The Provo Corpus: A large eye-tracking corpus with predictability norms.

Area of Science:

• Cognitive Psychology
• Computational Linguistics
• Neuroscience

Background:

• Eye-tracking research relies on predictable text to study reading processes.
• Existing corpora lack detailed predictability measures beyond orthographic form.

Purpose of the Study:

• Introduce the Provo Corpus, a novel dataset for reading research.
• Provide enhanced predictability norms for linguistic information.

Main Methods:

• Collected eye-tracking data.
• Developed predictability norms including cloze scores, morpho-syntactic, and semantic information.
• Compiled the Provo Corpus.

Main Results:

• The Provo Corpus includes comprehensive predictability norms.
• These norms extend beyond traditional orthographic predictability.
• The corpus is available for public access.

Conclusions:

• The Provo Corpus is a valuable resource for studying predictive reading.
• Its unique norms facilitate deeper insights into reading mechanisms.
• Researchers can access the data for further analysis.