known bilateral result. It was not externally registered. The package includes source records, a reusable NumPy tool, code, all bootstrap samples, results, figures, and run instructions. Fresh downloads and a complete replay matched the preserved files. A separate audit checked all 120,000 bootstrap draws.

hippocampus · longitudinal MRI · measurement error · covariance · brain aging · side pairing