Abstract: This work examines how adult second language learners (L2ers) and adult heritage speakers (HSs) process contrastive focus (CF) in Russian. While adult L2ers acquire the target language after puberty, HSs are exposed to it during early childhood, and are bilingual in it and the language of the society (in this case, English). Despite the different types of exposure, both groups acquire the target language under conditions of reduced input and often do not acquire it fully (cf. Montrul 2008). When proficiency is held constant, L2ers and HSs often exhibit similar patterns of non-target performance (e.g., Montrul & Ionin 2012), yet HSs also have selective advantages on phonology and early-acquired structures (see Montrul 2012 for an overview). Russian, like English, uses pitch accent to mark the element in CF. Unlike English, Russian has flexible word order, and the focused element can occur either in its base position or be moved leftward. Accenting an element not in contrastive focus is not felicitous in Russian, just as in English. Target-like processing requires that learners be aware of the word order flexibility of Russian, attend to the discourse to determine which element is in CF, and pay attention to the prosodic contour. According to the Interface Hypothesis (IH, Sorace 2011), the external interface syntax/discourse is particularly challenging in both L2-acquisition and L1-attrition (but see Leal et al. 2018, 2019 for evidence of success in this domain in L2/HS Spanish). In the case of Russian, there is much evidence that both HSs (e.g., Polinsky 2007) and L2ers (e.g., VanPatten et al. 2012) overrely on canonical word order, but very little is known about how learners integrate word order with prosody and discourse. Research questions: (a) Do HSs and/or L2ers of Russian (dominant in English) successfully acquire the relationship between discourse, word order and prosody with regard to CF realization? and (b) Do HSs have a selective advantage over L2ers in this domain? Methodology: We tested 20 HSs and 21 L2ers of Russian in the U.S.; the two groups were closely matched in proficiency via an independent test, and ranged from intermediate to advanced. 43 Russian native speakers (NSs) served as controls. We used a bimodal AJT, in which participants read and listened to 120 short dialogues that tested a variety of information-structure configurations, and rated the appropriateness of the answer on a 1-to-5 scale. Given that L2ers have an advantage on written tasks, and HSs - on aural tasks (Montrul et al. 2008), bimodal presentation ensured that neither group was favored. 24 of the items tested CF, with 2 felicitous response types and 2 infelicitous ones, 6 tokens per response type. Token sets were distributed across 4 presentation lists in a Latin-square design. Results: The results were analyzed using mixed-effects models for ordinal data, with items and participants as random effects, and answer type (4 levels) and participant group as a fixed effect. All three groups rated the felicitous answers significantly above the infelicitous, but the distinctions were much smaller for the L2ers, who rated the infelicitous continuations significantly higher than the HSs did. Our findings indicate that (i) Both learner groups were sensitive to the relationship between word order, discourse and prosody; yet (ii) HSs were much more target-like than L2ers. Our results provide evidence against the IH, indicating that success at external interfaces is possible, and also point to the advantage conferred by early exposure, even when proficiency and dominant language are controlled for. Lukyanchenko & Gor (2011) found that Russian HSs have an advantage over L2ers on the perception of phonological contrasts; our findings suggest that this advantage extends to prosody. Future research should investigate whether L2ers' difficulties are at the level of perception of prosody, or at the level of integrating prosodic information with the syntax and discourse information.