Gagl, B. 1, 2 , Richlan, F. 3 , Ludersdorfer, P. 3, 4 , Sassenhagen, J. 1 , Eisenhauer, S. 1 , Gregorova, K. . 1 & Fiebach, C. J. . 1, 2
1 Department of Psychology, Goethe University Frankfurt, Germany
2 Center for Individual Development and Adaptive Education of Children at Risk (IDeA), Frankfurt am Main, Germany
3 Centre for Cognitive Neuroscience, University of Salzburg, Austria
4 Wellcome Trust Centre for Neuroimaging, Institute of Neurology, University College London, UK
To characterize the role of the left ventral occipito-temporal cortex (lvOT) during visual word recognition in a quantitatively explicit and testable manner, we propose the lexical categorization model (LCM). In the LCM, we assume that lvOT categorizes perceived letter strings into words or non-words. This computation assures neuronal efficiency by abandoning high-level processing of meaningless letter strings in experiments and efficiently detects spelling errors and unknown words while reading. LCM simulations reproduce nine benchmark effects found with fMRI. Empirically, using fMRI, we demonstrate that LCM simulations predict lvOT brain activation that resembles the standard pattern: Consonant strings<Words<Pseudowords (N=18), but also patterns that deviate from the standard (N=35). Using electroencephalography (N=31), we show that LCM accounts for activations ~300ms after stimulus onset. Also, we found that word-likeness, the input to the LCM, is represented posterior to lvOT and before the lexical categorization. In contrast, a dichotomous word/non-word contrast, which is the assumed output of the LCM, could be localized to upstream frontal brain regions and after the lexical categorization. Thus, we propose a ventral-visual-stream processing framework for visual word recognition involving word-likeness extraction followed by lexical categorization, before accessing meaning. Finally, we implemented a lexical categorization intervention for German-language-learners resulting in a reading speed increase of 18% (N=89). Reaction times in the training showed an LCM effect, which increased with each session and, finally, we found a correlation of a person-specific estimate of this interaction and reading speed change. Hence, this evidence establishes that lexical categorization is fundamental to efficient reading.