Figure 7
The results of Exp. 4 where 793 participants completed the visual crowding experiment implemented using Virtual Chinrest on LabintheWild. (a) The average visual crowding effects were significantly different between target eccentricity of 4° (mean = 1.61°) and 6° (mean = 2.66°), and between participants with (e = 4◦: 1.90°; e = 6°: 3.03◦) and without (e = 4°: 1.62°; e = 6°: 2.58°) dyslexia. Error bars represent standard error. (b) The distribution of the percentage correctness of the crowding experiment across all participants. The average (indicated by the red vertical line) is 85.56%. (c) The distribution of the viewing distances across all participants calculated by Virtual Chinrest. Our participants’ viewing distances were between 17.4 cm and 68.3 cm with mean = 47.3 cm and sd = 8.9 cm. (d) The distribution of the within-subjects standard deviation (SD) of the viewing distances across all participants: the average is 3.9 cm (min = 0.003 cm, max = 22.7 cm). (e) The pairwise correlation of calculated viewing distances among three blind spot tasks at the beginning (s1), in the middle (s2) and at the end (s3) of the crowding experiment. The correlations of calculated viewing distances between s1 and s2, s2 and s3, s1 and s3 are 0.706, 0.805 and 0.630, respectively.
