On the role of syllabic neighbourhood density in the syllable structure effect in European Portuguese

Abstract

Previous lexical decision masked priming studies have shown that the advantage of syllable-congruent primes over syllable-incongruent primes is observed for CV (e.g., JU.ROS [interests]), but not for CVC first-syllable words (e.g., TUR.BO [turbo]), giving rise to the so-called syllable structure effect (e.g., ju.ral-JU.ROS < jur.ga-JU.ROS; tur.ta-TUR.BO = tu.res-TUR.BO). This effect is puzzling since it is not accounted for either by the distributional frequencies of CV and CVC syllables in European Portuguese (EP) or by syllable complexity. Here we examine whether the number of words of the same syllabic length sharing the same (first) syllable in the same (first) position, a measure taken as an index of syllabic neighbourhood density, may account for the syllable structure effect. To that purpose, 36 EP skilled readers performed a lexical decision masked priming task in which 48 CV and 48 CVC words matched in the number of syllabic neighbours, amongst other variables, were preceded by syllable-congruent (e.g., ju.ral-JU.ROS and tur.ta-TURBO), syllable-incongruent (e.g., jur-ga-JU.ROS and tu.res-TUR.BO), and unrelated primes (e.g., pu.cas- JU.ROS and binva-TURBO). Syllable priming effects were still observed only for CV words, even though CVC words with a CV phonological structure (e.g., PEN.TE[comb] - /p'eti/) tended to behave similarly to CV words, suggesting that EP syllable effects may be driven by phonological factors.