Back to search

National and Kapodistrian University of Athens

Orthographic and phonological representation and processing of lexical stress in the mental lexicon

Abstract

dc:description

Every word constitutes of its segmental and prosodic characteristics (Botinis, 2011). Lexical stress is a prosodic characteristic, referring to the prominence of one syllable of a word relative to the others (Ladd & Arvaniti, 2023). The acoustic correlates of a stressed syllable are increased duration and increased amplitude (Beckman, 1986; Laver, 1994). Lexical stress can enhance word disambiguation, especially in free-stress languages where stress position can vary (Arvaniti, 2020). In this case, shifting stress from one syllable to another in a given word can alter entirely a word meaning, as in the example of “γέρος” (ˈʝeros/old) και “γερός” (ʝeˈros/strong) in Greek that differ only in stress and their meaning is completely different. The current doctoral thesis aims at exploring lexical stress and more specifically the effects of stress in spoken word recognition and in silent reading. Furthermore, another goal was to test stress effects in the time course of lexical identification. Does stress affect word processing in the earliest stages of lexical access or does its effects become obvious in the latter stages of that process? The current study was conducted in Greek, a language that lends itself to the investigation of stress for the following reasons: (a) it is a free-stress language (Arvaniti, 2007) with the only phonological restriction of stress falling in one of the last three syllables of a word (Malikouti-Drachman & Drachman, 1989), (b) it is transparent language (Protopapas & Vlahou, 2009) with relatively small effects on vowel quality (Arvaniti, 2007; Fourakis, Botinis, & Katsaiti, 1999), and (c) the Greek language has an orthographic mark of stress placed on the vowel of the stressed syllable in every polysyllabic word (Protopapas, 2016). Regarding the effects of stress in spoken word recognition, several studies so far have demonstrated the effects of stress in word recognition, especially through the use of priming experiments (in English, Cooper et al., 2002; in Dutch, van Donselaar, Koster, & Cutler, 2005; in Italian, Tagliapietra & Tabossi, 2005; in Spanish, Soto-Faraco et al., 2001). In these experiments participants identified a target word faster when the auditory prime was a disyllabic fragment matching segmentally and prosodically to the following target, than in the condition of a neutral prime. Eye-tracking studies using the visual world paradigm further support the aforementioned findings suggesting that lexical stress affects spoken word recognition from the moment it becomes available in the acoustic signal αντιληπτός (in Dutch, Reinisch et al., 2010; in Italian, Sulpizio & McQueen, 2012). As a result, a primary goal of the current thesis was to check the replicability of the observed stress effects in Greek (chapters 1 and 2). In the domain of silent reading findings regarding the effects of stress are more limited and observed mainly in eye-tracking studies in English (Ashby & Clifton, 2005; Breen & Clifton, 2011, 2013). Moreover, in languages that bear an orthographic mark of stress in their orthography, a question rises concerning to what degree (if any) does the orthographic mark of stress affect visual word processing. Findings in this area are controversial since most studies designed to examine the effects of the orthographic mark of stress were conducted in Spanish (Gutierrez-Palma et al., 2021; Labusch et al., 2022; Marcet et al., 2021; Marcet & Perea, 2022; Schwab, 2015), a language where the stress mark is only of use in a limited number of cases (for a thorough description of the stress assignment in Spanish see Marcet and Perea, 2022). In the present doctoral thesis, it was attempted for the first time to investigate the processing of the orthographic mark of stress through both a possible parafoveal preview benefit before the eyes of the reader fixate on the word for the first time, as well as the silent reading of stress minimal pairs embedded in sentences while recording eye movements (chapter 3). A significant advantage of this study is the fact that changing the placement of the stress mark in a target did not constitute an orthographic error, rather the orthographic image of another lexical representation. Apart from the above, another equally important goal of the current thesis was the examination of the representation of stress in the mental lexicon and its relation to the segmental word characteristics. Levelt et al. (1999) in their model WEAVER++ proposed (a) the existence of abstract metrical frames and (b) the theory of stress Underspecification. According to stress Underpecification, only less frequent stress patterns in a given language are represented in the lexicon, while the most frequent (default) stress pattern is applied on words by rule. For this reason, it was attempted to observe “pure” stress priming in the absence of segmental overlap between prime and target (chapter 1), but also differences in the priming effects observed depending on whether the stress pattern under investigation was the default or a less frequent stress pattern (chapter 2). Overall, the findings of the present doctoral thesis confirm the effects of stress in spoken word recognition from the moment stress becomes available in the acoustic signal. In reading the effects of the orthographic mark of stress are lacking in the earliest stages of word recognition as indicated by the exploration of possible stress effects from the parafoveal preview of the initial syllable of the target word. Stress effects become evident though in the case of stress minimal pairs differing only in stress. Here, the stress mark affects significantly the access to a word’s meaning, by facilitating the process of reading when the stress mark matches sentence semantics, and by inhibiting the process when its placement on the word contradicts with sentence semantics. On the other hand, the absence of the stress mark demonstrates only minimal effects, possibly due to the lack of a complete coherent orthographic word presentation of the target. Finally, regarding the representation of stress in the lexicon, according to the current findings there is no evidence to support either the presence of abstract metrical frames or the theory of stress Underspecification. On the contrary, it is suggested that lexical stress (either the default or a non-default stress pattern) is fully specified in the lexicon as an integral part of the lexical representation.

Degree

thesis:*
Grantor dc:publisher
National and Kapodistrian University of Athens
Year dc:date
2024

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Andrikopoulou, Angeliki
  • Ανδρικοπούλου, Αγγελική

Subjects

dc:subject × 24

Rights

Language dc:language
eng

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/58934

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Andrikopoulou, Angeliki; Ανδρικοπούλου, Αγγελική. Orthographic and phonological representation and processing of lexical stress in the mental lexicon. National and Kapodistrian University of Athens, 2024. http://hdl.handle.net/10442/hedi/58934