Search

Reliability-weighted averaging of multiple perceptual estimates (or cues) can improve precision. Research suggests that newly-learned statistical associations can be rapidly integrated in this way for efficient decision-making. Yet, it remains unclear if integration of newly-learned statistics into decision-making can directly influence perception, rather than taking place only at a decision stage. In two experiments, we implicitly taught observers novel associations between shape and colour. Observers made colour matches by adjusting the colour of an oval to match a simultaneously presented reference. As the colour of the oval changed across trials, so did its shape according to a novel mapping of axis ratio to colour. Observers showed signatures of reliability-weighted averaging – a precision-improvement in both experiments and reweighting of the newly-learned shape cue with changes in uncertainty in Experiment 2. To ask whether this was accompanied by perceptual effects, Experiment 1 tested for “forced fusion” by measuring colour discrimination thresholds with and without incongruent novel cues. Experiment 2 tested for a “memory colour effect”, observers adjusting the colour of ovals with different axis ratios until they appeared grey. There was no evidence for forced fusion and the opposite of a memory colour effect. Overall, our results suggest that the ability to quickly learn novel cues and integrate them with familiar cues is not immediately (within the short duration of our experiments, and in the domain of colour and shape) accompanied by common perceptual effects.

Cue combination describes the use of two sensory cues together to increase perceptual precision. Internal relative bias describes a situation in which two cues to the same state of the world are perceived as signaling different states of the world on average. Current theory and evidence have difficulty accounting for many instances where cue combination is absent, such as in children under 10 years old, and in a variety of tasks. Here we show that internal relative biases between cues could be a key explanatory factor. Experiment 1, studying children's three-dimensional (slant) perception via disparity and texture, found a negative cross-sectional correlation between internal relative bias and cue combination behavior in 7- to 10-year-olds. Strikingly, children who had below-median levels of internal relative bias were able to combine cues, unlike the typical result for that age range. Experiment 2, studying adults' visual-auditory localization, found that cue combination behavior increased after an intervention designed to decrease internal relative bias. We interpret this as strong but preliminary evidence that internal relative bias can disrupt cue combination behavior. This provides a plausible mechanism to explain why children under 10 generally do not combine cues and why the audiovisual cue combination is so inconsistent in adults. Moving forward, we suggest that researchers who fail to find an expected cue combination effect should further investigate the possibility of issues with internal relative bias. Decreasing internal relative bias may also be an important goal for rehabilitation and sensory substitution or augmentation approaches to promoting efficient multisensory perception.

Previous studies suggest that to achieve color constancy, the human visual system makes use of multiple cues, including a priori assumptions about the illumination (“daylight priors”). Specular highlights have been proposed to aid constancy, but the evidence for their usefulness is mixed. Here, we used a novel cue-combination approach to test whether the presence of specular highlights or the validity of a daylight prior improves illumination chromaticity estimates, inferred from achromatic settings, to determine whether and under which conditions either cue contributes to color constancy. Observers made achromatic settings within three-dimensional rendered scenes containing matte or glossy shapes, illuminated by either daylight or nondaylight illuminations. We assessed both the variability of these settings and their accuracy, in terms of the standard color constancy index (CCI). When a spectrally uniform background was present, neither CCIs nor variability improved with specular highlights or daylight illuminants (Experiment 1). When a Mondrian background was introduced, CCIs decreased overall but were higher for scenes containing glossy, as opposed to matte, shapes (Experiments 2 and 3). There was no overall reduction in variability of settings and no benefit for scenes illuminated by daylights. Taken together, these results suggest that the human visual system indeed uses specular highlights to improve color constancy but only when other cues, such as from the local surround, are weakened.

Prior knowledge can help observers in various situations. Adults can simultaneously learn two location priors and integrate these with sensory information to locate hidden objects. Importantly, observers weight prior and sensory (likelihood) information differently depending on their respective reliabilities, in line with principles of Bayesian inference. Yet, there is limited evidence that observers actually perform Bayesian inference, rather than a heuristic, such as forming a look-up table. To distinguish these possibilities, we ask whether previously learned priors will be immediately integrated with a new, untrained likelihood. If observers use Bayesian principles, they should immediately put less weight on the new, less reliable, likelihood ("Bayesian transfer"). In an initial experiment, observers estimated the position of a hidden target, drawn from one of two distinct distributions, using sensory and prior information. The sensory cue consisted of dots drawn from a Gaussian distribution centered on the true location with either low, medium, or high variance; the latter introduced after block three of five to test for evidence of Bayesian transfer. Observers did not weight the cue (relative to the prior) significantly less in the high compared to medium variance condition, counter to Bayesian predictions. However, when explicitly informed of the different prior variabilities, observers placed less weight on the new high variance likelihood ("Bayesian transfer"), yet, substantially diverged from ideal. Much of this divergence can be captured by a model that weights sensory information, according only to internal noise in using the cue. These results emphasize the limits of Bayesian models in complex tasks.

Purpose.: To validate a novel, automated test of infant resolution acuity based on remote eye-tracking.\ud \ud Methods.: Infants aged 2 to 12 months were tested binocularly using a new adaptive computerized test of infant vision using eye tracking (ACTIVE), and Keeler infant acuity cards (KIAC). The ACTIVE test ran automatically, using remote eye-tracking to assess whether the infant fixated a black-and-white grating of variable spatial frequency. Test-retest reliability was assessed by performing each test twice. Accuracy was assessed by comparing acuity measures across tests and with established age-norms, and by comparing low-contrast acuity estimates in adults with data reported previously.\ud \ud Results.: All infants completed the ACTIVE test at least once. Median test duration was 101 seconds. Measured visual acuity increased with age (P < 0.001), and 90% of mean acuity estimates were within previously published 90% tolerance limits (based on acuity-card age norms). Acuity estimates were also correlated, within-subjects, with results from the KIAC (P = 0.004). In terms of reliability, 86% of acuity estimates deviated by ≤1 octave, with no significant difference in test-retest reliability between the ACTIVE and KIAC procedures (P = 0.461). In adults, acuity estimates from the ACTIVE test did not differ significantly from values reported by previous authors (P > 0.183).\ud \ud Conclusions.: An adaptive computerized test of infant vision using eye-tracking provides a rapid, automated measure of resolution acuity in preverbal infants. The ACTIVE performed comparably to the current clinical gold standard (acuity cards) in terms of testability, reliability, and accuracy, and its principles can be extended to measure other visual functions.

Purpose To characterize retinal structure and function in achromatopsia (ACHM) in preparation for clinical trials of gene therapy. Design Cross-sectional study. Participants Forty subjects with ACHM. Methods All subjects underwent spectral domain optical coherence tomography (SD-OCT), microperimetry, and molecular genetic testing. Foveal structure on SD-OCT was graded into 5 distinct categories: (1) continuous inner segment ellipsoid (ISe), (2) ISe disruption, (3) ISe absence, (4) presence of a hyporeflective zone (HRZ), and (5) outer retinal atrophy including retinal pigment epithelial loss. Foveal and outer nuclear layer (ONL) thickness was measured and presence of hypoplasia determined. Main Outcome Measures Photoreceptor appearance on SD-OCT imaging, foveal and ONL thickness, presence of foveal hypoplasia, retinal sensitivity and fixation stability, and association of these parameters with age and genotype. Results Forty subjects with a mean age of 24.9 years (range, 6–52 years) were included. Disease-causing variants were found in CNGA3 (n = 18), CNGB3 (n = 15), GNAT2 (n = 4), and PDE6C (n = 1). No variants were found in 2 individuals. In all, 22.5% of subjects had a continuous ISe layer at the fovea, 27.5% had ISe disruption, 20% had an absent ISe layer, 22.5% had an HRZ, and 7.5% had outer retinal atrophy. No significant differences in age ( P = 0.77), mean retinal sensitivity ( P = 0.21), or fixation stability ( P = 0.34) across the 5 SD-OCT categories were evident. No correlation was found between age and foveal thickness ( P = 0.84) or between age and foveal ONL thickness ( P = 0.12). Conclusions The lack of a clear association of disruption of retinal structure or function in ACHM with age suggests that the window of opportunity for intervention by gene therapy is wider in some individuals than previously indicated. Therefore, the potential benefit for a given subject is likely to be better predicted by specific measurement of photoreceptor structure rather than simply by age. The ability to directly assess cone photoreceptor preservation with SD-OCT and/or adaptive optics imaging is likely to prove invaluable in selecting subjects for future trials and measuring the trials' impact.

PURPOSE: To report on two population screening programs designed to detect significant refractive errors in 8308 8- to 9-month-old infants, examine the sequelae of infant hyperopia, and test whether early partial spectacle correction improved visual outcome (strabismus and acuity). The second program also examined whether infant hyperopia was associated with developmental differences across various domains such as language, cognition, attention, and visuomotor competences up to age 7 years. Linked programs in six European countries assessed costs of infant refractive screening. METHOD: In the first program, screening included an orthoptic examination and isotropic photorefraction, with cycloplegia. In the second program we carried out the same screening procedure without cycloplegia. Hyperopic infants (> or = +4 D) were followed up alongside an emmetropic control group, with visual and developmental measures up to age 7 years, and entered a controlled trial of partial spectacle correction. RESULTS: The second program showed that accommodative lag during photorefraction with a target at 75 cm (focus > or = +1.5 D) was a marker for significant hyperopia. In each program, prevalence of significant hyperopia at 9 to 11 months was around 5%; manifest strabismus was 0.3% at 9 months and 1.5 to 2.0% by school age. Infant hyperopia was associated with increased strabismus and poor acuity at 4 years. Spectacle wear by infant hyperopes produced better visual outcome than in uncorrected infants, although an improvement in strabismus was found in the first program only. The corrections did not affect emmetropization to 3.5 years; however, both corrected and uncorrected groups remained more hyperopic than controls in the preschool years. The hyperopic group showed poorer overall performance than controls between 1 and 7 years on visuoperceptual, cognitive, motor, and attention tests, but showed no consistent differences in early language or phonological awareness. Relative cost estimates suggest that refractive screening programs can detect visual problems in infancy at lower overall cost than surveillance in primary care. CONCLUSIONS: Photo/videorefraction can successfully screen infants for refractive errors, with visual outcomes improved through early refractive correction. Infant hyperopia is associated with mild delays across many aspects of visuocognitive and visuomotor development. These studies raise the possibility that infant refractive screening can identify not only visual problems, but also potential developmental and learning difficulties.

Advances in potential treatments for rod and cone dystrophies have increased the need to understand the contributions of rods and cones to higher-level cortical vision. We measured form, motion and biological motion coherence thresholds and EEG steady-state visual evoked potentials (SSVEP) responses under light conditions ranging from photopic to scotopic. Low light increased thresholds for all three kinds of stimuli; however, global form thresholds were relatively more impaired than those for global motion or biological motion. SSVEP responses to coherent global form and motion were reduced in low light, and motion responses showed a shift in topography from the midline to more lateral locations. Contrast sensitivity measures confirmed that basic visual processing was also affected by low light. However, comparison with contrast sensitivity function (CSF) reductions achieved by optical blur indicated that these were insufficient to explain the pattern of results, although the temporal properties of the rod system may also play a role. Overall, mid-level processing in extra-striate areas is differentially affected by light level, in ways that cannot be explained in terms of low-level spatiotemporal sensitivity. A topographical shift in scotopic motion SSVEP responses may reflect either changes to inhibitory feedback mechanisms between V1 and extra-striate regions or a reduction of input to the visual cortex. These results provide insight into how higher-level cortical vision is normally organised in absence of cone input, and provide a basis for comparison with patients with cone dystrophies, before and after treatments aiming to restore cone function.

Accurate measures of perceptual threshold are difficult to obtain in infants. In a clinical context, the challenges are particularly acute because the methods must yield meaningful results quickly and within a single individual. The present work considers how best to maximize speed, accuracy, and reliability when testing infants behaviorally and suggests some simple principles for improving test efficiency. Monte Carlo simulations, together with empirical (visual acuity) data from 65 infants, are used to demonstrate how psychophysical methods developed with adults can produce misleading results when applied to infants. The statistical properties of an effective clinical infant test are characterized, and based on these, it is shown that (a) a reduced (false-positive) guessing rate can greatly increase test efficiency, (b) the ideal threshold to target is often below 50% correct, and (c) simply taking the max correct response can often provide the best measure of an infant's perceptual sensitivity.

International audience

A technique commonly used to assess infants' visual acuity is forced-choice preferential looking, which relies on the propensity of infants to look at patterned stimuli over uniform grey. Existing tests use trained examiners to judge where an infant fixates; however, eye trackers offer the opportunity to measure eye movements quickly and automatically. In the present study infants' visual acuity was measured using a remote Tobii eye tracker (Tobii T120). Thirteen infants aged 4–8 months were seated before a screen and presented with a preferential looking task. On each trial a black and white chequerboard of variable spatial frequency was displayed on half of the screen, while uniform grey was displayed elsewhere. Eye tracking data were used to determine whether the infant fixated the chequerboard and to update the spatial frequency on subsequent trials according to a weighted up-down staircase targeting the 75% threshold. This initial study showed data in line with age norms are consistent across two testing sessions for some infants. Others yielded poor data or inconsistent performance. Future modifications will aim to make the test more engaging and to optimise its power and speed. These will provide a basis for a full assessment of the test's accuracy and reliability.

Unlike adults, children aged below 12 years did not use multiple depth cues to reduce their uncertainty when judging surface slant (Nardini et al 2010, PNAS 107:39, 17041–17046). This suggests that Bayesian cue integration in the human visual system is not present until late in its development. Here we asked whether individuals at the ‘transitional’ ages of 8 to 12 years show consistent cue integration behaviour across two testing sessions. Identifying observers matched in age who reliably do/do not integrate depth cues would allow us to investigate neuronal mechanisms underlying these individual differences. Thirty observers judged which surface, a 45° standard or a variable comparison, was the most slanted, based on texture alone, disparity alone, or both. The test was repeated, giving a total of 6 (3 conditions x 2 sessions) thresholds per observer. In a principal component analysis (PCA) of distributions of thresholds, the first component (60% of variance) described variation in global ability, while the second (19% of variance) described variation in the balance between single-cue and combined cue thresholds. Thus, children at this age vary in their use of combined depth cues, and do so consistently in measures taken across two sessions. Analyses of correlation and error quantify the stability of observers’ thresholds. Methods for identifying individuals who ‘do’ versus ‘do not’ integrate cues are suggested, based on PCA with elimination of unstable or outlying individuals. These findings provide a basis for investigating neuronal changes underlying development of the ability to integrate visual information to reduce uncertainty.

The Second Cambridge Population Infant Vision Screening Programme using the VPR-1 videorefractor without cycloplegia was undertaken in order to identify those infants with refractive errors who were potentially amblyogenic or strabismogenic. Infants identified at eight months were entered into a control trial of treatment with partial spectacle correction and underwent a long-term follow-up that monitored a wide range of visual, visuoperceptual, visuocognitive, visuomotor, linguistic and social development. In the present paper, the authors report on the outcome measures of visual acuity and strabismus. Poor acuity was defined as a best-corrected acuity of 6/12 or worse on crowded letters or 6/9 or worse on single letters, at age 4 years. Acuity was measured in 79 infants who were significantly hyperopic and/or anisometropic at 11-12 months of age, 23 who showed hyperopia of +3D but less than +3.5D, 196 control subjects, 14 controls with refractive errors, and 126 others who showed an accommodative lag on screening but were not significantly hyperopic on first retinoscopy. There was a poorer acuity outcome in the untreated group of hyperopes compared to controls (p < 0.0001) and to the children who were compliant in spectacle wear (p < 0.001) or who were prescribed spectacles (p < 0.05). Children who were significantly hyperopic at eight months were also more likely to be strabismic by 5.5 years compared to the emmetropic control group (p < 0.001). However, the present study did not find a significant difference in the incidence of strabismus between corrected and uncorrected hyperopic infants. Children who were not refractively corrected for significant hyperopia were four times more likely to have poor acuity at 5.5 years than infants who wore their hyperopic correction, supporting the findings of the First Cambridge Population Infant Vision Screening Programme.

In a population-based infant vision screening programme, 5295 infants were screened and those with significant refractive errors were followed up. To assess the relationship between the development of vision and other domains, we report a longitudinal study comparing infants with significant hyperopia, identified at age 9 months ('hyperopes') with infants with normal refractions ('controls'). Children are included who completed at each age a broad set of visual, cognitive, motor and language measures taken over a series of follow-up visits up to age 5.5 years. Hyperopes performed significantly worse than controls on the Atkinson Battery of Child Development for Examining Functional Vision at 14 months and 3.5 years and the Henderson Movement Assessment Battery for Children at 3.5 and 5.5 years. The Griffiths Child Development Scales, MacArthur Communicative Development Inventory and British Picture Vocabulary Scales showed no significant differences. Exclusion of those infants who became amblyopic and strabismic did not substantially alter these results, suggesting that the differences between groups were not a consequence of these disorders. These results indicate that early hyperopia is associated with a range of developmental deficits that persist at least to age 5.5 years. These effects are concentrated in visuocognitive and visuomotor domains rather than the linguistic domain.

Purpose. To longitudinally characterize retinal structure and function in achromatopsia (ACHM) in preparation for clinical gene therapy trials. Methods. Thirty-eight molecularly confirmed ACHM subjects underwent serial assessments, including spectral domain optical coherence tomography (SD-OCT), microperimetry, and fundus autofluorescence (FAF). Foveal structure on SD-OCT was graded and compared for evidence of progression, along with serial measurements of foveal total retinal thickness (FTRT) and outer nuclear layer (ONL) thickness. Fundus autofluorescence patterns were characterized and compared over time. Results. Mean follow-up was 19.5 months (age range at baseline, 6–52 years). Only 2 (5%) of 37 subjects demonstrated change in serial foveal SD-OCT scans. There was no statistically significant change over time in FTRT (P = 0.83), ONL thickness (P = 0.27), hyporeflective zone diameter (P = 0.42), visual acuity (P = 0.89), contrast sensitivity (P = 0.22), mean retinal sensitivity (P = 0.84), and fixation stability (P = 0.58). Three distinct FAF patterns were observed (n = 30): central increased FAF (n = 4), normal FAF (n = 11), and well-demarcated reduced FAF (n = 15); with the latter group displaying a slow increase in the area of reduced FAF of 0.03 mm2 over 19.3 months (P = 0.002). Conclusions. Previously published cross-sectional studies have described conflicting findings with respect to the age-dependency of progression. This study, which constitutes the largest and longest prospective longitudinal study of ACHM to date, suggests that although ACHM may be progressive, any such progression is slow and subtle in most patients, and does not correlate with age or genotype. We also describe the first serial assessment of FAF, which is highly variable between individuals, even of similar age and genotype.

Studying how we retrace our way in darkness provides a means of understanding the internal representations we use during navigation. In adults, path integration in darkness is influenced by previously presented visual information that conflicted with actual motion (Tcheang et al, PNAS, 2011, 108(3): 1152-7). Here we used immersive virtual reality to isolate visual and motor cues during encoding of a path in children and adults, and examined whether the cues' relative contributions to the stored path representation reflect optimal integration. Eighteen adults and fifteen 10-11-year-old children were guided along a two-leg path in darkness (motor-only), in a virtual room (visual-motor), or they watched a pre-recorded walk in the virtual room while standing still (visual-only). Participants then reproduced the path in darkness. By fitting bivariate normal distributions we obtained a measure of the dispersion of the end-points (variable error) and of their distances from the correct end point (constant error). While adults performed similarly across all three conditions, 10-11-year-old children reduced their variable error when encoding in the bimodal condition, indicating integration of cues. Both adults and children showed greater constant error in the visual-only condition than the motor-only (although significantly greater in adults than children), while the constant error for the visual-motor was intermediate between these two. A significant inverse correlation between visual-motor constant error and number of trials indicated that with only 10 repetitions children learnt to better combine the two cues. These results indicate that 10-11 year-old children can learn to optimally combine visual and motor information to encode a path. Adults' inability to do this could reflect their complete reliance on internal motion-related information because of its greater perceived reliability and relevance to the task. This suggests that the representations we use to navigate in darkness change significantly during the lifespan.

To perform everyday tasks, such as crossing a road, we greatly rely on our sight. However, certain situations (e.g., an extremely dark environment) as well as visual impairments can either reduce the reliability of or completely remove this sensory information. In these cases, the use of other information is vital. Here we seek to examine the development of haptic and auditory integration. Three different groups of adults and 5- to 12-year-old children were asked to judge which of a standard sized and a variably sized ball was the largest. One group performed the task with auditory information only, haptic only or both. Auditory information about object size came from the loudness of a naturalistic sound played when observers knocked the ball against a touch-pad. A second group performed the same conditions, while wearing a thick glove to reduce the reliability of the haptic information. Finally, a third group performed the task with either congruent or incongruent information. Psychometric functions were fitted to responses in order to measure observers’ sensitivities to object size under these different conditions. Integration of haptic and auditory information predicts greater sensitivity in the bimodal condition than in either single-modality condition. Initial results show that young children do not integrate information from haptic and auditory modalities, with some children aged below 8 years performing worse in the bimodal condition than in the auditory-only condition. Older children and adults seem able to integrate auditory and haptic information, especially when the reliability of the haptic information is reduced.