Abstract
Purpose:
Patients with very low vision or with retinal prosthesis can have false localization . However, the relationship between a target position and its localization by touching it has not been studied quantitatively. The purpose of this study was to investigate the accuracy of touching a fixed target under simulated very low vision.
Methods:
Twenty eight eyes of 14 healthy subjects (men, 5; women, 9; age, 22 to 39 years) wore spectacle frames to which 3 sheets of occlusion foil (<0.1 Ryser®) were attached to simulate hand motion vision. In the localization test, a white square target (visual angle; 10°) was displayed on a PC monitor at random positions, and the subjects were instructed to touch the center of the square. The touched positions relative to the center of the target were evaluated by dividing the screen area into 4 quadrants, and the distance between the touched position and the center of a target was calculated by a computer. The visual acuity of the subjects was examined with the PC monitor with grating targets.
Results:
The visual acuity ranged from 3.2 to 2.5 log MAR units, and all of the subjects were able to recognize the position of the target roughly but were not able to identify it relative to the position of the finger. The touched position deviated from the center of the target by a mean visual angle of 6.5±1.9°, but the distance was not correlated with the visual acuity (Pearson, r=0.0870, P=0.660). The touched position was more frequent in the lower or right quadrant than in the upper or left quadrant of the screen (One way ANOVA, P<0.0001).
Conclusions:
Because subjects with simulated very low vision were not able to see their own finger, visual feedback was not used in directing their finger. This probably caused the misdirection in the touch placement. These findings indicate that training will be required after an implantation of retinal prosthesis to learn the correct location of a target.