Volitional stimuli are felt to generate coarse spatiotopic responses of vergence which are subsequently refined using retinotopic stimuli. The aim of this study was to examine the accuracy of vergence responses to volitional stimuli.

Ten subjects with mean age of 24.5 y and a standard deviation of 4.2 y, participated in the experiment with informed consent. All subjects were emmetropic with vision of 6/6 or better in both eyes. The subjects were naive as to the purpose of the experiment. Subjects were shown a series of targets (Snellen E {6/9 equivalent size} printed on 35 mm slides) at vergences of 0.25, 0.50, 1.00, 2.00, 3.00 and 5.00 D. Subjects were then placed in darkness and instructed to imagine that they were looking at the same "target" distances again in a random order. They were then instructed to imagine that they were looking at a very distant target and a very near target. Vergence responses were measured using an infrared eye tracker with a resolution of ~0.25 MA.

All subjects produced appropriate vergence responses for imagining very near or very far distances. The responses to the remembered target distances showed considerable inter-subject variability, but tended to be very inaccurate. Subjects generally produced two flat response regions to volitional stimuli which corresponded to far (0.25 - 2.00 D) and near imagined distances (3.00 - 5.00 D). The subjects would switch between these responses depending upon the stimulus distance they were asked to imagine. The experimental results can be accurately modelled with a new model of vergence.

Volitional stimulation produces coarse, inaccurate responses of vergence in agreement with the spatiotopic/retinotopic classification. These responses can be grouped into a far response region for perceived distances of > 50 cm, and a near response region for perceived distances of < 50 cm.