Evidence for implicit learning in syntactic comprehension.

Area of Science:

• Psycholinguistics
• Cognitive Science
• Computational Linguistics

Background:

• Syntactic priming is a phenomenon where exposure to a particular sentence structure increases the likelihood of using that structure again.
• Existing theories of syntactic priming include activation-based and error-based learning accounts.
• The precise mechanisms underlying implicit statistical learning in language processing remain under investigation.

Purpose of the Study:

• To investigate the role of error signals in implicit learning during syntactic comprehension.
• To differentiate between activation-based and error-based accounts of syntactic priming.
• To explore the functional motivations for syntactic priming in language processing.

Main Methods:

• Reanalysis of existing data from a syntactic priming experiment (Thothathiri & Snedeker, 2008).
• Analysis of how error signals associated with syntactic primes influence subsequent syntactic expectations.
• Comparison of findings with predictions from different theoretical models of syntactic priming.

Main Results:

• Evidence suggests that error signals, not just activation, influence comprehenders' syntactic expectations.
• The results support error-based implicit learning accounts of syntactic priming.
• Findings challenge accounts that attribute syntactic priming solely to temporary increases in base-level activation.

Conclusions:

• Implicit learning plays a crucial role in syntactic comprehension, mediated by error signals.
• Error-based learning provides a more accurate framework for understanding syntactic priming than simple activation models.
• Further research is needed to understand the maintenance of implicit statistical knowledge and the functional role of syntactic priming.