Does anyone else have a little bit of a problem with the assumption that the "benchmark test" is enough of a gauge of general intelligence to be determinative i

Does anyone else have a little bit of a problem with the assumption that the "benchmark test" is enough of a gauge of general intelligence to be determinative in studies like this? Additionally, the discussion is mostly revolving around the physical size of the hippocampus, but instead of measuring that (arguably a difficult task), they instead rely on a "benchmark test" and base their results on that alone. What are the indicators that the benchmark test and hippocampus size are closely correlated?