Test-retest reliability of a new delay aversion task and executive function measure.

Despite the wide adoption of measures of executive functions and motivational tendencies in studies of developmental disorders and child psychopathology, few studies have investigated their test-retest reliability. The present paper examines the reliability of a new measure of delay aversion, three measures of working memory, a response inhibition measure and a measure of dual task performance. The children, aged between 7 and 15 years, performed the tasks twice, with a 2-week period in between the sessions. Using a relatively conservative criterion, only the delay aversion task and one of the working memory measures (delayed response alternation) demonstrated satisfactory test-retest reliability. The other two working memory measures (sentence span and counting span) showed modest reliability. For the inhibition measure (stop task) the results were mixed, with poor to modest reliabilities obtained for the various derived measures. The dual task failed to demonstrate adequate test-retest reliability. These differential reliabilities need to be borne in mind when interpreting the results of studies using these measures. In particular the effect of low reliability on statistical power and the Type II error rate should be considered.