Extraction of linguistic information from successive words during reading: Evidence for spatially distributed lexical processing

Chin An Wang, Albrecht W. Inhoff

Research output: Contribution to journalArticle

7 Citations (Scopus)


Two experiments examined whether word recognition progressed from one word to the next during reading, as maintained by sequential attention shift models such as the E-Z Reader model. The boundary technique was used to control the visibility of to-be-identified short target words, so that they were either previewed in the parafovea or masked. The eyes skipped a masked target on more than a quarter of the trials, and the following fixation must have been mislocated, if word recognition and saccade targeting progressed from one word to the next. Readers responded to the skipping parafoveally masked target words with relatively long viewing duration for the following posttarget word or with corrective saccades that returned the eyes from the posttarget word to the target. Experiment 2 manipulated the time-line of posttarget onset after target skipping, so that the posttarget word was either visible immediately upon fixation or after a short delay. The delay influenced posttarget viewing even when attention should have been focused at the target location according to E-Z Reader 10 simulations. These findings favor theoretical conceptions according to which lexical processing can encompass more than one word at a time.

Original languageEnglish
Pages (from-to)662-677
Number of pages16
JournalJournal of Experimental Psychology: Human Perception and Performance
Issue number3
Publication statusPublished - Aug 9 2013
Externally publishedYes



  • Attention gradient
  • Attentional spotlight
  • Eye movements
  • Reading
  • Word recognition

ASJC Scopus subject areas

  • Experimental and Cognitive Psychology
  • Arts and Humanities (miscellaneous)
  • Behavioral Neuroscience

Cite this