In this Research Topic we wish to take a critical look at fundamental assumptions about language architecture and about sub-components of the system, and to examine the validity of these assumptions for language modeling. Architectures of the language system have been postulated from varying perspectives, including theoretical linguistics, cognitive science, neuroscience, and artificial intelligence. These perspectives come in different flavors, focusing for instance on rules, constraints, conditional probabilities, or brain localization, and they incorporate varying degrees of non-linguistic, domain-general components such as perceptual systems, memory systems, and the executive function system. Some, though not all, give rise to models that not only describe the internal articulation of language but also make predictions about linguistic behavior at other levels of resolution e.g., (neuro)cognitive, computational.
This collection of articles has two objectives:
1. To bring together work which connects empirical and/or experimental results with clearly specified assumptions and hypotheses about language architecture and,
2. To reflect current understanding of language from a broad spectrum of empirical approaches --with the aim of highlighting their respective key vantage points-- as they exist in this third decade of the 21st century.
We welcome contributions from linguistics, psychology, cognitive (neuro)science, and computer science. We encourage manuscript submissions that address wider questions concerning:
- The structure of language as a mental system, including modeling of specific components (phonetics, phonology, morphology, syntax, semantics, and pragmatics); the interplay among these components and the division of labor between grammar and lexicon
- The role of language as a mental system in language comprehension and production
- The acquisition of language as a mental system under various conditions (e.g. L1, L2, SLE, dialect, language contact, home sign and emerging sign languages)
- The balance between domain-specific and domain-general components or principles of organization
- The computational, cognitive and biological plausibility of proposed language architectures
This Research Topic invites Original Research articles (up to 12,000 words) and is also open to shorter contributions such as Brief Research Reports (max. 4,000 words) and Perspectives (max. 3,000 words) that provide a snapshot of our current understanding of the language architecture on the basis of original research results. More information on article types accepted in Frontiers in Artificial Intelligence can be found
here.
