Neuropsychological assessment has long relied on magnitude-based statistics: correlation, effect size, and configuration metrics that quantify degree of difference while leaving distributional organization unexamined. These approaches serve severity estimation but may miss disruption to constraint architecture that is the regulatory network through which cortical regions shape cognitive output. This article develops a conceptual distinction between magnitude loss and behavioral network disruption, examining whether Kullback-Leibler (KL) divergence, which is sensitive to distributional structure, captures phenomena not detected by magnitude-based metrics. Secondary analysis of three published datasets, encompassing traumatic brain injury, dementia subtypes, a neurodevelopmental condition, and psychiatric comparison groups, revealed consistent directional asymmetry. Acquired neurological groups produced convergent KL profiles; neurodevelopmental attention deficit hyperactivity disorder (ADHD) did not, suggesting KL divergence may index constraint disruption rather than severity alone. Individually, KL divergence dissociated from magnitude-based metrics, indicating distributional differences even when correlations and effect sizes suggested moderate similarity. KL divergence appears sensitive to constraint architecture degradation rather than solely to magnitude of deviation. Characterizing patient-control differences through magnitude-based comparisons alone warrants reconsideration. KL divergence offers a statistical vocabulary for a clinical observation that has long preceded its measurement: acquired brain injury changes not only how much a system produces, but how it is organized.
James V. English (Thu,) studied this question.