Mental effort plays an important role in theoretical models of ADHD; it is central to both motivational and cognitive-energetic accounts, but it has been underspecified and under-examined empirically. It could however, provide the missing link between task performance of difficult and challenging tasks requiring cognitive control, and the effects of extrinsic rewards on performance in ADHD. In this exploratory study we examined two putative effort-related deficits in ADHD; First, impairments in the ability to apply effort and second, alterations in effort-related decision making due to an altered cost-benefit analysis relating to the application of effort. For this purpose, we examined 1) performance on an N-back task with five levels of difficulty (1-, 2-, 3-, 4-, 5-back) and choice behavior on a subsequent effort discounting task, 2) pupil dilation during the N-back task, and 3) subjective measures of mental demand and effort application during the N-back task. Fifteen boys with ADHD and 16 controls (aged 12-17 years) participated in this study. Results showed that with increasing level of N-back difficulty, performance deteriorated, and participants with ADHD performed significantly worse than controls, independent of level of difficulty. In the effort-discounting task, however, no main effect of group was found, indicating that adolescents with ADHD did not devalue effort-related rewards more than controls. This suggests that they are not more effort averse than controls. The overall pattern of pupil responses during the N-back blocks looked similar for ADHD and control participants: there was an initial increase in pupil dilation in the first 3-6 trials of a task block, after which a gradual decline in pupil response occurred back to baseline. In the control group, the peak of this pupil dilation increased with level of difficulty up to 3-back, while in the ADHD group such an effect was absent. Finally, groups did not differ in the subjective experience of the mental demand of the different levels of the N-back task, nor in effort application. Together, these results suggest that performance decrements in ADHD in a working memory task such as the task used here might be related to a suboptimal application of mental effort, which is not reflected in subjective experiences, and subsequently does not appear to lead to an altered cost-benefit analysis related to the application of effort.