Testing different metrics for syntactic and morphological complexity in aphasic discourse: A cross-genre comparison in Spanish
Abstract
Purpose: The aim of this study is to determine which of four tasks testing different discourse genres (descriptive, procedural, conversational, and narrative) is the most sensitive in detecting syntactic and morphological complexity deficits in Spanish aphasia. Additionally, we compare two approaches for measuring syntax in discourse to identify the most accurate set of metrics for discriminating group membership.
Method: In this comparative cross-sectional study, we analyzed the discourse of an aphasia group (N = 20) and a non-brain-damaged group (N = 14). Data was preprocessed using manual and automated procedures. We measured syntactic complexity through proxies (mean length of utterance, noun-verb ratio, and verbs per utterance), in addition to a competing ‘syntactic structures’ metric. The latter includes variables of subordination, left branching, relative clauses, and word integration, while morphological complexity was measured through an inflectional index.
Results: The descriptive and narrative tasks highlighted the most significant differences across syntactic complexity metrics between the groups. Additionally, nested binomial logistic regression models showed that the ‘syntactic structures’ metric predicted groups better than the proxies alone.
Conclusions: A comprehensive assessment of syntactic complexity can be obtained through the administration of both descriptive and narrative tasks. The measurement of ‘syntactic structures’ in discourse proved a fitter and more fine-grained approach for group discrimination than the incorporation of syntactic proxies alone. These findings provide guidance for a precise discourse assessment in Spanish aphasia.
Versions
Related
- A comparative analysis of syntactic complexity in English academic texts by Chinese MTI trainee translators and native academic writers
- Emended description of the genus Henriciella: morphological and phylogenetic analysis identifies two subgenera with different reproductive strategies
- Microbial Load and Morphological Characterization of Bacteria and Fungi in Dough Stored at Different Temperatures and a Contaminated Pickle Sample
- In vitro contact force evaluation of robotic magnetic navigation catheters; Head to head comparison of different catheters and different settings
- Metabolomics-based comparison of metabolite profiles in Capparis spinosa L. from different habitats and associated soil factors
BibTeX
@misc{chagoya2026testing,
title = {{Testing different metrics for syntactic and morphological complexity in aphasic discourse: A cross-genre comparison in Spanish}},
author = {Jorge Ricardo Hidalgo Chagoya and Xabi Ansorena and Roel Jonkers and Maddi Carrera and Ignacio Quemada Ubis and Simona Mancini},
howpublished = {PsyArXiv},
year = {2026},
month = nov,
doi = {10.31234/osf.io/dw2qa_v3},
url = {https://osf.io/preprints/psyarxiv/dw2qa_v3/},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Testing different metrics for syntactic and morphological complexity in aphasic discourse: A cross-genre comparison in Spanish
AU - Jorge Ricardo Hidalgo Chagoya
AU - Xabi Ansorena
AU - Roel Jonkers
AU - Maddi Carrera
AU - Ignacio Quemada Ubis
AU - Simona Mancini
PY - 2026
DA - 2026/11/02
PB - PsyArXiv
DO - 10.31234/osf.io/dw2qa_v3
UR - https://osf.io/preprints/psyarxiv/dw2qa_v3/
AB - Purpose: The aim of this study is to determine which of four tasks testing different discourse genres (descriptive, procedural, conversational, and narrative) is the most sensitive in detecting syntactic and morphological complexity deficits in Spanish aphasia. Additionally, we compare two approaches for measuring syntax in discourse to identify the most accurate set of metrics for discriminating group membership. Method: In this comparative cross-sectional study, we analyzed the discourse of an aphasia group (N = 20) and a non-brain-damaged group (N = 14). Data was preprocessed using manual and automated procedures. We measured syntactic complexity through proxies (mean length of utterance, noun-verb ratio, and verbs per utterance), in addition to a competing ‘syntactic structures’ metric. The latter includes variables of subordination, left branching, relative clauses, and word integration, while morphological complexity was measured through an inflectional index. Results: The descriptive and narrative tasks highlighted the most significant differences across syntactic complexity metrics between the groups. Additionally, nested binomial logistic regression models showed that the ‘syntactic structures’ metric predicted groups better than the proxies alone. Conclusions: A comprehensive assessment of syntactic complexity can be obtained through the administration of both descriptive and narrative tasks. The measurement of ‘syntactic structures’ in discourse proved a fitter and more fine-grained approach for group discrimination than the incorporation of syntactic proxies alone. These findings provide guidance for a precise discourse assessment in Spanish aphasia.
ER -