The English Sublexical Toolkit: Methods for indexing sound-spelling consistency

Area of Science:

• Cognitive Science
• Computational Linguistics
• Psycholinguistics

Background:

• Understanding sublexical units is crucial for explaining reading and spelling.
• Existing measures of sublexical regularity may not fully capture the complexity of the English lexicon.

Purpose of the Study:

• Introduce the English Sublexical Toolkit for analyzing sublexical knowledge.
• Provide superior and novel indices of sublexical regularity.
• Explain variance in reading and spelling behaviors.

Main Methods:

• Developed a suite of tools based on an experience-dependent learning framework.
• Quantified regularity of sublexical units (grapheme-to-phoneme, phoneme-to-grapheme) at multiple grain sizes.
• Extracted frequency and conditional probability indices for words and pseudowords.

Main Results:

• Measures from the toolkit explain significant variance in word naming and lexical decision tasks.
• The toolkit's indices outperform or match existing consistency measures.
• Novel indices like phonographeme frequency were introduced.
• Toolkit measures account for the majority of variance in pseudoword pronunciation probabilities.

Conclusions:

• The English Sublexical Toolkit offers a flexible framework and resources for analyzing sublexical regularities.
• These tools provide more accurate estimates of sublexical complexity.
• The toolkit aids in explaining reading and spelling behaviors for both real and pseudowords.