In this pilot study, the speech of 12 adult native speakers of English with intermediate to intermediate-high proficiency in Spanish as a second language (L2) was analyzed to determine whether L2 learners rely on distributional information in the process of L2 speech learning and if so, if similar or dissimilar distributional patterns of sounds are more easily acquired. The parameter for (dis)similarity was set around the notion of structural position in combination with native language (L1) and L2 phonemic inventories. The results show that the subjects were consistently more successful in producing the phonemes with overlapping distributional patterns in L1 and L2 than phonemes whose distribution differed in L1 and L2 as well as novel L2 contrasts. (Contains 15 notes and 5 tables.)