Morphology without semantics? The roles of vocabulary knowledge and language exposure in influencing the nature of lexical representations in a rote learning context

Morphology without semantics? The roles of vocabulary knowledge and language exposure in influencing the nature of lexical representations in a rote learning context

First Author: Siti Syuhada Binte Faizal -- Newcastle University
Additional authors/chairs: 
Ghada Khattab
Keywords: Arabic, Morphology, Lexical Decision, Vocabulary, Statistical learning
Abstract / Summary: 

Research on Arabic morphology has shown that discontinuous root and pattern morphemes are represented in Arabic speakers’ minds and play an important role in spoken and visual word processing. Given the intricate relationship between the semantic and morphological function of roots and patterns, this raises the question of whether non-concatenative morphological representation can take place without semantic representation. The unique context of rote learning of Arabic through Qur’anic memorisation - typical in Muslim communities in non-Arabic-speaking countries - offers a testing ground for the dissociation between semantic and morphological representations. We investigated whether non-Arabic-speaking Qur’anic memorisers implicitly gain morphological representations while reading and/or memorising the Qur’an, and thus whether they are primed by Qur’an Arabic roots in a lexical processing task. We also investigated whether the priming interacts with Qur’an vocabulary knowledge and amount of Qur’an memorised, thereby informing us of the roles of semantics and statistical exposure to the language in morphology. 153 Malay-English bilinguals (M.age = 15) were given a lexical decision task with a visual unmasked priming paradigm in which 26 orthographically and phonetically unambiguous targets were each paired with three different primes to generate three experimental conditions (+R+P, -R+P, Baseline; R=Root, P=Phonology). Linear mixed-effects regression analyses on RTs showed no statistical significance between the conditions, but interesting trends where the root priming effect is the largest in participants with more Qur’an vocabulary knowledge and memorisation, and minimal in participants with little vocabulary knowledge and memorisation. This supports the notion that root morphemes convey semantic information, and that semantic knowledge needs to be supported with statistical exposure to the language to strengthen morphological representations (and vice versa).