09:00 - 10:30
Parallel sessions 7
09:00 - 10:30
Room: HSZ - N4
Chair/s:
Benjamin Gagl
Visual word recognition and reading are central to human communication. Still, literacy rates are declining, increasing the need for better reading education and interventions for readers with low skill levels. At the beginning of such developments, one must understand the cognitions underlying reading. Here, we combine presentations that provide current developments in reading research, investigating how language, script, and memory influence visual word recognition processes in behavior and brain activation. We will start with a study by Sabrina Turker, which investigates the influence of language and memory skills on reading disabilities. The second study, by Benjamin Gagl, examines the influence of which items are stored in the lexicon on orthographic processing in visual word recognition behavior and brain responses. The third study, by Amelie Hague, investigates script familiarity on brain response dynamics. The fourth study, by Maz Mohamed, analyzes how learning to read in different languages influences the process of lexical access. Finally, Jana Hasenäcker presents a large-scale study of German lexicon decision data, which is essential to exploring novel hypotheses built on consensus-based guidelines, embracing open science methodology. The symposium relies on behavioral and brain findings across studies using implemented theoretical approaches through computational models, and offers an overview of the availability of novel datasets. Thus, this symposium delivers a comprehensive update on the neuro-cognitive processes implemented in reading and visual word recognition, including current theoretical advancements. 
Submission 701
Lexical Processing in English, Dutch, Estonian, and Malay: Pros and Cons of Combined N-Grams of Different Sizes in the Discriminative Lexicon Model
SymposiumTalk-04
Presented by: Maziyah Mohamed
Maziyah Mohamed 1, Sean Tseng 1, Shanshan Xu 1, Joshua Snell 2, Harald Baayen 1
1 University of Tübingen, Germany
2 Vrije Universiteit Amsterdam, Netherlands
The Discriminative Lexicon Model (DLM) works with linear or deep mappings from form embeddings to meaning embeddings. One simple way of designing form embeddings is to construct multiple-hot binary vectors that specify which n-grams (for fixed n, e.g., bigrams or trigrams) are present in a words orthographic or phonological representation. We report on explorations of combining n-grams for multiple n (e.g., form vectors specifying the presence of both bigrams, trigrams, and 4-grams), for four different languages, with special attention to trade-offs between gram sizes, the use of linear versus deep mappings, and language. Deep mappings offer higher prediction accuracy, but linear mappings tend to be more precise for predicting response latencies. Linear mappings with multiple-sized n-grams also show a clear advantage compared to models with only 3-grams, and tend to outperform models with deep mappings. The option of working will multiple n-grams will become available in the julia package JudiLing, which provides a computational implementation of the DLM.