Development in the Bilingual Interactive Activation Model: Simulations of L1 and L2 acquisition

dc.contributor.advisorDijkstra, A.F.J.
dc.contributor.advisorSprinkhuizen-Kuyper, I.G.
dc.contributor.authorHaga, F.
dc.description.abstractThe word learning performance of monolinguals and early and late bilinguals was simulated using the Bilingual Interactive Activation (BIA) model by Dijkstra and Van Heuven (2002), a model for visual word recognition. Increasingly large lexicons with increasing frequencies were used to simulate language acquisition in seven discrete developmental steps. When a particular word from a second language (L2) was assumed to be learned in a particular developmental step, it was incorporated in a lexicon shared with words of the first language (L1). For all words in the integrated lexicon, the word recognition time cycles were simulated. Five simulation studies were done, compared to each other, and statistically analyzed. The languages for simulations were Dutch and English. It was found that learning an L2 had an influence on the word recognition performance of the already known L1. Learning English as L2 had a negative effect on processing Dutch as L1, but learning Dutch as L2 had a positive effect on processing English as L1. Knowing an L1 had a negative influence on acquiring a L2; and learning an L2 early was more efficient than learning it later. However, the observed influences were surprisingly small.en_US
dc.thesis.facultyFaculteit der Sociale Wetenschappenen_US
dc.thesis.specialisationBachelor Artificial Intelligenceen_US
dc.thesis.studyprogrammeArtificial Intelligenceen_US
dc.titleDevelopment in the Bilingual Interactive Activation Model: Simulations of L1 and L2 acquisitionen_US
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
229.42 KB
Adobe Portable Document Format