Anne Fleur Kortekaas-Rijlaarsdam

123 EFFECTS OF MPH AND REWARD ON MATH PERFORMANCE 6 conditions were randomized and feedback from a preceding trial could thus positively affect performance on a consecutive trial (irrespective of condition). To this end we used a paired sample t-test to test whether mean final achieved level was significantly higher than mean entry level. Further, we calculated Pearson correlations between performance on our math task and the child’s performance on a similar math task (Speeded Arithmetic Test: De Vos, 1992). Second, we tested whether the impact of motivation condition on achieved difficulty level differed between groups (ADHD versus TD) using GLMM analysis with condition and group as fixed factors and a group*condition interaction. The effects of treatment (MPH versus placebo) on achieved math level and the possible interaction between effects of MPH and motivation were tested with a similar model with condition and treatment as fixed factors and a treatment*condition interaction, and a random intercept to control for within-subject dependency of measurements. Participation of 63 and 62 children in the ADHD and TD group, respectively, was sufficient to guarantee .80 power (α=.05) to detect group differences and MPH effects (Twisk, 2013). We checked whether there were group differences in entry level as determined in the practice phase to see whether we needed to control for entry level in the analyses of the experimental phase data. Using an independent samples t-test, we compared entry levels between our ADHD (when treated with placebo) and TD group. Using a paired samples t-test we compared entry levels between treatment conditions (MPH versus placebo treatment) in our ADHD group. When significant, and relevant (as determined by the accompanying effect size), entry level was included as a covariate. R E SU LT S Two children in the ADHD group did not complete the trial due to adverse events unrelated to the intervention (see Kortekaas-Rijlaarsdam et al., 2017b), resulting in 63 children in our ADHD group and 67 children in our TD group. Group characteristics of the ADHD and TD group and group comparisons are shown in Table 6.2. There were no group differences in age and gender and as expected, our ADHD group showed more inattentive-, hyperactive-impulsive symptoms and lower IQ, all typical for children diagnosed with ADHD. Table 6.3 displays the dependent variables for both groups and both treatment conditions.