Lexicogrammatical profiling of ASD: cognitive-functional mapping and diagnostic implications

Area of Science:

• Linguistics
• Psychology
• Neuroscience

Background:

• Previous research established a corpus of autism spectrum disorder (ASD) discourse and identified lexicogrammatical patterns distinguishing ASD from non-ASD discourse.
• High performance was achieved in distinguishing ASD discourse using computational models, with 46 significant discriminators identified.

Purpose of the Study:

• To examine 18 additional lexicogrammatical discriminators in ASD discourse.
• To map these discriminators onto cognitive-functional domains to clarify their diagnostic relevance.
• To refine the understanding of the language-cognition interface in ASD.

Main Methods:

• A Tag Linear Model was used to identify distinguishing lexicogrammatical features.
• Logistic regression with 10,000 bootstrap iterations established statistical significance.
• 18 previously unexamined discriminators were analyzed and mapped onto 12 cognitive-functional domains.

Main Results:

• Forty-six statistically significant discriminators (p < 0.05) were confirmed, with 18 analyzed in detail.
• Distinctive linguistic patterns were observed across domains, including reduced use of specific grammatical structures and resources.
• These patterns reflect constraints in abstraction, perspective-taking, and pragmatic orientation.

Conclusions:

• Lexicogrammatical patterns in ASD reflect domain-specific cognitive constraints.
• Mapping discriminators to cognitive-functional domains offers a linguistically grounded perspective on ASD neurocognition.
• Findings have implications for diagnostic and intervention research in ASD.