Convexity Biases in Stereoscopic Judgements of Ground Terrain. Hartle, B., Allison, R. S., & Wilcox, L. M. In CVR-CIAN Conference 2025: The Brain and Integrative Vision, pages 55. 2025. ![Convexity Biases in Stereoscopic Judgements of Ground Terrain [link]](https://bibbase.org/img/filetypes/link.svg "link")
-1 doi abstract bibtex Safely traversing across irregular terrain requires identification of complex variation in upcoming surfaces. We previously showed that stereopsis plays a significant role when discriminating terrain irregularities. However, observers were often biased towards perceiving convex surfaces, even when stereopsis indicated a concave depression. It was surprising that strong, yet ambiguous shading could activate a convexity bias and override stereopsis. The aim of the current study was to 1) determine whether the bias was generalizable to other viewing environments, and if so, 2) better understand its origin. We first replicated our previous study using a virtual reality headset. We then investigated whether the distribution of stereoscopic cues impacted the bias by varying the availability of binocular disparities either by 1) only showing one half of the surface or 2) presenting the full surface to one eye and half of the surface to the other (half- occlusion). All stimuli consisted of virtual grass terrain with a central mound or dip. We measured discrimination thresholds for the direction of terrain features (convex vs. concave). As in our previous study; under strong shading, half of observers exhibited a convexity bias. Observers that had elevated (but normal) stereoacuity thresholds were more likely to exhibit biases for convexity than observers more sensitive to stereoscopic depth. Judgements of half-occluded surfaces showed that eliminating the binocular disparity signal from one surface half did not impact the convexity bias. The bias was only present when the shading cue was strong. Thus, our results demonstrate that low illumination in the shaded region did not reduce the reliability of stereoscopic cues. Instead, the bias was attributable to the presence of strong shading. Despite all observers possessing normal stereoacuity, 53% relied on an ambiguous shading cue to interpret the ground relief. In impoverished environments, this could lead to potentially hazardous misjudgments during locomotion.
@incollection{Hartle:2025aa,
abstract = {Safely traversing across irregular terrain requires identification of complex variation in
upcoming surfaces. We previously showed that stereopsis plays a significant role when
discriminating terrain irregularities. However, observers were often biased towards
perceiving convex surfaces, even when stereopsis indicated a concave depression. It
was surprising that strong, yet ambiguous shading could activate a convexity bias and
override stereopsis. The aim of the current study was to 1) determine whether the bias
was generalizable to other viewing environments, and if so, 2) better understand its
origin. We first replicated our previous study using a virtual reality headset. We then
investigated whether the distribution of stereoscopic cues impacted the bias by varying
the availability of binocular disparities either by 1) only showing one half of the surface
or 2) presenting the full surface to one eye and half of the surface to the other (half-
occlusion). All stimuli consisted of virtual grass terrain with a central mound or dip. We
measured discrimination thresholds for the direction of terrain features (convex vs.
concave). As in our previous study; under strong shading, half of observers exhibited a
convexity bias. Observers that had elevated (but normal) stereoacuity thresholds were
more likely to exhibit biases for convexity than observers more sensitive to stereoscopic
depth. Judgements of half-occluded surfaces showed that eliminating the binocular
disparity signal from one surface half did not impact the convexity bias. The bias was
only present when the shading cue was strong. Thus, our results demonstrate that
low illumination in the shaded region did not reduce the reliability of stereoscopic
cues. Instead, the bias was attributable to the presence of strong shading. Despite all
observers possessing normal stereoacuity, 53\% relied on an ambiguous shading cue to
interpret the ground relief. In impoverished environments, this could lead to potentially
hazardous misjudgments during locomotion.},
annote = {JUNE 17-19, 2025
SECOND STUDENT CENTRE
YORK UNIVERSITY},
author = {Brittney Hartle and Robert S. Allison and Laurie M. Wilcox},
booktitle = {CVR-CIAN Conference 2025: The Brain and Integrative Vision},
date-added = {2025-07-26 06:20:14 -0400},
date-modified = {2025-07-26 06:20:14 -0400},
doi = {10.25071/10315/42927},
keywords = {Stereopsis},
pages = {55},
title = {Convexity Biases in Stereoscopic Judgements of Ground Terrain},
url-1 = {https://doi.org/10.25071/10315/42927},
year = {2025},
bdsk-url-1 = {https://doi.org/10.25071/10315/42927}}
Downloads: 0
{"_id":"ZCYND6X3ArvAYyQ2r","bibbaseid":"hartle-allison-wilcox-convexitybiasesinstereoscopicjudgementsofgroundterrain-2025","author_short":["Hartle, B.","Allison, R. S.","Wilcox, L. M."],"bibdata":{"bibtype":"incollection","type":"incollection","abstract":"Safely traversing across irregular terrain requires identification of complex variation in upcoming surfaces. We previously showed that stereopsis plays a significant role when discriminating terrain irregularities. However, observers were often biased towards perceiving convex surfaces, even when stereopsis indicated a concave depression. It was surprising that strong, yet ambiguous shading could activate a convexity bias and override stereopsis. The aim of the current study was to 1) determine whether the bias was generalizable to other viewing environments, and if so, 2) better understand its origin. We first replicated our previous study using a virtual reality headset. We then investigated whether the distribution of stereoscopic cues impacted the bias by varying the availability of binocular disparities either by 1) only showing one half of the surface or 2) presenting the full surface to one eye and half of the surface to the other (half- occlusion). All stimuli consisted of virtual grass terrain with a central mound or dip. We measured discrimination thresholds for the direction of terrain features (convex vs. concave). As in our previous study; under strong shading, half of observers exhibited a convexity bias. Observers that had elevated (but normal) stereoacuity thresholds were more likely to exhibit biases for convexity than observers more sensitive to stereoscopic depth. Judgements of half-occluded surfaces showed that eliminating the binocular disparity signal from one surface half did not impact the convexity bias. The bias was only present when the shading cue was strong. Thus, our results demonstrate that low illumination in the shaded region did not reduce the reliability of stereoscopic cues. Instead, the bias was attributable to the presence of strong shading. Despite all observers possessing normal stereoacuity, 53% relied on an ambiguous shading cue to interpret the ground relief. In impoverished environments, this could lead to potentially hazardous misjudgments during locomotion.","annote":"JUNE 17-19, 2025 SECOND STUDENT CENTRE YORK UNIVERSITY","author":[{"firstnames":["Brittney"],"propositions":[],"lastnames":["Hartle"],"suffixes":[]},{"firstnames":["Robert","S."],"propositions":[],"lastnames":["Allison"],"suffixes":[]},{"firstnames":["Laurie","M."],"propositions":[],"lastnames":["Wilcox"],"suffixes":[]}],"booktitle":"CVR-CIAN Conference 2025: The Brain and Integrative Vision","date-added":"2025-07-26 06:20:14 -0400","date-modified":"2025-07-26 06:20:14 -0400","doi":"10.25071/10315/42927","keywords":"Stereopsis","pages":"55","title":"Convexity Biases in Stereoscopic Judgements of Ground Terrain","url-1":"https://doi.org/10.25071/10315/42927","year":"2025","bdsk-url-1":"https://doi.org/10.25071/10315/42927","bibtex":"@incollection{Hartle:2025aa,\n\tabstract = {Safely traversing across irregular terrain requires identification of complex variation in\nupcoming surfaces. We previously showed that stereopsis plays a significant role when\ndiscriminating terrain irregularities. However, observers were often biased towards\nperceiving convex surfaces, even when stereopsis indicated a concave depression. It\nwas surprising that strong, yet ambiguous shading could activate a convexity bias and\noverride stereopsis. The aim of the current study was to 1) determine whether the bias\nwas generalizable to other viewing environments, and if so, 2) better understand its\norigin. We first replicated our previous study using a virtual reality headset. We then\ninvestigated whether the distribution of stereoscopic cues impacted the bias by varying\nthe availability of binocular disparities either by 1) only showing one half of the surface\nor 2) presenting the full surface to one eye and half of the surface to the other (half-\nocclusion). All stimuli consisted of virtual grass terrain with a central mound or dip. We\nmeasured discrimination thresholds for the direction of terrain features (convex vs.\nconcave). As in our previous study; under strong shading, half of observers exhibited a\nconvexity bias. Observers that had elevated (but normal) stereoacuity thresholds were\nmore likely to exhibit biases for convexity than observers more sensitive to stereoscopic\ndepth. Judgements of half-occluded surfaces showed that eliminating the binocular\ndisparity signal from one surface half did not impact the convexity bias. The bias was\nonly present when the shading cue was strong. Thus, our results demonstrate that\nlow illumination in the shaded region did not reduce the reliability of stereoscopic\ncues. Instead, the bias was attributable to the presence of strong shading. Despite all\nobservers possessing normal stereoacuity, 53\\% relied on an ambiguous shading cue to\ninterpret the ground relief. In impoverished environments, this could lead to potentially\nhazardous misjudgments during locomotion.},\n\tannote = {JUNE 17-19, 2025\nSECOND STUDENT CENTRE\nYORK UNIVERSITY},\n\tauthor = {Brittney Hartle and Robert S. Allison and Laurie M. Wilcox},\n\tbooktitle = {CVR-CIAN Conference 2025: The Brain and Integrative Vision},\n\tdate-added = {2025-07-26 06:20:14 -0400},\n\tdate-modified = {2025-07-26 06:20:14 -0400},\n\tdoi = {10.25071/10315/42927},\n\tkeywords = {Stereopsis},\n\tpages = {55},\n\ttitle = {Convexity Biases in Stereoscopic Judgements of Ground Terrain},\n\turl-1 = {https://doi.org/10.25071/10315/42927},\n\tyear = {2025},\n\tbdsk-url-1 = {https://doi.org/10.25071/10315/42927}}\n\n","author_short":["Hartle, B.","Allison, R. S.","Wilcox, L. M."],"key":"Hartle:2025aa","id":"Hartle:2025aa","bibbaseid":"hartle-allison-wilcox-convexitybiasesinstereoscopicjudgementsofgroundterrain-2025","role":"author","urls":{"-1":"https://doi.org/10.25071/10315/42927"},"keyword":["Stereopsis"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"incollection","biburl":"www.cse.yorku.ca/percept/papers/self.bib","dataSources":["2KKYxJNEDKp35ykmq","BPKPSXjrbMGteC59J"],"keywords":["stereopsis"],"search_terms":["convexity","biases","stereoscopic","judgements","ground","terrain","hartle","allison","wilcox"],"title":"Convexity Biases in Stereoscopic Judgements of Ground Terrain","year":2025}