• PDF / 1,348,624 Bytes
• 29 Pages / 439.37 x 666.142 pts Page_size
• 0 Downloads / 196 Views

DOWNLOAD

REPORT

---

Prosodic Disambiguation of Morphological Ambiguities in Turkish Nazik Dinçtopal Deniz1 

© Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract This study investigates the production and processing of lexical prosody in morphological ambiguities in Turkish. Native speakers of Turkish took part in two read-aloud and two lexical decision experiments. The results showed that in speaking, for both genuine and pseudo words that contrasted in stress, participants changed the fundamental frequency (F0) and intensity to disambiguate; and they changed duration (but not F0 or intensity) to disambiguate words and pseudo-words that did not contrast in stress. In listening, the participants were sensitive to the prosodic (mis)match in stress-contrasting pairs, but not to durational (mis)match presumably because the durational differences between the comparison pairs were shorter than perceivable. The findings show that Turkish speakers use prosody to disambiguate morphologically ambiguous word pairs and that they are sensitive to prosodic cues (at least to those used in stress contrast) when they hear them. Their behavior for pseudo-words suggests that they do so not on the basis of individual word knowledge but productively. The comparison pairs in the current study were segmentally identical, allowing us to attribute the observed prosodic variation only to the morpho-syntactic structure of the ambiguous pairs. Keywords  Turkish · Morphological processing · Prosody · Morpho-syntax

Introduction There has been an increased interest in the role of prosody in language processing recently. Although the role of prosody has now been widely investigated in sentence processing (Deniz and Fodor 2017; Kjelgaard and Speer 1999; Marslen-Wilson et  al. 1992; Nagel et al. 1996; Speer et al. 1996; Stoyneshka et al. 2010 among many others), research that examines the role of prosody in morphological processing has been limited, but informative. Previous research has shown that prosodic information in the form of lexical stress is represented autonomously (Cappa et al. 1997; Laganaro et al. 2002; Levelt et al. 1999; Miceli and Caramazza 1993) in the left hemisphere (Arciuli and Slowiaczek 2007) and its computation may involve a direct or an indirect route depending on its regularity (Cappa * Nazik Dinçtopal Deniz [email protected] 1



Department of Foreign Language Education, Boğaziçi University, 34342 Bebek, Istanbul, Turkey

13

Vol.:(0123456789)



Journal of Psycholinguistic Research

et al. 1997; Laganaro et al. 2002; Miceli and Caramazza 1993). It was also shown that morphological complexity of words affected their prosody in the form of duration (Seyfarth et al. 2018) and ambicategorical words were disambiguated with prosodic cues in speech (Conwell 2017). Research examining the processing of word-level prosody showed that the prosodic structure of a temporarily ambiguous word informed the human language processing mechanism in its parsing decisions (Davis et al. 2002; Isel et al. 2003; Kemps et al.