Vergence eye movements are not essential for stereoscopic depth

Active Vision in Binocular Depth Estimation: A Top-Down Perspective

Depth estimation is an ill-posed problem; objects of different shapes or dimensions, even if at different distances, may project to the same image on the retina. Our brain uses several cues for depth estimation, including monocular cues such as motion parallax and binocular cues such as diplopia. However, it remains unclear how the computations required for depth estimation are implemented in biologically plausible ways. State-of-the-art approaches to depth estimation based on deep neural networks implicitly describe the brain as a hierarchical feature detector. Instead, in this paper we propose an alternative approach that casts depth estimation as a problem of active inference. We show that depth can be inferred by inverting a hierarchical generative model that simultaneously predicts the eyes' projections from a 2D belief over an object. Model inversion consists of a series of biologically plausible homogeneous transformations based on Predictive Coding principles. Under the plausible assumption of a nonuniform fovea resolution, depth estimation favors an active vision strategy that fixates the object with the eyes, rendering the depth belief more accurate. This strategy is not realized by first fixating on a target and then estimating the depth; instead, it combines the two processes through action-perception cycles, with a similar mechanism of the saccades during object recognition. The proposed approach requires only local (top-down and bottom-up) message passing, which can be implemented in biologically plausible neural circuits.

Does vision extract absolute distance from vergence?

Since Kepler (1604) and Descartes (1637), 'vergence' (the angular rotation of the eyes) has been thought of as one of our most important absolute distance cues. But vergence has never been tested as an absolute distance cue divorced from obvious confounding cues such as binocular disparity. In this article, we control for these confounding cues for the first time by gradually manipulating vergence and find that observers fail to accurately judge distance from vergence. We consider several different interpretations of these results and argue that the most principled response to these results is to question the general effectiveness of vergence as an absolute distance cue. Given that other absolute distance cues (such as motion parallax and vertical disparities) are limited in application, this poses a real challenge to our contemporary understanding of visual scale.

Correlations between clinical measures and symptoms: Report 1: Stereoacuity with accommodative, vergence measures, and symptoms

AIM

The aim of this study was to explore associations between reduced stereoacuity and clinical measures of accommodation, vergences, and symptoms which could facilitate the development of quick and reliable screening tools.

METHODS

Using a multi-stage random cluster sampling, 1211 high school students (481 males and 730 females) between 13 and 18 years of age, were selected and examined. Visual acuity, stereoacuity and suppression, refractive errors, near point of convergence, heterophoria and fusional vergences, as well as, amplitude of accommodation, accommodative response, facility and relative accommodation were evaluated. Correlations among variables and the validity of Randot stereoacuity to distinguish between children with and without defective clinical measures as well as symptomatic versus asymptomatic children were characterized by the sensitivity and specificity of the tests.

RESULTS

The overall mean stereoacuity was 43.9 ± 25.23 s arc, and 18.9% [95% Confidence Interval, 16.6-21.4%)] of the participants had reduced stereoacuity (defined as ≥60). Stereoacuity values and symptoms scores were worse in children with defective clinical measures. The Receiver Operation Curve showed that maximum sensitivity and specificity was obtained with near point of convergence break (≥10 cm) of (0.70 95% confidence interval: 0.63-0.77) with Randot stereoacuity test (defined as ≥60 s arc). The correlations between reduced stereoacuity and symptoms scores was moderately strong and statistically significant (Pearson's, r = 0.507, p = 0.01). The Receiver Operation Curve showed that maximum sensitivity and specificity obtained with the Convergence Insufficiency Symptoms Survey was 0.57 (95% Confidence interval = 0.53-0.62, p = 0.001), sensitivity of 90.26%, and specificity 15.26% with the Randot stereoacuity test.

CONCLUSION

Reduced stereoacuity, defective clinical measures and symptoms of asthenopia were prevalent among sample of school children studied. Randot stereoacuity test could fairly distinguish between defective and normal clinical measures; though the accuracy to differentiate between symptomatic and asymptomatic school children is poor. These findings highlight the need for validation of a simple and fast screening tool in school settings. Further studies to confirm above findings will be needed.

Mice Discriminate Stereoscopic Surfaces Without Fixating in Depth

Stereopsis is a ubiquitous feature of primate mammalian vision, but little is known about if and how rodents such as mice use stereoscopic vision. We used random dot stereograms to test for stereopsis in male and female mice, and they were able to discriminate near from far surfaces over a range of disparities, with diminishing performance for small and large binocular disparities. Based on two-photon measurements of disparity tuning, the range of disparities represented in the visual cortex aligns with the behavior and covers a broad range of disparities. When we examined their binocular eye movements, we found that, unlike primates, mice did not systematically vary relative eye positions or use vergence eye movements when presented with different disparities. Nonetheless, the representation of disparity tuning was wide enough to capture stereoscopic information over a range of potential vergence angles. Although mice share fundamental characteristics of stereoscopic vision with primates and carnivores, their lack of disparity-dependent vergence eye movements and wide neuronal representation suggests that they may use a distinct strategy for stereopsis.SIGNIFICANCE STATEMENT Binocular vision allows us to derive depth information by comparing right and left eye information. We characterized binocular integration in mice because tools exist in these animals to dissect the underlying neural circuitry for binocular vision. Using random dot stereograms, we find that behavior and disparity tuning in the visual cortex share fundamental characteristics with primates, but we did not observe any evidence of disparity-dependent changes in vergence angle. We propose that mice use a distinct strategy of stereopsis compared with primates by using a broad range of disparities to encode depth over a large field of view and to compensate for nonstereoscopic changes in vergence angle that arise during natural behavior.

Stereoacuity and refractive, accommodative and vergence anomalies of South African school children, aged 13–18 years

Aim: The aim of this study was to explore possible associations between stereoacuity and refractive, accommodative and vergence anomalies.Methods: The study design was cross-sectional and comprised data from 1056 high school children aged between 13 and 18 years; mean age and standard deviation were 15.89 ± 1.58 years. Using a multi-stage random cluster sampling, participants were selected from 13 high schools out of a sample frame of 60 schools in the municipality concerned. In the final sample, 403 (38%) were males and 653 (62%) females. Refractive errors, heterophoria, near point of convergence, fusional vergences and accommodative functions (amplitude, facility, response and relative) were evaluated. Stereoacuity was evaluated using the Randot stereotest and recorded in seconds of arc where reduced stereoacuity was defined as worse than 40 s arc.Results: Overall, the mean stereoacuities (in seconds of arc) of the children with anomalies were the following: those with refractive errors (52.6 ± 36.9), with accommodative anomalies (53.1 ± 34.1) and with vergence anomalies (48.29 ± 31.1). The mean stereoacuity of those with vergence anomalies was significantly better than that of those with either refractive errors or accommodative anomalies (p = 0.02). In the refractive error category, only anisometropia had significantly reduced mean stereoacuity compared to emmetropia (Mann–Whitney U: p = 0.01). The mean stereoacuity of cases of accommodative anomalies was significantly reduced compared to those without such anomalies (Mann–Whitney U: p = 0.01). Similarly, the mean stereoacuity of cases with vergence anomalies was significantly reduced compared to those without vergence anomalies (p = 0.02).Conclusion: Refractive errors, accommodative or vergence anomalies are more likely to have reduced mean stereoacuity than cases without such anomalies. Refractive errors or accommodative anomalies had significantly more reduced stereoacuity than vergence anomalies. These findings suggest that the Randot stereotest could be used to identify those with such anomalies, and this study extends knowledge regarding the possible use of stereoacuity as a useful tool to screen for binocular anomalies.

Congenitally Impaired Disparity Vergence in Children With Infantile Esotropia

Purpose

We examined whether congenital impairment of disparity vergence in infantile esotropia (ET) exists in children with short duration ET (≤3 months) compared with long-duration ET and healthy controls. A short duration of misalignment would allow for a substantial amount of balanced binocular input during the critical period of binocular disparity development.

Methods

A total of 19 children aged 5 to 12 years and treated for infantile ET with a short (≤3 months; n = 10) or long (≥5 months; n = 9) duration of constant misalignment before alignment were enrolled. A total of 22 healthy control children were enrolled as a comparison group. Eye movements during disparity vergence and accommodative vergence were recorded using an EyeLink 1000 binocular eye tracker. Mean response gain was compared between and within groups to determine the effect of duration of misalignment and viewing condition.

Results

Compared with controls, children with short (P = 0.002) and long (P < 0.001) duration infantile ET had reduced response gains for disparity vergence, but not for accommodative vergence (P = 0.19).

Conclusions

Regardless of duration of misalignment, children with infantile ET had reduced disparity vergence, consistent with a congenital impairment of disparity vergence in infantile ET. Although early correction of misalignment increases the likelihood that some level of binocular disparity sensitivity will be present, normal levels may never be achieved.

Sensitivity of vergence responses of 5- to 10-week-old human infants

Infants have been shown to make vergence eye movements by 1 month of age to stimulation with prisms or targets moving in depth. However, little is currently understood about the threshold sensitivity of the maturing visual system to such stimulation. In this study, 5- to 10-week-old human infants and adults viewed a target moving in depth as a triangle wave of three amplitudes (1.0, 0.5, and 0.25 meter angles). Their horizontal eye position and the refractive state of both eyes were measured simultaneously. The vergence responses of the infants and adults varied at the same frequency as the stimulus at the three tested modulation amplitudes. For a typical infant of this age, the smallest amplitude is equivalent to an interocular change of approximately 2° of retinal disparity, from nearest to farthest points. The infants' accommodation responses only modulated reliably to the largest stimulus, while adults responded to all three amplitudes. Although the accommodative system appears relatively insensitive, the sensitivity of the vergence responses suggests that subtle cues are available to drive vergence in the second month after birth.

The (In)Effectiveness of Simulated Blur for Depth Perception in Naturalistic Images

We examine depth perception in images of real scenes with naturalistic variation in pictorial depth cues, simulated dioptric blur and binocular disparity. Light field photographs of natural scenes were taken with a Lytro plenoptic camera that simultaneously captures images at up to 12 focal planes. When accommodation at any given plane was simulated, the corresponding defocus blur at other depth planes was extracted from the stack of focal plane images. Depth information from pictorial cues, relative blur and stereoscopic disparity was separately introduced into the images. In 2AFC tasks, observers were required to indicate which of two patches extracted from these images was farther. Depth discrimination sensitivity was highest when geometric and stereoscopic disparity cues were both present. Blur cues impaired sensitivity by reducing the contrast of geometric information at high spatial frequencies. While simulated generic blur may not assist depth perception, it remains possible that dioptric blur from the optics of an observer’s own eyes may be used to recover depth information on an individual basis. The implications of our findings for virtual reality rendering technology are discussed.