Computational models, educational implications, and methodological innovations: The realm of visual word recognition

Abstract

This article aims to provide an overview of the current status of visual word recognition research, from the main models and their current challenges, to the educational and methodological implications of studies in this field. Visual word recognition is a critical reading process that connects visual sensation and perception with linguistic (sentence, text) processing. For this reason, it has captured the interest of researchers in cognitive science. Importantly, it is particularly easy to model quantitatively and researchers have developed a number of computational models to explain the processes involved. Recent years have witnessed an increasing number of corpora in several languages, including average identification times of thousands of words, allowing virtual simulations of experiments to test the predictions of theoretical models without the recruitment of participants. Nevertheless, despite the advances achieved in the understanding of word processing, models still have outstanding questions to be answered, such as the role of visual information during word recognition, or how diacritics are represented at the letter level. On the applied side, word recognition research has also contributed to the improvement of educational techniques, such as the development of friendly fonts for different populations, along with methodological innovations in cognitive psychology, such as the use of linear-mixed effects models, Bayesian methods and multi-laboratory approaches.