Clinical vision

Cortical mechanisms of discomfort in migraine

Sarah Haigh, Olivera Karanovic, Frances Wilkinson, Nick Cooper and Arnold Wilkins

Migraineurs show subtle differences in perceptual sensitivity that have been attributed to cortical hyperexcitability. Coutts et al. [Coutts, Cooper, Elwell and Wilkins, 2012, Cephalalgia, in press] found that the haemodynamic response at the offset of aversive visual patterns is more rapid in migraineurs than non-migraineurs. We measured contrast thresholds for aversion to drifting, vibrating and static patterns in migraineurs and headache-free controls. For both migraineurs and controls, drifting and vibrating patterns were more aversive than static patterns. Static patterns are epileptogenic in patients with photosensitive epilepsy; vibrating patterns are more so and drifting patterns less so [Binnie, Findlay and Wilkins, 1985, Electroencephalogr Clin Neurophysiol,61,1-6]. There was therefore no simple relationship between the epileptogenic properties of the pattern and the aversion it evoked, even though, overall, the migraineurs showed greater discomfort than controls. We recorded the haemodynamic response to the patterns (using near-infrared spectroscopy). There were no differences in the amplitude of the haemodynamic response, but there was a steeper response at the offset of the drifting pattern, which was the most aversive. Our findings suggest that this difference between grating patterns (and between migraineurs and controls as shown by Coutts et al.) may reflect the aversion.

Neurological and neuropsychological characteristics of occipital stroke

Cathleen Grimsen, Antje Kraft, Maren Prass, Markus Bahnemann, Freimuth Brunner, Andreas Kastrup, Stephan Brandt and Manfred Fahle

We investigated a large number of stroke patients (N=128) to better characterize neurological and neuropsychological deficits caused by occipital damage. The aim was 1) a systematic description of subacute clinical and neuropsychological deficits and 2) to correlate the exact localization of lesions with specific perceptual deficits. Therefore we evaluated vision-related subjective and objective impairments by means of detailed anamnesis and clinical examination (such as Demtect, Rey-Osterrieth-Figure, Lang Stereo Test). Voxel-based lesion-symptom mapping (VLSM) was performed for those deficits appearing in at least ten patients. Subjectively, patients often reported visual field defects and phosphenes, as well as anomia, difficulties in reading and memory deficits. Visual disorders, such as achromatopsia, akinetopsia or prosopagnosia, were rarely reported. Objectively, a wide range of neuropsychological deficits were present. Visual field defects, reading disorders and the perception of phosphenes were associated with lesions of the calcarine sulcus. Anomia was associated with lesions of the occipital inferior gyrus while memory deficits were correlated with damage to the lingual gyrus and the hippocampus. Overall, we provide detailed descriptions of neurological and neuropsychological symptoms, infarct topography and the behavioral-structural relationship, as revealed by VLSM. The results underline the clinical relevance for neurological and neuropsychological assessments after occipital stroke.

Vision among sight recovery patients with various visual experience

Michaela Porubanova, Radovan Sikl and Michal Simecek

Sight recovery patients, both with acquired (Sikl et al., submitted) or congenital (Fine et al., 2003; Ostrovsky et al., 2006, 2009) blindness, show deficits particularly in the domains of face perception, visual space perception and object recognition. However, many patients with acquired blindness gain visual experience during short sight recovery periods, i.e., their blindness is repeatedly interrupted. It is not well known how these 'interruptions' facilitate patients` capacity for post-treatment improvement. We examined three sight-recovery patients with acquired blindness who differed in terms of period of blindness, and also in terms of presence of those short sight recovery periods during their blindness. Various experimental tasks examining above mentioned afflicted areas were used. The results illustrate that the patient with a continuous (uninterrupted) period of blindness shows remarkably impoverished performance especially in the domain of object recognition in sub-optimal conditions in comparison with two other patients with interrupted visual deprivation. The findings implicate the importance of short visual periods in maintenance of rich and robust visual representations in post-treatment visual development.

Investigating the connectivity and integrity of visual pathways following damage to V1

Sara Ajina, Rebecca Millington, Christopher Kennard and Holly Bridge

Few studies have investigated functional or structural connectivity in the visual pathway following visual cortex damage in adulthood. GY, who sustained left V1 damage aged 8 reportedly showed altered connectivity of thalamocortical and cortico-cortical visual pathways when compared to controls. The authors suggest this may indicate plasticity within the visual system. It remains unclear, however, whether any such structural changes are necessary for visual function, or indeed whether such changes are possible when damage occurs in adulthood. We recruited four patients with chronic V1 damage sustained in adulthood, three of whom showed BOLD activation of ipsilesional V5/MT with motion stimuli presented in the blind hemifield. We used diffusion-tensor imaging (DTI) to look for the presence of tracts subserving V5/MT, and to measure tract integrity. In both patients and age-matched controls we found pathways between V5/MT and LGN, as well as superior colliculus. Despite activation of V5/MT, the integrity of these pathways was impaired (mean FA=0.368 vs. 0.462 controls, 95%CI=-0.139,-0.04). Cortico-cortical connections between V5/MT bilaterally were present only in patients demonstrating V5/MT activation and tract numbers were significantly greater than controls (normalized mean=403.0 vs 42.3, 95% CI=27.78,693.6). Employing DTI to investigate changes in connectivity may uncover possible structural changes underlying blindsight.

In schizophrenia patients, high-level but not low-level motion processing is impaired

Farid Kandil, Anya Pedersen and Patricia Ohrmann

One of the prominent perceptuo-motor deficits in schizophrenia patients is the inability to perform smooth-pursuit eye-movements. While recent studies have shown that frontal motor areas as FEF are rather unaffected and that higher-order motion computation in area MT (motion integration, speed discrimination and conscious motion perception) is highly impaired, it has remained unclear whether also low level motion processing in V1 is compromised. Here we contrast performance of schizophrenia patients with normal controls in two tasks. In the first experiment, low-level motion information defines the figure in a figure-ground segregation task. In the second, similar motion information defines events that have to be used in a temporal order judgement task. While solving the first requires processing in V1 and the ventral pathway (unaffected in Schizophrenia), the second relies on processing in V1 and the dorsal pathway (compromised in Schizophrenia). Results show that patients'thresholds are greatly prolonged in the judgement task but only mildly impaired in the segregation task. As both tasks rely on the same low-level motion information the good results in the segregation task demonstrate that low-level motion processing is relatively spared in Schizophrenia. This result restrains the locus of the impairment in Schizophrenia to areas beyond the primary visual cortex and thus contributes to a finer localization of the deficit. For visual neuroscience, in turn, the findings add further evidence to the notion that also motion-defined figure-ground segregation is processed in the areas of the ventral pathway.

Measurements of Contrast Sensitivity Functions Show Recovery from Extended Blindness

Amy Kalia, Luis Lesmes, Michael Dorr, Peter Bex, Tapan Gandhi, Piyush Swami, Garga Chatterjee and Pawan Sinha

We report on the recovery of vision in ten patients who were treated for early onset (< 1 year) cataracts after long-term deprivation (8-15 years). Unlike previously reported cases of sight restoration [Fine, et al., 2003; Maurer, Ellemberg and Lewis, 2006], we had the opportunity to study a large number of patients in India who were blind since early in life and treated past the age for normal development of contrast sensitivity (Project Prakash). Patients ranged in age from 8 to 17 years and were treated 3 months to 2 years prior to testing. We used Lea symbols as stimuli and a Bayesian adaptive method to estimate contrast sensitivity functions [Lesmes, et al., 2010]. Results show that all patients recovered some vision in the lower spatial frequency range, and some patients even showed peak sensitivities (thresholds<1%) that were equivalent to sighted controls, but shifted to lower frequencies. The age, level of pre-operative vision (light perception vs. finger counting), and time since treatment did not predict the extent of recovery. These results suggest that the human visual system has the potential to develop contrast sensitivity after long-term deprivation, but it is still unknown which factors influence the extent of recovery.

How does the optical quality of the eye affect letter recognition and reading? Task-specific effects are revealed for optical distortions of text

Laura Young, Hannah Smithson, Gordon Love and Simon Liversedge

The optical components of the eye have inherent imperfections that degrade the image falling on the retina. Simple examples are defocus and astigmatism, which are routinely corrected, but higher-order aberrations remain. These optical distortions can have dramatic effects on the retinal image; making letters more difficult to identify and potentially making one more similar to another. We imposed defocus, coma and secondary astigmatism (higher-order astigmatism) in the rendering of text. The types of aberration can be compared on the same scale via an index of confusability based on letter-by-letter cross-correlation. These aberrations have qualitative differences: defocus and secondary astigmatism primarily affect letter form while coma smears the text in one direction whilst leaving the form relatively intact. In one experiment we measured subjects' contrast thresholds for single-letter identification and in another we measured reading performance (average fixation duration over a sentence). Defocus and secondary astigmatism specifically disrupted word recognition whereas coma did not. Coma was particularly disruptive with letter strings since in reading (but not letter identification) equivalent impairment occurred at lower confusability. This was possibly caused by crowding or disruptions to saccade planning since coma tends to fill in the spaces between letters and words.

Beyond blindsight: perceptual re-learning of visual motion discrimination in cortical blindness improves static orientation discrimination

Anasuya Das and Krystel R. Huxlin

Damage to the primary visual cortex or its immediate afferents results in a dense scotoma, termed cortical blindness (CB). A majority of CB subjects however, still retain the ability to detect stimuli that have high temporal and low spatial frequency content. Furthermore, recent work from our lab as well as others has shown the effectiveness of repeated perceptual stimulation in shifting the blindfield border. Specifically, we showed that global direction discrimination training restores direction integration at the trained, blindfield locations. Broad transfer of learning was observed across different moving stimuli. We asked whether visual motion relearning is specific to motion stimuli or whether it transfers to non-motion stimuli that traditionally do not elicit blindsight. We trained CB subjects on a global direction discrimination task in their blind field until they recovered normal direction integration thresholds. Contrast sensitivity for orientation discrimination of static gabors was measured pre- and post-training at the same locations. Motion training improved contrast sensitivity for static gabors presented at the training location, with greater improvements at spatial frequencies lower than 5cpd. Orientation difference thresholds also improved significantly for low frequency (0.5cpd) static gabors. These results suggest that perceptual learning in cortically blind fields generalizes to a broader range of stimuli than predicted by blindsight alone.

Effects of cloudiness on reading –A comparative experiment on the contrast polarity effect using a cloudy ocular media simulation–

Yasushi Nakano, Tetsuya Arai, Ataru Yoshino, Kensuke Oshima and Tsutomu Kusano

Legge et al. (1986) showed that reading performance is higher with white letters on a black background (white-on-black; W/B) than black letters on a white background (black-on-white; B/W) in eye disorders with cloudy ocular media. In this research, cloudy ocular media was simulated with a Bangerter filter, and an examination was conducted using MNREAD-J of which aspects that white/black inversion has an effect on (e.g., logMAR acuity, reading acuity, critical print size, maximum reading speed). The participants were 12 adults with normal visual function. Results of a t-test showed that for logMAR acuity, reading acuity, and critical print size, the white/black inversion condition exhibited significantly higher performance, but there was no difference in maximum reading speed. Also, the difference in performance was greatest for logMAR acuity (B/W condition 0.49, W/B condition 0.23), followed by reading acuity (B/W condition 0.45, W/B condition 0.32) and critical print size (B/W condition 0.73, W/B condition 0.60).

Training on lateral interactions improves peripheral visual acuity in patients with macular degeneration

Marcello Maniglia, Gianluca Campana, Andrea Pavan, Daniela Guzzon and Clara Casco

People suffering from macular degeneration are forced to use a new fixation point in the peripheral visual field (preferential retinal locus - PRL) for visual tasks previously performed by the macula. However, besides a different pattern of connectivity in the retina, the peripheral visual field is characterized, respect to the central visual field, by stronger inhibitory lateral interactions between detectors at early levels of cortical processing, which hampers visual resolution. Here we investigated whether training on lateral interactions could weaken these inhibitory interactions and whether learning transferred to untrained, higher-level visual functions (acuity, crowding). We previously reported that modulation of lateral interactions in the periphery with perceptual learning reduced crowding in normal-sighted subjects. Training consisted in a contrast detection of a Gabor target with iso-oriented high contrast Gabor flankers, at different target-to-flankers separations. Training increased contrast sensitivity at all separations, and, despite not having any effect on crowding, improved visual acuity and contrast sensitivity at the trained spatial frequencies. The absence of crowding reduction could be due to a different pattern of connectivity respect to normal-sighted subjects, resulting from the loss of central vision. Nevertheless, the improvement of visual acuity opens new perspectives for the rehabilitation of patients with macular degeneration.

Could contrast-modulated optotypes aid in the detection of amblyopia?

Sarah J. H. Lalor, Monika A. Formankiewicz and Sarah J. Waugh

Amblyopes show a specific loss of sensitivity to contrast-modulated stimuli and are thought to suffer more from crowding. Stronger contour interaction/crowding has been found for detecting large contrast-modulated (CM) letters, than for luminance-modulated (LM) letters (Chung et al, 2007, Journal of Vision, 7(2), 1-13); and for visual acuity using CM rather than LM Cs (Hairol et al, 2010, Journal of Vision, 10(7), 1332). Neither crowding, nor contour interaction, has been investigated with luminance-modulated and contrast-modulated, child-friendly optotypes. Contour interaction was assessed for LM and CM Kay Pictures, Lea Symbols and HOTV optotypes for normal adult observers. Visual acuity, with and without surrounding contours, was measured using a method of constant stimuli and contour separations of 0 to 10 stroke-widths. A repeated measures ANOVA with Greenhouse-Geisser correction found the magnitude of contour interaction to be significantly larger (at p<0.05) for CM optotypes (~1 line of acuity loss) than for LM optotypes (~0.5 line of acuity loss). The extent of crowding (in stroke widths) was not significantly different for the two stimulus types (p>0.10). The larger magnitude of contour interaction, combined with a specific loss of sensitivity to contrast-modulated stimuli in amblyopia, may mean that contrast-modulated optotypes could aid in amblyopia detection.

The effects of termination rules on contour interaction in a resolution acuity task with luminance-modulated and contrast-modulated Cs.

Monika A. Formankiewicz, Sarah J. Waugh and M. Izzuddin Hairol

The larger magnitude of contour interaction for visual acuity using contrast-modulated (CM) than luminance-modulated (LM) stimuli, may be useful in the detection of amblyopia (e.g. Hairol et al., 2010, Journal of Vision, 10(7), 1332). Performance level at which acuity is taken varies for different paradigms and charts. To investigate whether different termination rules affect the magnitude of contour interaction measured, we analysed psychometric functions for a LM and CM resolution task, with and without flanking bars. Psychometric function slopes were flatter for CM (~4.5) than LM stimuli (~5.0), but this does not contribute to the greater contour interaction reported for CM, than LM stimuli. Slopes for the abutting condition only, were different (flatter) than the isolated condition, especially for LM stimuli. This causes the peak of contour interaction to sometimes occur away from the abutting condition and its magnitude to increase up to 0.1 logMAR (1 line on a chart) with increasing performance level (and size). Different psychometric function slopes for abutting bars only, can affect contour interaction magnitudes reported in research, depending on termination rules. However clinically, where bars are placed some distance away from the target, magnitudes would not likely be affected.

Effective eye movements training of patients with tunnel vision- a randomized and controlled study.

Iliya V. Ivanov, Annika Vollmer, Claudia Gehrlich, Angelika Cordey, Nhung X. Nguyen and Susanne Trauzettel-Klosinski

Tunnel vision is a severe peripheral visual field loss condition, impairing mobility. It affects orientation, navigation and eye movement pattern while walking through obstacles in daily-life. We trained patients with tunnel vision caused by retinitis pigmentosa (RP) to extend their eye movements, causing increase in dynamic visual field (VF). Patients were randomly assigned into two training groups: saccadic (experimental) or reading (placebo). Eye positions while walking in real environment were recorded before and after training by Tobii Glasses Eye Tracker. Eye movement pattern was assessed by calculating the sample horizontal and vertical standard deviations (dispersions) of the eye positions. We compared dispersions of patients with reading and saccadic training with a control group of normals. Before and after training, saccadic patients with residual VF > 15° and reading patients did show wider horizontal than vertical dispersions, which were not significantly different from the normals. However, saccadic patients with residual VF < 15° showed equal horizontal and vertical dispersions before training. After training horizontal dispersion significantly increased. We conclude that the specific eye movement training assigned to the RP patients with VF < 15° effectively increased their dynamic VF.

BOLD evidence linking deficits in early vision to Schizophrenia.

Kiley Seymour, Timo Stein, Lia Sanders, Ines Theophil and Philipp Sterzer

It has been proposed that the cognitive deficits associated with schizophrenia can be attributed to an impaired use of contextual information to interpret stimuli. In support of this, recent evidence suggests that some low-level contextual interactions in vision may be weaker in schizophrenia, linking early neural processing to a more intrinsic disturbance of cortical architecture associated with the illness. One of the best-studied examples of contextual influence in vision is that of orientation processing. Here, in schizophrenic patients and healthy age-matched controls, we examine orientation-specific contextual suppression of the BOLD signal in occipital cortex. We find evidence indicating that basic visual processing of contextual information is altered in schizophrenia, supporting recent suggestions that early sensory disturbances may underlie the cognitive deficits commonly associated with the disorder.

Population receptive field mapping in patients with schizophrenia

Elaine J. Anderson, Marc S. Tibber, D. Samuel Schwarzkopf, Geraint Rees and Steven C. Dakin

Schizophrenia (SZ) is a disorder characterised not only by high-level cognitive dysfunction, but also by low-level changes in visual perception. In particular, patients with SZ are less susceptible to the influence of visual context, resulting in weak effects on a range of contextual visual illusions, including size, contrast, motion and orientation judgements. In healthy observers, the strength of these illusions anti-correlates with the surface area of V1 i.e. individuals with a large V1 experience weaker illusions. We therefore hypothesised that those patients with SZ who demonstrate reduced susceptibility to contextual effects will have larger V1 surface area, smaller population receptive field (pRF) sizes and increased cortical magnification factor (CMF). We used high-field fMRI to calculate pRF sizes, CMF and the surface area of V1 to V3 in 20 patients with SZ and 20 healthy controls. As predicted, patients with SZ had consistently smaller pRF sizes in V1- V3, compared to healthy controls. These functional differences in the architecture of the visual cortex in patients with SZ may play a causal role in the observed changes to visual perception.

Patterns of fMRI visual cortex activation in patients with hemianopia

Rebecca Millington, Clarissa Yasuda, Guilherme Beltramini, Panitha Jindahra, John Barbur, Gordon Plant, Christopher Kennard, Fernando Cendes and Holly Bridge

Homonymous field deficits reflect damage to the post-chiasmal visual pathway. Damage to different locations along this pathway can cause differences in visual field loss and varying access to subcortical nuclei. It is of interest to establish whether variability in residual visual function is related to the location of the lesion. We used fMRI to examine activity in striate cortex (V1) and the motion area (V5/MT) in fourteen patients with damage to the optic tract, lateral geniculate nucleus, optic radiation, or occipital cortex. Moving dot stimuli were employed to stimulate the affected hemifield, unaffected hemifield or centrally spanning both hemifields. All hemianopia patients showed reduced activity in the affected hemisphere, regardless of lesion location. Ten patients showed little or no occipital activation to stimuli presented in the blind hemifield. Quadrantanopia patients (n=3) showed no V1 activity to stimuli presented in the blind quadrant, but did exhibit activation in the affected hemisphere when the stimulus crossed the horizontal meridian. Interestingly, some patients showed unexpected activation patterns, such as similar contralesional V1 activation to stimuli presented in either hemifield. These unusual responses may reflect structural reorganisation of the visual pathways since they do not appear to be related to lesion age or location.

Multisensory Cortical Integration In Patients With Hereditary Cone Dystrophy: A fMRI And DTI Study

Sofia Crespi, Stefania Bianchi Marzoli, Giacinto Triolo, Francesco Bandello, Maria Concetta Morrone and Andrea Falini

Hereditary cone dystrophies (HCD) determine deprivation of normal visual input to the portion of V1 representing the macula. In congenitally blind subjects, Functional MRI studies (fMRI) demonstrated that the 'deafferented' V1 responds to stimuli of different sensory modalities. We aim to understand if a similar reorganization is present in HCD. In 7 HCD patients and 7 normal subjects we measured fMRI cortical response to audio, visual and audio-visual stimuli and used MR Diffusion Tensor Imaging (DTI) to quantify the integrity of the optic radiations and the connections between V1 and A1. The ON stimuli were expanding-contracting visual flow motion (extending 3 deg within the scotoma), auditory looming-receding flow motion, or a combination of the two. The OFF stimulus was a grey background with peripheral square stimulus to help fixation. Although in normal blind folded subjects no response to sound was ever observed in V1/V2, in HCD patients acoustic stimuli activated deafferented V1. DTI results revealed no differences in white matter microstructure between patients and controls along the Inferior Longitudinal Fasciculus or the Inferior fronto-occipital fasciculus. This data suggests a cortical reorganization of function in deafferented V1, despite the late onset of the disease and the lack of major anatomical changes.

Global motion coherence performance after extended blindness: stretching the window

Garga Chatterjee, Amy Kalia, Gandhi Tapan, Piyush Swami and Pawan Sinha

Global motion perception is an important building block of visual development. Is this ability available for use in individuals who gain sight late in life? Examples of global motion assessment after recovery from extended blindness are scanty (Ellemberg et al., 2002; Fine et al., 2003). As a part of Project Prakash, we studied 8 subjects who were blind for the first 8 to 17 years of life because of very-early onset bilateral cataracts. Global motion perception was assessed by measuring coherence thresholds using random dot kinetograms (Newsome and Paré, 1988). The assessments were conducted 4 months to 2 years post cataract removal. All subjects had above chance global motion coherence thresholds with the performance of some subjects approaching near-normal levels. Evidence for even partial acquisition of global motion coherence is significant in that it shows that even extended periods of blindness starting near birth do not entirely preclude the development of this ability. The present data greatly extends the window of very early onset blindness duration after which acquisition or development of global motion coherence sensitivity is possible.

Prism Adaptation in Cerebellar Patients

Karoline Spang, Dagmar Timann-Braun, Mitch Glickstein and Manfred Fahle

Eye-hand adaptation to prisms shifting the visual input horizontally is far more complex than meets the eye. Depending on the exact experimental conditions, the felt position of the hand adapts, and/or the perceived position of the eye in the orbit as well as the perceived position of the head relative to the trunk. We tested the textbook wisdom that cerebellar patients lack prism adaptation. 18 patients suffering from unilateral cerebellar lesions due to strokes and 14 students performed ballistic movements towards a visual target with and without prisms shifting the visual input by 170 to the left with the hands visible only at the endpoint of the movement. This procedure leads to an adaptation mostly of the visual input (felt eye position relative to trunk). Speed of adaptation and re-adaptation as well as size of both direct and after-effect did not differ between our group of patients and the control group of students! This difference to textbook results may be due to the fact that opposite to previous studies, we did not ask our patients to ignore their pointing errors, a procedure aimed at eliminating cognition-based adaptation. Still, we found a significantly smaller after-effect in the lesioned versus intact side.

Metamorphopsia and Visual Acuity

Emily Wiecek, Steven Dakin and Peter Bex

Metamorphopsia (visual distortion) is a common problem for patients with visual impairment, including macular degradation, macular edema, epiretinal membrane, and amblyopia. Metamorphopsia is currently assessed with Amsler charts and acuity; however, the contribution of distortion to visual acuity loss is unknown. Here we examined the relationship between spatial distortion (both intrinsic and extrinsic) and letter acuity. Four normally-sighted observers and two observers with self-reported metamorphopsia completed a (10AFC) Sloan-letter identification task on six letter sizes undergoing different levels of spatial distortion (applied using a pixel remapping algorithm). Metamorphopsia was confirmed and quantified with a dichoptic pointing task in which observers aligned two stimuli presented separately to either eye. Spatially band-passed filtered noise was used to remap the pixels and varied over six spatial frequencies and six amplitudes. Performance was modeled using ordinal logistic regression. Letter identification decreased as spatial frequency and magnitude of distortion increased (p<0.01) and letter size decreased (p<0.01). Normally-sighted observers performed significantly better than observers with metamorphopsia. Letter size and an interaction term between magnitude and letter size were significant predictors of performance. This nonlinear relationship means that acuity is an unreliable indicator of visual impairment in the presence of metamorphopsia.

Effects of different levels of retinal illuminance on grating and letter acuities using luminance- and contrast-modulated stimuli

Norsham Ahmad, Sarah J. Waugh and Monika A. Formankiewicz

Letter and grating acuities are degraded with decreasing retinal illuminance for standard luminance stimuli and investigating their interrelationship has proved helpful in understanding normal and amblyopic vision (Levi & Klein, 1982, Nature, 298, 268-270). Degrading normal spatial vision by using neutral density filters might mimic amblyopia. Grating and letter acuities were measured for luminance (L), luminance-modulated noise (LM) and contrast-modulated noise (CM) Gabors and C stimuli. Observers wore goggles that incorporated a 2.5mm diameter pupil and different neutral density filters resulting in retinal illuminances between ~0.5-500 tds. Grating acuities were degraded with reduced retinal illuminance at a steeper rate for CM stimuli, depending on the noise (slope ~0.4-0.7) than for L and LM stimuli (slope ~0.3). Letter acuities for all stimulus types behaved similarly (~0.4-0.5). A comparison of the effects on gratings versus letters showed approximate equivalence; or greater effects on gratings for CM stimuli were found. For L and LM stimuli, reduced retinal illuminance degrades normal vision in a way similar to that previously found in anisometropic amblyopia. It is unlike the effects found in strabismic amblyopia, where letter acuity is more affected than grating acuity. For most CM stimuli, degraded vision affects grating acuity more than letters.