Abstract
Inaccuracy in the vergence eye position ("fixation disparity") can occur despite a fusion stimulus. When measured with eye trackers, this inaccuracy is referred to as "objective fixation disparity". It is a matter of debate whether objective fixation disparity can be estimated with a technically simple psycho-physical procedure, i.e. the perceived offset of aligned dichoptic nonius targets, referred to as "subjective fixation disparity". To investigate the relation between these two measures, simultaneous tests were made in far vision when placing prisms in front of the eyes (for a few seconds) in order to induce forced vergence, i.e. to vary the absolute disparity (from 1 deg divergent to 3.4 deg convergent). Frequent repeated measurements in 12 observers allowed for individual analyses. Generally, fixation disparity values and the effects of prisms were much smaller in the subjective than in the objective measures. Some observers differed systematically in the characteristics of the two types of prism-induced curves. Individual regressions showed that the subjective vs. objective slope was 8% on the average (with largest individual values of 18%). This suggests that sensory fusion shifts the visual direction of the (peripheral) binocular targets by the full amount of objective fixation disparity (since single vision was achieved); however, for the (central) monocular nonius lines this shift was more or less incomplete so that the dichoptic nonius targets indicated an individual percentage of objective fixation disparity. The subjective-to-objective ratio seems to be an individual characteristic of fixation disparity in terms of the amount and in terms of the effect of prism-induced forced vergence. Therefore, on the group level the subjective measures do not allow for a precise prediction of the objective measures.