Figure 2.

An example of the boundary paradigm. The asterisk underneath each sentence indicates the reader’s fixation location. An invisible boundary is placed in a sentence in the space immediately before a target word (the lines in the example above represent the location of the boundary, but this is not visible to participants). A preview letter string is available in the target word’s location prior to the reader making a saccade that crosses this invisible boundary. After the reader’s eyes cross the boundary, they move to directly fixate the target word. Then, a display change occurs wherein the preview letter string changes to the correct target word. By manipulating certain characteristics of the overlap (e.g., phonological similarity) between the preview string and the target word, parafoveal pre-processing can be studied. For example, phonologically consistent (e.g., brake) and inconsistent (e.g., bread) previews can be presented for a target (e.g., break) to examine the extent to which a reader is undergoing phonological pre-processing prior to direct fixation. If a reader does extract phonological information during parafoveal pre-processing, then reading times on the target word should be shorter following a consistent preview than an inconsistent preview. This decrease to reading times is referred to as preview benefit. If preview benefit is found, i.e., shorter reading times, on a word that was parafoveally available compared to when the parafoveal preview word was masked, this is strongly indicative of parafoveal pre-processing having occurred, as lexical identification has been facilitated. As such, parafoveal pre-processing and this paradigm enable researchers to investigate pre-lexical effects, as manipulations are conducted outside of direct fixation (i.e., lexical processing): if the manipulated characteristic of a given word in the parafovea confers preview benefit to the reader, the word must have been pre-lexically processed to some extent prior to it receiving a direct fixation.