Table 1 Summary of task design, manipulations, results and implications for the Local Features papers included in the review
Local features | |||||
|---|---|---|---|---|---|
Citation | Low vision type | Task paradigm | Main manipulations | Results | Implications for design |
Bochsler et al. (2012) | Blur: 2 severity levels Mild: P–R = 0.8 logMAR = 0.88 (20/152) Severe: P–R = 0.6 logMAR = 1.65 (20/884) | Viewed and identified steps up or down, ramps up or down, or flat surfaces | Distance: 1.52, 3.05, and 6.10 m Blur severity Locomotion versus static viewing High-contrast checkerboard texture versus uniform gray surface | 1. Poorer performance with textured compared to uniform surface with severe blur 2. Locomotion improved accuracy over static viewing | Avoid high contrast texture on ground surfaces Designing for active locomotion could facilitate feature recognition |
Bochsler et al. (2013) | LV individuals (variety of types) Acuity range from logMAR 0.98 to 2.18 (20/191–20/3000) Field loss range of peripheral, central, or both | Viewed and identified steps up or down, ramps up or down, or flat surfaces | Distance: 1.52, 3.05, and 6.10 m Lighting: near and far Target-background contrast Locomotion versus static viewing | 1. People with LV outperformed those with simulated LV and were not strongly affected by target-background contrast 2. Similar to simulated LV, people with LV showed better identification for step-up compared to step-down and benefited from closer distances and locomotion | Designing for active locomotion could facilitate feature recognition |
Kallie et al. (2012) | Blur (severe) P–R = 0.6 Snellen 20/900 | Viewed boxes and cylinders (2–6 ft) Detection, confidence for detection, shape, and height | Distance: 3.05, 5.18, 7.32 m Lighting: overhead or window Color: White, gray Object height: short, medium, tall Shape: box, cylinder | 1. Cylinders were easier to identify compared to boxes 2. Advantage for color (white vs gray) depended on the lighting 3. Better performance resulted with larger and closer objects | Consider potential interactions of color, object shape, and lighting Overhead and window lighting may not differentially affect performance |
Legge et al. (2010) | Blur: 2 severity levels Mild: P–R = 0.8 logMAR = 0.81–.85 Severe: P–R = 0.6 logMAR = 1.64–1.67 | Viewed and identified steps up or down, ramps up or down, or flat surfaces Uniform gray surface with varying backgrounds | Distance: 1.52, 3.05, and 6.10 m Lighting: overhead, near window, far window Blur severity Background color: black or gray | 1. Steps up were more visible than steps down 2. Local geometric cues for identification (e.g., shape of edge contours of a walkway in an image) were dependent on viewing distance and contrast 3. A cue for identifying a ramp was its elevation in the image | Visibility of steps down is of particular concern and may be enhanced by contrast between riser and contiguous surface, and directional lighting |
Rand et al. (2011) | Blur (severe) logMAR = 1.60 (20/791) | Distance perception to targets | Visual horizon height (wall-floor boundary): actual or raised Distance: 3, 4.5, 6 m | 1. When the “horizon” was raised, the angle of declination to the target increased, and viewers judged the distance to targets on the ground to be closer | Create high contrast between wall and floor to make the visual horizon salient |
Rand et al. (2012) | Blur (severe) P–R = 0.46 logMAR = 1.51 (20/647) | Distance and size perception of targets on stands | Distance: 1.5, 2.7, 4 m Color of stands that the targets were placed on–created high or low contrast with ground plane Blur versus normal vision | 1. Distance and size judgments were accurate to the targets presented on the visible black stands 2. When the stands were not visible (painted gray and viewed through blur goggles), participants overestimated size and distance to the target | Increase visibility of information for grounding targets when they are located above the ground surface |
Tarampi et al. (2010) | Blur (severe) P–R = 0.36 logMAR = 1.53 | Distance perception by blind walking to targets | Distance: 1.5, 3.1, 6 m Blur versus normal vision | 1. Relatively accurate distance perception with blur, although with increased variability | Accuracy in distance estimation may be increased with salient visual horizon cues |