Anne Fleur Kortekaas-Rijlaarsdam

121 EFFECTS OF MPH AND REWARD ON MATH PERFORMANCE 6 the motivation condition a correct answer resulted in rewarded feedback on 85% of trials (15% of correct trials were not followed by rewarded feedback). Rewarded feedback was given in this probabilistic manner to avoid that children would deploy certain calculation strategies reinforced in the motivation condition. If a certain addition calculation strategy was correct, partial feedback made sure that such a strategy was not automatically reinforced (there was no rewarded feedback in 15% of correct trials in the motivation condition). However, incorrect answers were not rewarded, in order to avoid that children would become frustrated by the task (e.g., feelings of unfairness). The effect of motivation, was manipulated by the color of the screen anticipating the math calculation. The tracking mechanism ran independently for both the motivation and neutral conditions to make sure that possible differences between the achieved difficulty levels represented the condition-specific reward related gain for each participant. The dependent variable was the difference in achieved math level at the end of the task between the motivation and neutral condition. Difficulty levels In mathematics, addition problem difficulty can be determined by different factors. The most commonly studied variable is problem size, which is the magnitude of the (total) sum (e.g. 2+2 is easier than 34+52). Reaction times as well as the number of errors increase when the problem size increases (Ashcraft, 1992; Dehaene, 1992). A second variable influencing problem difficulty is the number of carry operations (Widaman, Geary, Cormier, & Little, 1989), which is the mental operation of carrying a calculation to another decade (e.g. 2+7 has no carry operations, 8+5 has one carry operation and 67+78 has two carry operations). Third, similarity between numbers can decrease difficulty. Ties (2+2) are generally the easiest type of sums, but also sums with intermediate similarity (23+22) are perceived as being easier than sums consisting of unique digits (43+21) (Ashcraft, 1992). Fourth, formulation of the sum (i.e. the distribution of the digits) can increase or decrease difficulty, with A+B being perceived as easier (lower RTs, less errors) than AB+C. In this line of reasoning, AB+CD is perceived as easier than ABC+D (higher RTs, more errors). Using this theoretical framework, 12 different levels of problem difficulty (addition only) were created (see Table 6.1). Problem size was considered most influential, followed by the number of carry operations, followed by the formulation (Ashcraft, 1992; Widaman et al., 1989). Ties (2+2) or calculations with high similarity (e.g. 24+28) were excluded. Procedure The current study has been carried out in accordance with the Declaration of Helsinki ant was approved by the local ethics committee. Parents and children above the age of