Processing of native and nonnative inflected words: Beyond affix stripping

•Native speakers show additional processing costs for oblique-case inflected nouns.•Processing costs persist for zero-inflected oblique nouns with no affix to strip.•Decomposition costs for oblique-case inflected nouns are incurred during checking.•Nonnative speakers may use affix stripping to access the stem and underuse checking.•Nonnative sensitivity to the inflectional paradigm improves with proficiency.

Two auditory lexical decision tasks explore the role of case form (citation or oblique) and the type of inflection (overt or zero). In native speakers, the study reports an additional processing cost for both overtly and zero-inflected oblique-case nouns compared to the same nouns in the citation form. It is interpreted as the cost of checking the recomposed word within the inflectional paradigm rather than the cost of affix stripping, because there is no affix to strip in zero-inflected words. Conversely, nonnative speakers of Russian in Experiment 1 do not show additional processing costs either for case form or inflection type, which suggests that they do not process the morphological information encoded in the inflection. In Experiment 2, we add a new manipulation to the nonword condition such that the nonwords illegally combine real stems and real inflections to emphasize the need for processing the inflection. This time, nonnative speakers show additional processing costs for oblique-case nouns, and their sensitivity to case increases with proficiency, with only high-proficiency nonnative speakers demonstrating native-like sensitivity. We show that citation forms are processed faster than oblique forms regardless of inflection, and that nonnative speakers' engagement of morphological information is task and proficiency-dependent.