Visual crowding refers to the impairment of recognizing peripherally presented objects flanked by distractors. Crowding is characterized by a radial-tangential anisotropy, which describes an ellipsoid shape of the crowding area elongated along the radial target-fovea axis. Crowding effects, exhibiting a “critical distance” between target and flankers, can be reduced by perceptual learning. In this experiment we investigated the learning-induced reduction of crowding in normally-sighted participants of three age groups (<40 years, N=30; 40-60 years, N=21; >60 years, N=14), who were trained to detect the orientation of a Landolt-C optotype flanked by distractors in the upper left and right quadrant of the visual field at an eccentricity of 6.5° visual angle over four sessions. A staircase procedure (2-down, 1-up) adjusted the critical target-to-flanker distance needed to achieve 70.7% correct responses during training. The results show that all three age groups improved significantly over four sessions (p < .001), indicated by a decrease of this target-to-flanker distance, with no significant difference between the age groups (p = .562). Interestingly, the radial-tangential anisotropy was less pronounced in the two older age groups in comparison to the youngest group. Overall, the anisotropy tended to decrease with training (p = .058). The results show that perceptual learning of a crowding task with optotypes is also feasible for older age groups, and could as such be a promising tool in rehabilitation programs to help improve peripheral vision e.g. in patients with central vision loss.

Acknowledgments: Supported by Deutsche Forschungsgemeinschaft (DFG: PL641-1/1), ZSER e.V. c/o Universitätsklinikum Regensburg