Lieberman, Borovsky, Hatrak, and Mayberry (2015) used a modified version of the visual-world paradigm to examine the real-time processing of signs in American Sign Language. They examined the activation of phonological and semantic competitors in native signers and late-learning signers and concluded that their results provide evidence that the mental lexicon of late learners is organized differently from that of native signers. In particular, they claimed that late-learning signers, in contrast to native signers, do not activate phonological competitors during the real-time recognition of spoken words. I argue that this claim receives no substantive support from the data and the inferential statistics.