Sources of position-perception error for small isolated targets

It has often been reported that, in the presence of static reference stimul i, briefly presented visual targets are perceived as being closer to the fi xation point than they actually are. The first purpose of the present study was to investigate whether the same phenomenon can be demonstrated in a si tuation without static reference stimuli. Experiment 1, with position namin g as the task, showed that such a central shift is also observed under thes e conditions. This finding is of importance because it completes an explana tion for central near-location errors in the partial-report bar-probe task. The second purpose of the present study was to provide an explanation for these central shifts. For this explanation information about the exact size of the central shift is required. In Exps. 2, 3, and 4, with cursor settin g as the task, it was attempted to assess more precisely the size of the ce ntral shifts. These experiments revealed that two different factors determi ne the results in cursor setting tasks; a factor "target position" and a fa ctor "cursor position." Experiment 5 showed that it is the point of fixatio n, not the fixation point. that serves, at least in part, as the reference point in this type of task. All the findings together allow us to conclude that the target positions are underestimated by about 10%. From vision rese arch it is known that saccadic eye movements, performed for bringing a targ et in the fovea, also show an undershoot of about 10%. It is therefore conc luded that the system in charge of saccadic eye movements also provides the metric in visual space within a single eye fixation.