Aristotle University Of Thessaloniki (AUTH)
The processing of non-canonical structures in Greek
Abstract
dc:descriptionThis dissertation investigates the processing and comprehension of relative clauses (RCs) in Greek, with a particular focus on similarity effects and the use of heuristics. The investigation of similarity effects is framed within featural Relativized Minimality (fRM, e.g., Rizzi, 2004, 2018) and similarity-based interference (SBI) accounts (e.g., Gordon et al., 2001; Lewis et al., 2006), both predicting that object relative clauses (ORCs) will cause greater difficulty when the head and the embedded DP match in features. The use of heuristics is framed within the Good Enough (GE) processing approach (e.g., Ferreira, 2003), which expects that the parser might resort to fast and frugal heuristics, such as the plausibility heuristic, during processing and comprehension. The study also examines how individual differences in cognitive and (meta)linguistic abilities modulate RC processing and comprehension. The research comprises two main studies. Study 1 employed a self-paced reading task to investigate number and pragmatic similarity effects in ORCs in a group of 42 adult speakers. Study 2 consisted of two eye-tracking experiments examining both subject relative clauses (SRCs) and ORCs manipulated for plausibility. Experiment 1 compared 40 university students with humanities background to 40 with non-humanities background, while Experiment 2 compared 50 adults to 50 adolescents (mean age: 14.06). Across all experiments, participants completed tasks measuring working memory, cognitive control, reading fluency, and metalinguistic abilities, namely morphosyntactic awareness, knowledge of metalanguage and ambiguity awareness. Results from Study 1 revealed both number and pragmatic similarity effects during online processing, with the latter appearing only on the RC DP. In offline comprehension, there was an interaction showing that number match disadvantage appeared only in pragmatic match conditions. Overall, number similarity effects are compatible with both fRM and SBI accounts, whereas pragmatic similarity can only be accounted for by SBI models. Study 2 demonstrated an ORC disadvantage across experiments and illustrated the use of the plausibility heuristic by both adults and adolescents, in line with the GE processing approach. Regarding individual variation, non-humanities participants were slower than humanities ones and sometimes exhibited greater sensitivity to syntactic manipulations. Similarly, adolescents showed increased processing times and lower accuracy than adults, as well as a greater influence by syntactic information. Working memory alleviated the online ORC disadvantage for humanities participants. Finally, reading fluency exerted a positive effect on participants’ online and offline behavior, while metalinguistic abilities either facilitated or inhibited their performance.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Douka, Gerakini
- Δούκα, Γερακινή
Subjects
dc:subject × 14- Προτασική επεξεργασία
- Χρονομετρικά πειράματα
- Αναφορικές προτάσεις
- Ατομικές διαφορές
- Sentence processing
- Online experiments
- Relative clauses
- Individual differences
- Γλώσσα και Λογοτεχνία
- Ανθρωπιστικές Επιστήμες και Τέχνες
- Γλώσσα και Γλωσσολογία
- Languages and Literature
- Humanities and the Arts
- Language and Linguistics
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/59005