What is it about?
We used the Connectionist Dual-Process model of reading to understand how different processes people use when reading work. Despite CDP specifying many different processes used in reading, we show that it can still help elucidate what the most important aspects of reading are, even when it has access to only very limited amounts of data.
Featured Image
Photo by Michał Parzuchowski on Unsplash
Why is it important?
Reading uses many different process and so models trying to explain how people read are often quite complicated. Rather than ignore potentially important aspects of reading by reducing the scope of what the Connectionist Dual-Process (CDP) can potentially explain we instead show that: (a) The important parameters of CDP can be identified; (b) the results CDP produces are robust even when the model is optimized on very small datasets; and (c) CDP performs quantitatively better than statistical methods that fit the data. These findings are important because they show that CDP could be used to help understand individual differences in reading and reading disorders where data is very limited.
Perspectives
We wrote this article because we wanted to show that even when very limited data sets are available, it is still possible to understand how people read. In the future, we hope to extend this work to understand variation in the way children with dyslexia read and to predict what strategies might help them the most.
Conrad Perry
University of Adelaide
Read the Original
This page is a summary of: Modelling a complex cognitive system with limited data: Optimization and generalization in a computational model of reading aloud, PLOS Complex Systems, November 2025, PLOS,
DOI: 10.1371/journal.pcsy.0000074.
You can read the full text:
Contributors
The following have contributed to this page
