Most cognitive training programs are adaptive, despite limited direct evidence that this maximizes children's outcomes. This randomized controlled trial evaluated working memory training with difficulty of activities presented using adaptive, self-select, or stepwise compared with an active control. At baseline, immediately, and 6-months post-intervention, 201 Australian primary school children (101 males, 7-11 years) completed working memory tests (near and intermediate transfer) and the Raven's Standard Progressive Matrices, and caregivers completed the attention-deficit/hyperactivity disorder-Rating Scale-5 (far transfer). The intervention comprised ten 20-min sessions delivered in class. For each training condition, compared with the active control, there was no evidence of transfer immediately or 6-months post-intervention (negligible to small effects). This trial provides no evidence that adaptive working memory training maximizes children's outcomes.