Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Processing bottlenecks in visual word recognition
Poster A117 in Poster Session A, Tuesday, October 24, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Alex White1; 1Barnard College
When reading this page, you can see many words at once. But can you process the meaning of multiple words simultaneously? I will present psychophysical and fMRI data about the processing bottlenecks that constrain visual word recognition, including with the poor quality of peripheral vision and capacity limits in the ‘visual word form area’ (VWFA). In many experiments we have found that when two words are briefly flashed together, readers are able to recognize only one of them. This is consistent with a serial model of recognition in which attention must be focused single words. Concurrent fMRI data show that the anterior portion of the VWFA responds to the lexical features of only one attended word at a time. Additional studies, however, suggest important exceptions to this bottleneck when the experimental conditions better approximate natural reading. First, if two unrelated words are short enough and arranged horizontally close together, many (but not all) participants can process them both simultaneously. This suggests that there is a narrow window around the point of gaze fixation within which two words may be recognized independently and in parallel. Second, when two words are presented outside that window, semantic relations between them may nonetheless be detected. Specifically, if the two words happen to form a compound word (like “water + fall”), participants have a better chance of recognizing them both simultaneously. Altogether, this research points towards a limited, context-dependent capacity for parallel processing of multiple written words. The large individual differences in this capacity will be a major topic of future research.
Topic Areas: Reading, Computational Approaches