The effect of differing scoring methods for the Tower of London task on developmental patterns of performance

The Tower of London (TOL) task is frequently used to assess executive functioning in both adults and children, although there remains considerable controversy over what it measures and how to score it. In this study we compare two scoring methods and find that correlations between them were high for 7-year-olds (.86), and dropped to a low of .47 for adults. These results demonstrate that the TOL necessarily has different construct validity in adults and children. Second, results of one method (based on errors only) show a developmental trend in performance from middle childhood to adulthood, while the other (based on errors and time) shows a developmental progression from ages 7 to 13, but not between 13-year-olds and young adults. Thus, scoring method influences the resulting developmental model.