PONG: A computational model of visual word recognition through bihemispheric activation

Area of Science:

• Cognitive psychology
• Neuroscience
• Computational linguistics

Background:

• Orthographic processing and visual word recognition are complex, with existing models struggling to explain all observed phenomena.
• Recent research findings highlight limitations in current theoretical frameworks for visual word recognition.

Purpose of the Study:

• To propose a novel theory, Positional Ordering of N-Grams (PONG), to explain how the brain computes letter positions.
• To provide a unified framework that accounts for both established and emerging findings in orthographic processing.

Main Methods:

• Introduction of the PONG theory, positing hemisphere-specific N-gram detectors sensitive to visual field location.
• Modeling word activation as a function of N-gram identity and relative laterality.
• Simulations to test the PONG framework against known and novel phenomena.

Main Results:

• The PONG model successfully explains classic and recent findings in visual word recognition.
• The framework demonstrates efficacy in accounting for cross-linguistic differences unexplained by other models.
• Simulations indicate PONG's robust explanatory power for orthographic processing.

Conclusions:

• PONG offers a new perspective on the neural computation of letter positions in orthographic processing.
• The theory provides a more comprehensive account of visual word recognition than previous models.
• Further research is warranted to explore the neurophysiological plausibility and applications of PONG.