Explaining the Sentence Superiority Effect and N400s Elicited by Words and Short Sentences with OB1-Reader

Area of Science:

• Cognitive Science
• Computational Neuroscience
• Psycholinguistics

Background:

• Computational models enhance understanding of reading behavior.
• Linking model 'physiology' to behavior increases predictive power.
• OB1-reader simulates human reading processes and has explained prior findings.

Purpose of the Study:

• Extend OB1-reader's scope to new experimental paradigms.
• Simulate behavioral performance and electroencephalography (EEG) activity.
• Investigate word recognition and sentence processing.

Main Methods:

• Utilized OB1-reader to simulate two word-recognition tasks: flanker and sentence reading.
• Compared simulated behavioral data with experimental results.
• Generated virtual event-related potentials (ERPs) from OB1 node activity and compared to human ERPs.

Main Results:

• OB1 simulated key behavioral findings, including the sentence superiority effect.
• Model activity predicted the N400 component size in human readers.
• The model's lexical activity correlated with N400 but not N250 amplitude.

Conclusions:

• OB1-reader demonstrates capacity to simulate complex reading behaviors and neural correlates.
• The model's 'lexical activity' is a relevant predictor for the N400 component in reading.
• Further refinement may be needed to capture earlier ERP components like N250.