| Moment-to-moment fluctuation in performance (i.e. across-trial inconsistency) was examined in 2-choice reaction time data in three separate samples of participants: healthy controls ranging in age from 6 to 81 years, and both children and adolescents with Attention Deficit/Hyperactivity Disorder (ADHD). A methodological approach was employed that allowed for the examination of inconsistency while controlling for the effects of practice, fatigue, and group differences in mean level of performance. Among healthy controls, a U-shaped curve defined the relationship between age and inconsistency with optimal performance found in those aged approximately 20 years old. In children (aged 6--12 years old), those with ADHD were significantly more inconsistent than controls, and inconsistency was related to ratings of symptoms of ADHD both at home and school. Group differences were also observed in adolescents (aged 12--17 years old) with ADHD, however, in general, differences were observed only in those participants who also had reading difficulties (RD). Inconsistency was also examined separately in each end of the reaction time distribution. Evidence for a specific effect selectively affecting the slow portion of the distribution was found across all three samples. However, in addition, there was also evidence for effects that were general to both the fast and slow portions of the distribution (in the normal population in those age 6 to 20 years), as well as evidence for an effect selectively affecting the fast portion of the distribution (in adolescents with and without ADHD). The findings indicate that in addition to traditional outcomes of interest (e.g. mean level of performance), moment-to-moment fluctuation in cognitive performance is an important phenomenon which should be taken into account in future research in developmental psychology and psychopathology. |
