Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Does L2 morphological processing engage the same ventral pathway as L1? A masked priming fMRI study
Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai
This poster is part of the Sandbox Series.
Longyun Hu1, Xavreila Sheung-Wa Ng1, Enoch Yee-Lok Wan1, Matthew King-Hang Ma1, Marc F. Joanisse2, Manson Cheuk-Man Fong1; 1Hongkong Polytechnic University, 2The University of Western Ontario
Introduction. Visual word recognition models propose that morphologically complex words (e.g., teacher) are rapidly decomposed during early lexical access. In L1 readers, masked morphological priming consistently engages the ventral reading stream, particularly left occipitotemporal cortex (vOT), reflecting morpho-orthographic decomposition. However, the neural basis of this process has remained under-characterized in L2 learners: only two fMRI studies have directly addressed L2 derivational processing, with mixed findings on whether L2 readers recruit the same ventral mechanisms as native speakers or rely more on phonologically mediated pathways given reduced reading automaticity. Behavioral evidence regarding L2 processing is mixed: research indicates either native-like morpho-orthographic decomposition or a primary reliance on surface form over underlying morphological structure. The present study addresses this gap by comparing masked morphological priming in L1 and L2 English readers. Methods. L1 Cantonese–L2 English readers (N = 14 of target 24; mean age = 20) performed a masked primed lexical decision task on a Siemens 3T Prisma (TR = 800 ms, voxel size = 2 mm isotropic); an age-matched L1 English group will be recruited in London, Canada. Each trial presented a 66-ms masked prime followed by a target requiring word/nonword judgement. Five Condition × Relatedness pairings were tested: Transparent (speaker–SPEAK), Opaque (dormant–DORM), Form (corner–CORN), Phonological (selfish–CELL), Semantic (legend–MYTH). Departing from prior designs that grouped etymological with pseudo-morphological pairs under 'Opaque', Opaque pairs were etymologically related but semantically opaque; Form pairs were orthographically overlapping but etymologically unrelated. Phonological and Semantic pairings served as controls. Functional ROIs from an orthogonal word > nonword contrast were analysed with per-ROI Condition × Relatedness ANOVAs. Results. Data from the L2 group has been analyzed. Behaviorally, RTs showed a significant main effect of Relatedness, F(1,20.9) = 11.55, p = .003, and a marginal Condition × Relatedness interaction, F(4,181.6) = 2.29, p = .061. Posthoc t-tests revealed that priming emerged only for Transparent, p < .001, and Opaque, p < .001. Crucially, the presence of Opaque priming but an absence of Form priming is consistent with the process of morpho-orthographic segmentation being driven by morphological structure rather than surface overlap. Functional activation analysis revealed significant Condition × Relatedness interactions in the left ventral occipitotemporal cortex (vOT), F(4,52) = 3.12, p= .022, and left occipital pole, F(4,52) = 3.13, p = .022. The occipital pole showed form-overlap suppression, p = .042, whereas Form priming was absent behaviorally. The absence of behavioral form priming may be due to cancellation by facilitatory graphical process and inhibitory orthographical process. In the left middle frontal gyrus, semantic priming was significant, p = .047, while phonological priming approached significance, p = .091, potentially reflecting inhibitory phonological interference in L2 lexical access. Discussion. These preliminary findings favour an ortho-morphological decomposition in L2. A significant Opaque priming (behaviourally and in vOT) alongside an absent Form priming indicates morphemic parsing at the early visual-orthographic stage. The remaining data collection will enable direct between-group comparison to test whether L2 readers recruit these ventral morpho-orthographic mechanisms to the same extent as native speakers.
Topic Areas: Morphology, Reading