Individual differences in reading aloud: a mega-study, item effects, and some models

Area of Science:

• Cognitive Psychology
• Neuroscience
• Computational Linguistics

Background:

• Individual differences in reading are often overlooked in visual word recognition models.
• Existing models primarily focus on item response time effects and neuropsychological disorders.
• Understanding individual variations can refine theories of reading processes.

Purpose of the Study:

• To investigate how individual differences in reading aloud performance can inform and test computational models.
• To assess the ability of established models (DRC, CDP+) to capture individual differences in response times.
• To develop an improved model that better accounts for individual variations in reading.

Main Methods:

• 100 participants performed a word reading aloud task.
• Parameters for the Dual Route Cascaded (DRC) and Connectionist Dual Process Plus (CDP+) models were estimated for each individual.
• A modified model, DRC-FC, was developed by altering the frequency effect's locus.

Main Results:

• Neither the DRC nor CDP+ models adequately captured the observed individual differences and their correlations.
• The novel DRC-FC model demonstrated improved ability to account for correlations among individual differences.
• Significant individual differences in reading persist even after controlling for general processing speed.

Conclusions:

• Individual differences in reading are crucial for testing and advancing models of visual word recognition.
• The DRC-FC model offers a more accurate representation of individual reading variations.
• This research provides a valuable dataset and framework for future modeling efforts integrating individual differences.