Poster Session B, Sunday, March 24, 8:00 – 10:00 am, Pacific Concourse
Semantic training affects online formation of memory traces for novel morphology
Viktória Roxána Balla1, Yury Shtyrov1,2,3,4, Miika Leminen1,3, Alina Leminen1,3; 1University of Helsinki, 2Higher School of Economics, 3Aarhus University, 4Saint Petersburg State University
Learning to recognize morphemic boundaries is crucial for fluent language use. The question of morphological learning is especially relevant in languages with a rich morphology, such as Finnish. Neurocognitive studies propose separate systems for decomposition and storage, which are flexibly used during the processing of polymorphemic inflections and derivations. Nevertheless, neural mechanisms underlying the acquisition of novel morphology remain unexplored. To address this question, we trained 19 native Finnish-speaking participants with new derivational suffixes through a word-picture association task. Following the short training session we used magnetoencephalography (MEG) to record the participants’ brain responses to trained and untrained suffixes combined with real and pseudoword stems. We compared event-related fields recorded during the first and last 4 minutes of the passive listening task to assess the rapid online learning of novel suffixes. We found a response increase in the left frontal and temporal sensors for the trained suffixes compared with the untrained at the 55-85 ms, 100-140 ms and 180-260 ms time-windows following the suffix onset. This response increase suggests that a short semantic training of novel affixes can facilitate morphological decomposition and speed up suffix memory trace formation. At the same time, the left temporal sensors showed enhanced effects for untrained suffixes at 180-260 ms towards the end of exposure, suggesting the online formation of memory traces even without previous semantic training. Overall, our findings suggest immediate formation of memory representations for novel affixes, with a facilitative effect of semantic training on morphological parsing.
Topic Area: LANGUAGE: Semantic