Abstract
Decisions about when and where to move the eyes are commonly assumed to be made independently (e.g., Findlay & Walker, 1999). Consequently, the influence of an experimental manipulation on fixation times in reading is typically analyzed without considering the possible impact of this manipulation on fixation locations. However, during reading, it is well-known that fixation location within the word influences fixation durations. Specifically, first-fixation duration is longer for fixations located near the center of the word than for fixations near the beginning or the end of the word. Thus, it is necessary to establish the extent to which location effects can interact with lexical variables such as word frequency, to rule out the possibility that lexical effects are at least in part due to subtle differences in fixation location. In a previous study (Reingold, Reichle, Glaholt, & Sheridan, accepted), we demonstrated that location effects on fixation duration were largely independent of the impact of word frequency (high vs. low) and parafoveal preview validity (valid vs. invalid preview). Extending this research, the present study directly manipulated a variable that was expected to influence fixation location. Specifically, we contrasted a normal text condition with a modified text condition that contained random numbers between 2 and 9 instead of spaces (e.g., John3decided8to5sell9the7table). Relative to the normal text condition, the modified text condition produced landing positions that were closer to the beginning of the word, as well as shorter saccades, longer fixations, and slower overall reading times. However, the normal and modified text conditions showed a similar pattern of word frequency and preview validity effects. The present results indicate that location effects during reading are largely independent of the influence of lexical variables, and more generally, our findings support the assumption that when and where decisions constitute independent factors in eye-movement control in reading.
Meeting abstract presented at VSS 2012