Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Implicit and explicit learning of Russian grammatical gender
Abstract
Implicit and rule-based learning of inflectional morphology corresponding to Russian grammatical gender agreement rules were compared in 82 Russian-naïve adult participants immediately after the training and three weeks thereafter. The learning was assessed with the grammaticality judgment task and a gender decision test. First, grammatical classification accuracy produced by the rule-based learning conditions was higher than that produced by implicit learning in both immediate and delayed tests, which demonstrates benefits of rule-based learning in accuracy and transferability to novel contexts. However, a dramatic deterioration of performance over time was observed in the rule-based learning condition, which did not occur in the implicit learning condition. This finding suggests that implicit learning has long-term retention benefits. Second, comparisons of performance from three exemplar presentation modes (auditory-and-visual, auditory-only, and visual-only) produced evidence in support of the auditory bias in implicit learning of naturalistic language stimuli---when exemplars displayed phonetic/orthographic dissociation, the underlying patterns were learned in accordance with the phonetic pattern of invariance, but not the orthographic. Third, it was also established by the error analysis that implicit learning in the two conditions that involved auditory presentation of exemplars produced the same acquisition patterns (overgeneralization of the Feminine gender and the poor learning of the Neuter) with those observed in the natural course of language acquisition by Russian children. Also, the learning patterns when training was exemplar-based were reflective of the cue-to-category predictive strength---more detectable and more frequently-occurring cues were acquired with more accuracy than more ambiguous and less frequent cues as also happens in the natural language acquisition. Next, in comparable implicit and rule-based learning conditions, response times were slower when learning was rule-based than when it was exemplar-based, which suggests that implicit learning offers greater fluency benefits in language learning. Finally, when learning was exemplar-based, a dissociation was observed between the accuracy of grammatical classifications in test performance and rule articulation accuracy. In the three implicit learning conditions, the above-chance grammatical accuracy on tests was coupled with little-to-zero awareness of the rules.
Subject Area
Cognitive therapy|Linguistics|Language arts
Recommended Citation
Kosheleva, Yelena A, "Implicit and explicit learning of Russian grammatical gender" (2006). ETD collection for University of Nebraska-Lincoln. AAI3215815.
https://digitalcommons.unl.edu/dissertations/AAI3215815