Garagnani, Shtyrov and Pulvermüller (2009) showed that strong attention was necessary for adults with normal language performance to allow them to memorize new words. In the case of low attention, the subjects could not remember unknown words but only known words. The link between attention and learning new words could explain the comorbidity between attention disorders and language impairment in young children. We wanted to test whether young children with or without SLI were able to learn new words with or without attention to the task. Our hypothesis was that learning new words would be especially difficult for children with SLI, which means that their performance would drop more when learning new words without attention than in any other situation. Ten children with SLI were matched by age to ten children with typical language development. Age ranged from 6;9 to 10;03. Children were presented during training with lists containing four non-words and four real words. Training was conducted under attention or without attention. Testing contained three non-words distractors and three real words distractors in addition to the words presented during training. Results showed that control children had better results than children with SLI, t(317) = 2.97, p = 0.003, that results under attention were better than results without attention, t(317) = 3.69, p = 0.0002, and that real words were better memorized than non-words, t(317) = 4.31, p < 0.0001. There was an interaction effect between Children's type (Ctrl vs. Sli) and Attention (with vs. without), but only with real words, F(1,76) = 5.88, p = 0.01, not with non-words. Our main hypothesis was only partly confirmed, but results showed that children with SLI had clear problems with learning non-words and learning words under no attention. Full results will be presented and discussed during the presentation of the paper. Garagnani, M., Shtyrov, Y., & Pulvermüller, F. (2009). Effects of attention on what is known and what is not: MEG evidence for functionally discrete memory circuits. Frontiers in human neuroscience, 3, 10.