To further investigate the usefulness of 3 purported measures of executive function (EF) in head injured children, we administered the Twenty Questions Test (TQT), Tower of London (TOL), and the Wisconsin Card Sorting Test (WCST) to 151 children who had sustained a closed head injury (CHI) of varying severity about 3 years earlier. In addition, we tested 89 normal controls. Fifty-seven of the patients were included in a longitudinal study that compared performance at 3 months and 36 months. All of the head injured children underwent magnetic resonance imaging for investigational purposes. Severity of CHI, as defined by the lowest Glasgow Coma Scale (GCS) score, affected performance on all 3 EF measures. Focal lesion volume incremented prediction of performance on TOL and WCST, but not TQT. Moderate intercorrelations of the test variables were obtained. Although all three EF measures depicted changes in performance over 3 years, a ceiling effect detracted from the sensitivity of the TOL to the impact of CHI on development. Implications of the findings for clinical applications are discussed.