Retinopathy and Visual Dysfunction

This session will review recent advances in the detection of early changes in oculo-visual function leading to blindness, how genetic polymorphisms influence susceptibility to retinal disease and how stem cell therapy and transplants of normal RPE within the same eye are advancing treatment. The speakers will describe the latest functional and structural techniques performed in animals or humans, how association between structural changes and loss of visual sensitivity and the importance of how identification of polymorphic changes in the gene encoding complement factor H might be associated with susceptibility to retinal disease.

Moderator: Peter Gouras, Columbia University

9:15 AM: Fred Fitzke, University College London

Retinal imaging and functional studies in the living human eye to reveal fine structural changes that accompany loss of visual sensitivity

In patients with hereditary retinal degenerations and age related macular degeneration (AMD) novel optical imaging techniques provide new opportunities for understanding the biological basis of visual loss and determining whether innovative forms of intervention are effective in preventing blindness. Advances in Adaptive Optics imaging, Optical Coherence Tomography and Molecular Imaging allow the investigation of photoreceptor distribution, thickness of the different retinal layers and the effects of specific molecular species associated with the pathologies. Autofluorescence (AF) imaging which targets lipofuscin in the Retinal Pigment Epithelial cell layer allows the study of its role in Stargardts disease and AMD. Newer methods currently being used in animal models are expected to soon be translated to patient investigations. These include direct visualisation of the fundamental biological process of apoptosis (programmed cell death) in the living eyes of rodent and primate models. The psychophysical measurement of rod and cone mediated visual function on a microscopic scale in the eyes of patients with localised regions of retinal dysfunction provides an understanding of the functional consequences of the abnormalities seen by the imaging studies. In some cases of AMD striking abnormalities in AF can co-exist with nearly normal cone function while rod function is severely compromised. These techniques allow investigation of the biological basis of visual loss, its natural history and the potential beneficial effects of novel treatments including gene therapy.

9:35 AM: Yi-Zhong Wang, Retina Foundation of the Southwest

Early detection/assessment/monitoring of vision loss in AMD and Stargardt disease

Recent progress in understanding the pathogenesis of retinal degenerative diseases, such as age-related or juvenile macular degeneration, has led to the exciting possibility of treatment trials for these diseases. It is well documented that early detection of eye diseases or early detection of the change of disease status is crucial to successful treatment. Furthermore, the effectiveness of any treatment for eye diseases must ultimately be assessed in terms of its ability to preserve visual function. Hence, there is an increasing demand for new visual function tests that are more sensitivity than current standard tests such as visual acuity in evaluating vision loss in macular degeneration. Given that the damage from macular degeneration is inhomogeneous across the macula and the subretinal deposits do not always affect visual acuity, we hypothesized that it would be more difficult for patients with macular degeneration to perform distortion detection tasks requiring global visual integration over a larger retinal area than to perform a localized test such as visual acuity. We have developed a new shape discrimination paradigm to test this hypothesis, and have shown that shape discrimination sensitivity is significantly reduced in patients with macular degeneration, even though they may still retain normal visual acuity. We are evaluating the efficacy of the shape discrimination test, together with other functional tests and retinal imaging methods, in assisting early detection, monitoring the progression, and evaluating treatment outcomes of macular degeneration.

9:55 AM: David G. Birch, Retina Foundation of the Southwest

The role of electrophysiology in detecting and following retinal dystrophies

Full-field electroretinography (ERG) has long been recognized as a useful tool for the detection and characterization of retinal degenerative disease (RDDs). Traditional ISCEV-standard ERGs separate rod-mediated from cone-mediated responses, but can only discriminate among patients on the basis of peak-to-peak amplitude and b-wave implicit time. In an attempt to broaden the phenotypic assessment to better match genetic heterogeneity, we have sought to identify physiologically relevant parameters in the photoreceptor component that relate to activation and deactivation parameters of phototransduction. These new parameters help discriminate among patients with distinct disease-causing mutations.

Clinical trials in RDDs benefit from the detection of sub-microvolt responses in advanced disease and from the analysis of local responses from discrete regions. To provide a more comprehensive analysis of local function, we are attempting to relate multifocal ERG amplitudes and static perimetric thresholds to fine structural measures obtained through spectral domain optical coherence tomography (SD-OCT).

10:15 AM: Peter Coffey, University College London

Animal models of ARMD and RPE transplant in human patients

The London Project to Cure Blindness aims to make the most of human embryonic stem cells to prevent blindness and restore sight in patients with age-related macular degeneration (AMD) by 2011. Our goal is to replace cells essential for "seeing" lost by disease at the back of the eye. We aim to repair and regenerate the aged diseased eye using human embryonic stem cells which have been transformed into the cells affected in AMD: the support cells for the photoreceptors (retinal pigment epithelium) and the photoreceptors. The cells will be surgically implanted into a clinical population of AMD patients.

10:35 AM: Discussion

Close Window

Contributed Talk Session: Color

Moderator: Jay Neitz, Medical College of Wisconsin

1:45 PM: Osamu Masuda, University of Rochester

Arrangement of the trichromatic cone mosaic in peripheral retina of a color-normal and a deutan carrier

Previous studies have concluded that the packing arrangement of foveal L and M cones is not distinguishable from random in most eyes. However, these measurements have been confined to retinal eccentricities of only about 1 deg from the foveal center. Here, we measured the packing arrangement in extrafoveal retina.

We classified the L, M, and S cones in a deutan carrier and a color-normal male subjects with adaptive optics imaging combined with retinal densitometry at 1.25, 4, and 10 deg in the temporal retina. We evaluated the packing arrangement of the 3 cone classes by comparing the frequencies of distances between all cones of the same type with those expected based on a random pigment assignment rule.

Peripheral L and M cones of the deutan carrier exhibited significant clumping at 4 deg whereas those at 1.25 deg did not. The arrangement in the carrier at 10 deg had a tendency toward clumping but was not significant. The arrangement of the normal male was not significantly different from random both in the parafovea and in the periphery.

The organization of L and M cones outside the fovea, in a deutan carrier, shows a tendency toward clumping of cones of like type, as expected from X chromosome inactivation. This clumping may have implications for the strength of red-green color vision in peripheral retina since it increases the probability that peripheral midget cell centers will be driven by predominantly one class of cone.

2:00 PM: Paul Martin, National Vision Research Institute of Australia

Contribution of blue (S) cone signals to classical and extraclassical receptive fields in the lateral geniculate nucleus

Extraclassical inhibition (ECI) in the retina and lateral geniculate nucleus (LGN) acts to suppress responses to stimuli presented in the classical receptive field (CRF). Here we asked 1) do koniocellular (blue-on and blue-off) cells exhibit ECI? 2) does the ECI show chromatic selectivity? 3) what is the relative strength of short-wavelength sensitive (S, "blue") and medium/long wavelength sensitive (ML) cone inputs to CRF and ECI? We recorded extracellular action potentials from the LGN of sufentanil-anesthetized marmosets. Weights of S and ML cone inputs were measured using 1) cone-selective and achromatic stimuli presented in variable apertures and annuli, 2) modulation in an ML/S color plane [1], and 3) modulation of chromaticity and luminance through the D65 white point [2]. We found ECI in all cell classes is dominated by ML cone inputs. Average response attenuation (S-cone selective annuli vs. achromatic annuli) was 6% vs. 52% for PC cells (n=50), 20% vs. 63% for MC cells (n=10), 20% vs 45% for blue-on cells (n=21), and 19% vs. 43% for blue-off cells (n=3). The MC and PC cells we recorded only rarely received detectible input from S cones. Where detectible, S cone inputs mostly influenced PC cells, and were weak and inhibitory. We conclude that S cone signals are functionally isolated to the classical receptive field of blue-on and blue-off cells in the LGN.

2:15 PM: Rigmor Baraas, Buskerud University College

Comparison of local versus lateral S-cone modulation on rod thresholds

Rod and S-cone signals have been shown to interact in an additive manner at threshold. S-cone ganglion cells receive color-opponent input without center-surround antagonistic structure, which is unlike ganglion cells that receive color-opponent input from L- and M-cones. Interaction between rods and S-cones may therefore depend on whether rod and S-cone modulation covers the same spatial location (local) or different spatial locations (lateral) (see review by Buck 2004).

Here, rod thresholds were measured at different phases for a rod pulse superimposed on an Scone sinusoidally-modulated background. Rod and S-cone isolating stimuli (silent substitution) were generated with a 2-channel 4-primary Maxwellian view system. Detection thresholds for a 100-ms 1-deg rod-isolating pulse were measured against one of four possible adapting backgrounds: 1) a 12-deg steady full-field, 2) a 1-deg cone-modulating field (local), 3) a 12-deg cone-modulating annulus with a 1-deg inner diameter (lateral) or 4) a 12-deg cone-modulating full-field (local and lateral together) (Lee, Decay, Smith & Pokorny, 1999). The cone modulation was a temporal sinusoid along the S axis and did not modulate rods. The stimulus was presented 6-deg extra-foveal (nasal) at 200 Td and 20 Td.

Results show that, at 200 Td, rod thresholds varied with phase in a systematic way for all conditions tested. At 20 Td, however, the variation with phase was only observed for the fullfield condition. This suggests that the spatial configuration of the adapting background affects the type of interaction observed, but differently to what is seen for rod and L-/M-cone interaction.

2:30 PM: Andy Salzwedel, Medical College of Wisconsin

Functional magnetic resonance imaging (fMRI) investigation of the circuitry for blue-yellow color vision

The neural circuit underlying blue-yellow color vision in the rat was investigated using high resolution functional magnetic resonance imaging (fMRI) under a variety of stimulus conditions. Rodents and other mammals share evolutionarily conserved neural circuits for processing the output of two classes of spectrally distinct cone photoreceptor. Unlike other mammals, their availability combined with their brain and body size make them ideal subjects for experiments using a high field strength 9.4T MRI system coupled with pharmacological intervention. Herein, the well established agonist of metabotropic glutamate receptors, 2-amino-4-phosphonobutyric acid (APB), was used intravitreously to block the ON-pathways in the rat retina. A computer controlled binocular visual stimulator was designed to operate in the MRI scanner. Light from arrays of colored LEDs was delivered via fiber optic bundles to produce binocular stimulation. fMRI responses were compared under stimulus conditions that isolated either S or M cones. Regions analyzed in the functional images included the dorsal lateral geniculate nucleus, the lateral posterior nucleus, the superior colliculus, the primary visual cortex, and higher visual areas. Standard theory attributes the perception of blueness to comparisons between S-cones and M-cones made by small bistratified ganglion cells. However, when the ON-pathways (i.e. S-cone input) were inhibited the fMRI showed cortical activation, indicating that the small bistratified cells normally provide inhibitory rather than excitatory input to cortex. This suggests that the sensation of blueness is mediated by OFF responses. Based on known circuitry the best candidate for generating these OFF responses is horizontal cell feedback from S-cones on neighboring M-cones.

2:45 PM: Sang Wook Hong, Vanderbilt University

Interocular suppression selectively affects achromatic and chromatic pathways

Results from a series of psychophysical experiments show that interocular suppression produced by continuous flash suppression (CFS) differentially affects visual features of a target viewed by the other eye. When CFS stimuli are defined by luminance contrast, target color can be reliably identified but percent-correct discrimination of target orientation is near chance. When the colored target is moving, color identification deteriorates with motion speed but direction of motion discrimination improves with target speed. Color's immunity to suppression is also weakened when interocular suppression is induced by equiluminant CFS stimuli that presumably stimulate the chromatic pathway. These results imply that the strong interocular suppression induced by CFS may differentially operate on parvo- and magno-cellular mechanisms depending on the chromatic properties of the stimuli inducing CFS.

3:00 PM: Michael Engles, University of Georgia

Macular pigment and contrast sensitivity: testing the acuity hypothesis

The macular pigment (MP) is composed of diet-derived carotenoids that screen blue light and are concentrated in and around the fovea. Several functions for the MP have been proposed. The earliest hypothesis, termed the Acuity Hypothesis, predicts that increased MP optical density should improve spatial vision by reducing the deleterious effects of chromatic aberration, an optical phenomenon whereby short-wave light is blurred to a much greater extent than longwave light. Previous tests of this hypothesis have an insufficient number of subjects (typically only two), have neglected to measure MP optical density, or have only tested the highest spatial frequencies of the greater contrast sensitivity function (CSF). To adequately test the Acuity Hypothesis, a specially designed optical apparatus was constructed and a rigorous psychophysical procedure was used. Complete CSFs could be obtained in broadband achromatic light (unfiltered xenon; attenuated by the MP) and in shortwave-deficient light (bandpass filter with a 550nm cutoff; not attenuated by MP). The results of this study do not support the predictions of the Acuity Hypothesis, and our findings are in agreement with those of previous studies. This is not to say that the MP does not influence visual function in other contexts (e.g. the visibility of distant targets as viewed through Earth's atmosphere). To this end, measuring CSFs in the presence of simulated blue haze carefully matched to sky light is already underway.

Close Window

Festschrift in Honor of Dr. Walt Makous

Moderator: David Williams, University of Rochester

4:30 PM: Don MacLeod, University of California-San Diego

Color in the neural maze

I will review some of the presumed successes, and some acknowledged and unacknowledged obscurities, in our physiological understanding of the dimensionality of color space. As we trace the flow of information from object to image and thence through the retina and visual pathway, it becomes harder, not easier, to discern an isomorphism between color as we perceive it and the neural representation of color in the visual system. Even the physiological explanation of trichromacy presents as yet unanswered challenges in the common situation where more than 3 visual pigments are involved.

Color charts commonly adopt a coordinate system with two bipolar axes, one for redness (positive) vs. greenness (negative) and the other for yellowness vs blueness. Perceptually unitary ('unique') red and green, and blue and yellow, are found in opposite directions from white at the origin. But the pervasive rectifying nonlinearity of neural responses, and the marked asymmetry between excitatory and inhibitor response dictated by the relatively low spontaneous firing rate, are equally supportive of an idealization with multiple monopolar signals for redness, greenness, yellowness and blueness. With mutually orthogonal monopolar coordinates for the four primary signals, the isoluminant colors lie on the surface of a hypercube in the 4D space, with the achromatic point at one corner. To test this scheme, MacLeod, Pallett and Krizay asked whether red, green, yellow, and blue colors equidistant from white are perceptually equidistant from each other, as the hypercube model predicts. Results are close to the predictions of the hypercube surface model.

4:50 PM: Julie Schnapf, University of California-San Francisco

Electrical networks in mammalian photoreceptors

Photoreceptors are known to be coupled electrically through gap junctions, but the functional consequences of coupling are not well understood. We've studied photoreceptor coupling in guinea pig and macaque. Dye injections show that rods are coupled to each other in pools of 1- 12 rods. Statistical variations in light responses indicate that some rods are uncoupled, but most show signal spread to neighboring rods. Taken with our measured junctional conductance of ~350 pS, the electrical data are consistent with patchy networks of 1-10 rods. This is in agreement of our dye pool measurements as well as the rod networks derived from mouse electron micrographs. We modeled rod coupling and its effect on downstream retinal processing and absolute threshold. Coupling is expected to impact scotopic performance and optimal postreceptoral filtering.

In macaque retina, rods are electrically coupled to red and green cones. In addition, about half of blue cones also show rod input. The magnitude of rod input is comparable in all three cone types. Our analysis suggests that rod-cone coupling will influence color perception. Indeed psychophysical studies demonstrate that rod excitation causes shifts in color percepts. Using our direct measurements of rod and cone light responses, we modeled the extent to which rod-cone coupling could account for the rod influence on color. For most dim stimuli, rod excitation is predicted to shift color percepts towards blue-green.

5:10 PM: Bill Geisler, The University of Texas-Austin

Natural systems analysis

The environments in which an organism lives and the tasks it performs within those environments shape its perceptual systems through evolution and experience. This is an obvious statement, but it implies several fundamental components of research that are needed if we are going to gain a deep understanding of perceptual systems. The first is to identify the natural tasks and sub-tasks that are performed by the organism under natural conditions. The second is to measure and analyze those specific environmental properties (natural scene statistics) relevant for performing the tasks. The third is a computational analysis to determine how a rational (ideal) perceptual system would exploit the measured environmental properties to perform the tasks. This component is critical because it provides insight into the information contained in the natural stimuli and it can suggest principled hypotheses for the neural mechanisms the organism might use to exploit that information. The fourth component is to formulate specific hypotheses for neural mechanisms, based on the first three components, and test them in physiological and behavioral studies that capture the essence of the natural task. This general approach is illustrated with a study of contour grouping that combines measurements of natural scene statistics, derivation of ideal Bayesian observers that exploit those statistics, and psychophysical experiments that compare human and ideal performance. This study and other recent studies demonstrate the great potential of "natural systems analysis" for producing advances in behavioral science and systems neuroscience.

5:30 PM: Peter Bex, Harvard University

Contrast perception in natural scenes

Contrast sensitivity is routinely measured with sine wave grating stimuli presented on homogenous grey backgrounds, whereas natural images are composed of a broad range of spatial and temporal structure. I examine how contrast sensitivity varies with the context in which it is measured in an effort to extend our band-pass channel-based models of visual processing for application in natural conditions.

Close Window

Gene Therapy Approaches to Basic and Clinical Vision Sciences

Treatment of visions disorders has motivated development of gene therapy approaches to restore visual function. In addition the tools developed have proved to be useful for exploring basic properties of the visual system. This session will review some of the clinical and basic science advances made possible through the use of gene therapy in the retina.

Moderator: Jay Neitz, Medical College of Wisconsin

8:30 AM: Matt Mauck, Medical College of Wisconsin

Using gene therapy to probe the circuit for color vision

We have successfully treated color blindness in an adult primate using gene therapy. This raises the major question of what are the properties of the neural circuit that make the addition of a new dimension of color vision possible. Toward answering this, we probed the circuit for color vision in the rodent using a gene therapy approach in which new long-wavelength sensitivity was targeted to either S-cone or M-cone pathways. Gene therapy using cone-class specific transcriptional regulatory elements enabled us to express human L-opsin in a mosaic of either M- or S-cones. Functional consequences of the expanded spectral sensitivity were explored after therapy by measuring the neuronal response throughout the visual system with fMRI and behavioral tests of color vision. Dramatic expansion of color vision to include a red-green dimension of color vision in these rodents was observed when L-opsin was targeted to S-cones but not when targeted to M-cones. This result indicates that the novel red-green color vision observed in animals treated with gene therapy is served by a portion of the preexisting circuit for blue-yellow color vision involving the S-cone pathway. This implies that the expanded color vision was generated with a change only at the receptor level with gene therapy, without plastic neural changes in high-order cortical or subcortical circuitry. This presumably parallels the origins of the circuit for red-green color vision in primates which must have arisen from a preexisting circuit serving blue-yellow color vision rather than from evolution of a new circuit de novo.

8:50 AM: Andras Komaromy, University of Pennsylvania

Restoration of cone function in dog models of rod monochromacy

In an estimated 75% of patients with achromatopsia or rod monochromacy the disease phenotype is caused by mutations in the alpha (CNGA3) or beta (CNGB3) subunits of the cone cyclic nucleotide-gated (CNG) channels. Our work focuses on studying the molecular disease mechanisms and on developing new therapies in two canine models with CNGB3 channelopathies. By using these models, we have provided proof of principle that gene expression can be targeted specifically and robustly to cone subclasses and their function restored by recombinant adeno-associated virus (rAAV)-mediated gene replacement therapy. rAAV of serotype 5 containing human CNGB3 cDNA was injected into the subretinal space of dogs affected by either a null or missense mutation (D262N) of CNGB3. The transgene was under control of cone-specific promoters, either a variation of the human red cone opsin or the human cone arrestin promoter. Restoration of cone function was evaluated by behavioral visual testing under photopic conditions and by recording of full-field flash electroretinograms. The robustness of the rescue was promoter-dependent but mutation-independent. Our results are promising for future gene therapy in human patients with achromatopsia, or other diseases that affect cones.

9:10 AM: Ken Greenberg, University of California-Berkeley

Electrophysiology of channel rhodopsin in rabbit retina

One common feature of diseases causing photoreceptor loss is that inner retinal neurons are preserved long after photoreceptor apoptosis occurs. Some degree of vision may potentially be restored if these remaining neurons could directly respond to light and transmit meaningful information to visual centers. Recently, light-sensitive cation channels (Channelrhodopsins) and anion pumps (Halorhodopsins) were shown to effectively excite and inhibit neural activity in response to visible light. By targeting these rhodopsins appropriately in a diseased retina, a virtual ON or OFF signaling pathway may potentially be generated. The human retina contains at least 30 morphologically distinct bipolar and ganglion cell subtypes, however ON and OFF-center signal detection is the most significant division among visual features extracted by ganglion cells. Imparting light sensitivity specifically to ON or OFF-center bipolar and ganglion cells with excitatory or inhibitory rhodopsins may yield new insights into signal processing and could allow light perception in the absence of rod and cone-mediated vision. Using patterned light stimuli with electrophysiological recording techniques, we are attempting to understand the quality of vision that could be restored using this prosthetic approach.

9:30 AM: Kate Kolstad, University of California-Berkeley

Engineered photo-switch driven cortical responses in models of inherited blinding diseases

An alternative therapy for blinding diseases, such as retinitis pigmentosa and macular degeneration, is to confer light sensitivity on second and/or third order neurons in the retina. One benefit of this approach is it provides a therapy for late stages of retinal degeneration, where the majority of photoreceptors are already lost. We have delivered engineered light sensitive proteins to retinal ganglion cells in an animal model of retinal degeneration. These light sensitive proteins are capable of finely tuned control of single neuron activity in cultured hippocampal cells and retinal explants (Szobota et al. 2007, Borges and Greenberg, unpublished data) However, an important question is―How might engineered visual input be interpreted by the brain? To answer this question we have recorded cell population responses in V1 when visual input is limited to engineered channel induced activity in the retina. Our data show that these "photo-switches" in the retina are capable of restoring cortical responses to full field light flashes in previously "blind" animals.

9:50 AM: Discussion

Close Window

Contributed Talk Session: Vision

Moderator: Greg Appelbaum, Duke University

11:00 AM: Michael Falconbridge (to be given by Don MacLeod), University of California San Diego

Invisible gratings exposed by their effect on subsequent test gratings

Early stages of the visual system respond to very rapid changes in visual input. By the time signals reach perception, this high temporal resolution is lost. We presented a rapidly counterphasing, masked grating that was invisible to observers. When asked to discern the orientation of the grating in a 2 alternative forced-choice procedure, observers performed at chance levels. These invisible gratings nevertheless caused a specific reduction in sensitivity to subsequently presented test gratings that were oriented parallel to the invisible grating. This adaptation effect allowed observers in a separate experiment to ascertain the orientation of the invisible grating on the basis of its effects on the appearance of the subsequent test gratings. Further to this, when a rapidly modulating adapting grating is presented to one eye, and adaptation is tested in the other eye, the level of adaptation is only about 20% smaller than if both stimuli were presented to both eyes. Taken together, the results imply that 1) the site of adaptation is no earlier than input layers of V1 (where orientation is first encoded) with a significant portion of adaptation being mediated at or beyond the stage where input from both eyes is integrated, 2) sensitivity to rapid change penetrates to this level, and 3) the neural mechanisms driving conscious perception either ignore, or do not have access to the information about rapid modulations that drive the adaptation effect.

11:15 AM: Peng Zhang, University of Minnesota

Long-term orientation-specific contrast reduction reveals plasticity of mechanisms of contrast appearance

Full contrast stimuli that differ in spatial pattern usually appear equally intense, despite sensitivity differences at threshold. This contrast constancy may arise from observers' experience with the environment. To test this, we placed observers in an environment where we gradually reduced the contrast of image components at a specific orientation. Six subjects viewed the world using an "altered reality" system, comprised of a head-mounted video camera that fed into a laptop computer that in turn drove a head-mounted display. Energy at a narrow range of orientations across all spatial frequencies was removed from the video images; this filtering was done in real-time on the laptop computer. Over the course of four hours, subjects performed everyday activities while the strength of the filtering increased linearly from a 0% to a 38% reduction in contrast. At four equally spaced intervals, subjects performed a contrast matching task on a calibrated LCD display viewed through the altered reality system. Subjects adjusted the contrast of an orthogonally oriented Gabor pattern to match a maximum contrast test pattern at the filtered orientation. The test pattern contrast reduced from 100% to 62% across test sessions, as filtering strength increased. Average matched contrast to the test pattern stayed relatively constant, decreasing, from 92% to 80% over the four sessions. The difference between the test and match contrast increased reliably across sessions. These data reveal plasticity in mechanisms of contrast appearance driven by the distributions of contrast in the observers' environment.

11:30 AM: Betina Ip, University of Oxford

Spatial, but not feature attention, modulates responses to stereoscopic structure-from-motion in human visual cortex

Perceiving solid objects involves extracting information about their 3D structure, often using a combination of motion and binocular cues. Little is known about how visual attention affects these processes. fMRI-studies suggest that objects can be selected as a unit of spatial attention [1] and that attending to features within objects can change responses across the visual field [2]. Do spatial and feature attention modulate cortical responses to stereoscopic structure-frommotion (SFM) stimuli?

To direct spatial attention, a cue appeared before each trial, directing attention left or right. Two rotating SFM-cylinders appeared, one left and one right of the central fixation point. Subjects performed a behavioural task on the cued cylinder. The cortical response to spatial attention should give differences in fMRI-responses between the attended vs unattended cylinder in visual areas (V1-V7, LO1, LO2, hMT+), while feature attention should be revealed by different response to similar or dissimilar directions of rotations of the two cylinders. We found that spatial attention strongly increased responses in both dorsal and ventral visual areas in a retinotopic manner. Higher areas in the intra-parietal-sulcus and the frontal-eye-fields were also modulated. However, neither behavioural nor fMRI-responses reflected the featural similarity between the two cylinders. One mechanism for this could be the reallocation of spatial attention.

The cortical responses to SFM-cylinder are modulated by spatial attention, but not by feature attention based on similarity. Preliminary data suggest that behavioural relevance of the feature of interest may be required to modulate the responses at the level of component features.

11:45 AM: Ying Geng, University of Rochester

In vivo imaging of rat retinal capillaries and fluorescently labeled retinal ganglion cells, dendrites and axons

Rodent transgenic and knockout models enhance our understanding of normal and diseased retina. However, in vivo microscopic resolution of retinal cells is compromised by ocular aberrations. We have characterized the performance of the Rochester fluorescence adaptive optics scanning laser ophthalmoscope (fAOSLO) for imaging cells in the living rat retina.

GFP was expressed in ganglion cells of normal SD rats via intravitreal injections of AAV vectors. An fAOSLO acquired simultaneous reflectance and fluorescence retinal images. For comparison, histological images were obtained in the same eyes using a confocal SLO. In vivo resolution with the fAOSLO was characterized by comparing the transverse cross-section across individual dendrites from in vivo images to cross-sections from histology.

Capillaries and fluorescently-labelled ganglion cell bodies, axons, and dendrites were clearly resolved after AO correction. However, the FWHMs of dendrite cross-sections taken in vivo were ~40% larger than the expected FWHMs from histology.

If the aberrations of the rat eye were completely corrected, the resolution could be ~2X that of the human eye. While our instrument corrects a substantial fraction of the aberrations as indicated by the RMS wavefront error after AO correction, direct measurements of retinal image quality reveal some blur beyond that expected from diffraction. Nonetheless, subcellular features of ganglion cells can be resolved, which offers promise for using adaptive optics to investigate the rodent eye in vivo.

12:00 PM: Girish Kumar, University of Houston

Localized distortions within saccadic mapping contribute to the variance in saccadic landing points

Previous work 1 showed that saccades placed stimuli within a zone of approximately 5-10 arc minutes in diameter within the fovea. Our aim is to determine the contribution of localized targeting distortions towards this variability.

An Adaptive Optics Scanning Laser Ophthalmoscope was used to collect videos from 3 normal subjects who fixated a 7 arc minute square that jittered randomly among nine positions within a 2 degree field. Eye movements were extracted and saccades identified using a velocity criterion of 15 o/sec. The retinal locations of the stimulus in the frame prior to a saccade was taken as its "starting location", while the location following the saccade was the "ending location". Saccades were grouped based on the proximity of their starting locations, and the spread of the ending locations between these groups and randomly chosen saccades were compared.

Spread for all ending locations ranged from 10 to 17 arc minutes. For two subjects, a sign test showed a statistically significant proportion (r2 – 0.16 to 0.18).

Our findings show saccades place stimuli within a "preferred retinal zone" that tends to undershoot a central foveal location. However, local retinal regions have unique and separate targets within this zone, leading us to infer the presence of localized distortions within saccadic mapping.

12:15 PM:Ethan Rossi, University of California, Berkeley

The relationship between the cone photoreceptor mosaic and visual acuity in normal observers and blue cone monochromat carriers

Determine the relationship between the photoreceptor mosaic and visual acuity (VA) in normal observers and blue cone monochromat (BCM) carriers.

The adaptive optics scanning laser ophthalmoscope (AOSLO) was used to project an AO-corrected stimulus onto the retina of 9 observers (6 normal; 3 BCM carriers). High contrast photopic letter acuity was measured using a 4AFC tumbling E test at the preferred retinal locus (PRL) and temporal parafovea. Stimuli were presented at 840 or 658 nm with simultaneous photoreceptor imaging at 840 nm.

Cones were well resolved at the PRL for all BCM carriers and two normal observers. BCM carriers had lower cone density and greater cone spacing than normal observers. Voronoi analysis revealed irregular cone packing and cone loss in BCM carriers. Power spectra of mosaic images confirmed irregular packing, with no Yellot's ring at test locations of BCM carriers. BCM carriers performed worse in the VA task than normal observers. VA was better than or matched the Nyquist limit for all normal observers and one BCM carrier at the PRL, but worse than the Nyquist limit beyond the PRL.

VA is not sampling limited (ie. limited by the spacing of rows of photoreceptors) beyond the PRL. Cone spacing closely matches VA outside the PRL, but there is a trend toward underperformance (VA drops faster than spacing). Although these BCM carriers had normal VA tested conventionally, their performance with AO-correction shows that AO-corrected tests may reveal cone loss before such loss is detectable with conventional methods. A large discrepancy between cone spacing and VA in two BCM carriers may indicate a post-receptoral deficit.

Close Window

Color and Motion Processing

The question is the extent to which color supports or does not support motion processing, and so has direct implications for the functional separability of the dorsal and ventral streams. The idea can be approached from a number of perspectives.

Moderator: Karen Dobkins, University of California-San Diego

2:30 PM: Brian White, Queen's University

Color signals in the primate superior colliculus

Color is important for segmenting objects from backgrounds, which can in turn facilitate visual search in complex scenes. However, neural substrates that control overt visual orienting towards interesting objects (i.e., saccadic eye movements) are not believed to encode visual features such as color (Bichot, Schall, & Thompson, 1996; McPeek & Keller, 2002), despite direct projections from color-related areas to saccadic substrates such as the Superior Colliculus (SC)(Lock, Baizer, & Bender, 2003). Using a delayed saccade task, we measured the activity of visually responsive SC neurons (N=68) in two monkeys to stimuli derived from the DKL color space (Derrington, Krauskopf, & Lennie, 1984). We show the first evidence that neurons in the intermediate SC layers can respond to pure chromatic stimuli with the same magnitude as the best luminance stimulus (100% contrast). This activity is not adequately explained by residual luminance signals because 1) neurons with an equal or better color response showed significantly longer visual onset latencies (25-30ms), and 2) the luminance control-stimulus was more than 20cd/m2 different from the color stimuli. Furthermore, the color response was not seen for phasic-visual neurons obtained from the superficial SC layers. This implies that the color-related activity in the SC involves different pathways than luminance, and projects primarily to the intermediate SC layers. The delayed color response is also reflected by the longer saccadic reaction times previously reported for chromatic DKL stimuli (White, Kerzel, & Gegenfurtner, 2006). We conclude that this activity represents a true colour response only one stage from the brainstem premotor circuitry that drives the eyes.

2:50 PM: Jonathan Nassi, Harvard Medical School

Specialized circuits relaying primate parallel visual pathways to area MT

Parallel pathways in the primate visual system parse the retinal input into relatively independent magnocellular (M), parvocellular (P), and koniocellular (K) streams. Chromatic opponency carried along the P and K pathways remains anatomically separate and distinct from the non-opponent signals carried along the M pathway through the lateral geniculate nucleus (LGN) and into the input layers of primary visual cortex (V1). Within V1, however, these pathways appear to intermix and it is unclear how each pathway contributes to the computations performed in extrastriate visual cortex. If color supports motion perception, then it is likely that color and motion cues integrate somewhere along the cortical visual hierarchy. Dorsal stream area MT is known to be specialized for motion processing and is a likely location where color signals could meaningfully contribute to the perception of motion. Nevertheless, MT is thought to be dominated both functionally and anatomically by inputs from the M pathway, with little or no contributions from the P or K pathways. We used rabies virus as a transynaptic tracer of anatomical circuits to assess the contributions of M, P, and K pathways to area MT of macaque monkey. We were particularly interested to determine whether MT receives any inputs from the chromatically opponent P or K pathways. We found that the main ascending input through layer 4C of V1 to MT is indeed dominated by the M pathway. Nevertheless, the P pathway does reach MT along alternative routes, including a surprisingly robust, disynaptic input from the P layers of the LGN. Additional input from the P pathway may arrive into MT through a less direct connection involving layer 4C of V1 and extrastriate cortical area V2. While the functional implications of P and K inputs to MT remain unclear, each of the specialized circuits described above likely provides MT with a unique combination of M, P, and K pathway signals and informs highly specific visual computations and tasks.

3:10 PM: Shin'ya Nishida, NTT Communication Science Laboratories

Trajectory integration of color signals for motion deblurring

Whether fundamental visual attributes, such as color, motion and shape, are analyzed separately in specialized pathways has been one of the central questions of visual neuroscience. Although recent studies have revealed various forms of cross-attribute interactions, including significant contributions of color signals to motion processing, it is still widely believed that color perception is relatively independent of motion processing. A challenge to this belief is 'motion-induced color mixing' (Nishida et al., 2007, Current Biology), in which moving bars, the color of which alternates between two colors (e.g., red and green), are perceived as the mixed color (e.g., yellow) even though the two colors are never superimposed on the retina. This phenomenon suggests that color signals may be integrated not only at the same retinal location, but also along a motion trajectory, as in the case of trajectory integration of pattern signals (Nishida, 2004, Current Biology). It is possible that this neural mechanism helps us to see veridical colors for moving objects by reducing motion blur. In line with this hypothesis, we also found 'motion-induced color segregation' (Watanabe & Nishida, 2007, Journal of Vision), in which temporal alternations of two colors on the retina are perceptually segregated more veridically when they are presented as moving patterns rather than as stationary alternations at the same rate. Additionally, this improvement of objective temporal resolution is similar to subjective motion deblurring (Bedell & Rott, 1996, Current Biology) in that both effects are enhanced when an observer views a stationary object while making a pursuit eye movement, in comparison with when an observer views a moving object without moving eyes (Terao et al., 2008, VSS).

3:30 PM: Declan McKeefry, University of Bradford

Color in motion revealed by motion after-effects

The analysis of the colour and motion of visual stimuli was traditionally believed to take place within segregated processing pathways in the primate visual system. However, it has become increasingly more apparent that this segregation cannot remain absolute and that there must be some capacity for integration across these sub-modalities. In this talk I will describe some psychophysical experiments which explore the extent to which colour and luminance cues can be either combined or segregated in motion perception.

In studies of the traditional motion after-effect (MAE), where prolonged inspection of a unidirectional moving stimulus results in illusory motion in the opposite direction, we have found evidence of chromatic selectivity. Induced MAEs were strongest when the adapting and test stimuli were of the same chromatic (or luminance) contrast composition, but fell to a minimum when the two stimuli were modulated along orthogonal axes in colour space. However, when we examined the chromatic selectivity of a second after-effect, in which motion adaptation induces misperceptions in the spatial position of stationary objects, we found it to be completely insensitive to chromatic composition of the test and adapting stimuli. Similarly, in experiments where we measured misperceptions in the speed of moving grating stimuli, induced by placing flanking or 'modifier' stimuli in close proximity, we found that the magnitude of the effects were the same regardless of the chromatic or luminance contrast composition of the stimuli.

The results from these experiments suggest that there are separable chromatic and luminance inputs to motion processing, at least in its earliest stages. However, when motion cues are utilised for more complex aspects of perception, such as position estimation or the computation of stimulus speed, there is the capacity to effectively combine both colour and luminance information to achieve specific perceptual goals.

3:50 PM: Discussion

Close Window

Long-term Adaptive Effects in Color Vision

There are emerging reports that human color appearance can show long term changes, for example in compensating for the variations in macular pigment density across the visual field or for age related color changes in lens pigment density. These would reflect novel, long-term adaptation mechanisms that govern color appearance. The speakers will present their research into the existence and nature of these mechanisms.

Moderator: Angela Brown, Ohio State University

8:30 AM: Billy Hammond, University of Georgia

Compensation for macular pigment: color appearance and sensitivity regulation

A long-standing question in color vision research is how the visual system is able to compensate for the significant absorbance of short wave light by the crystalline lens and macular pigment (MP). The significant attenuation of short-wave light due to prior filtering must require subsequent enhancement of the blue signal in order to maintain color constancy across the retina where MP levels are changing quickly and dramatically. We studied this compensation mechanism by measuring MP spatial density profiles, S-cone sensitivity, and hue cancellation functions across the central retina. Despite large variations in MP across the retina, hue cancellation values for the Y-B system across the central retina were constant. For example, one subject's MP density declined from a central peak of 0.99 to near zero at 7° (near 95% transmission difference) yet thresholds for the Y-B system were unaffected. In contrast, the G-lobe of the R-G system was directly correlated with MP density. Absolute sensitivity levels between foveal and parafoveal locations were also very similar for the peak of the s-cone pathway (440 nm). The results for sensitivity and color appearance are consistent. The intense yellow-background used in increment thresholds, for instance, probably adapts the yellow lobe (which receives its input from mid-and-long wave cones) of the Y / B system so strongly that it is the B-lobe of the system exclusively mediating the increment threshold response (i.e., isolating just the S-cone input to the system). Taken together, the results suggest that the visual system compensates for differential filtering of SW-light by MP across the retina by regulating the Y / B opponent system.

8:50 AM: Jack Werner, University of California-Davis

What the aging lens can tell us about color appearance

Substantial age-related change in the spectral distribution of the light reaching the retina is known to occur as a result of lenticular senescence. These changes are accompanied by neural losses in cone pathways. It is, therefore, often expected (and even stated) that color perception will be altered in the elderly, especially when tested with stimuli that remove the normal cues for color constancy. This expectation is wrong, as we have demonstrated by measuring age-related changes in color appearance using several different approaches and exploring a wide gamut of color space. Here we summarize experiments that support the view that this stability of color perception across the life span is due to compensation by the visual system for lenticular senescence and neural losses in order to renormalize chromatic mechanisms.

9:10 AM: Rhea Eskew, Northeastern University

Potential mechanisms of long-term adaptation in color vision, and a failure to find evidence for them

In an interesting and provocative paper, Neitz and colleagues (2002) provided evidence that wearing colored contact lenses over an extended period could produce long-term changes in color vision, as indexed by shifts in unique yellow settings. They interpreted the shifts as being due to the long-term chromatic exposure altering the relative gains (cone weights) within the red-green color mechanism, but there are other plausible interpretations. One possibility that we considered is that the lenses cause a relative change in the internal noise associated with the L and M-cone inputs to the hue mechanism, which could produce larger shifts in hue at low retinal illuminances than at high, an idea that might be consistent with the very recent claim of Belmore & Shevell (2008) of intensity-dependent hue shifts after long-term adaptation This talk will describe this and other possible reasons for the shifts in unique yellow, and describe a study that was designed to test between some of these alternatives. Unfortunately, however, we were unable to consistently produce shifts in unique yellow by long-term wearing of red lenses. Although initial results looked promising (encouraging data from two observers were presented at FVM in 2006), additional observers and additional data on the first two observers ultimately failed to replicate the basic finding of a hue shift. In some cases, our observers (like those of Belmore & Shevell) failed to show a convincing return to baseline, suggesting perhaps that whatever small changes we observed were either permanent, which seems unlikely, or spurious.

9:30 AM: Aline Bompas, Cardiff University

Eye movements participate in color appearance

The retinal image is far from homogeneous. In particular, the density variation of macular pigment with eccentricity introduces major centre-periphery differences in the spectrum and intensity of light sampled by our photoreceptors. Although this leads to substantial perceptual differences when measured in laboratory conditions, these are not detected in real life. Most interestingly, objects seen in peripheral vision do not suddenly change color when we look straight at them, even if one pays particular attention to possible color changes. I have investigated the hypothesis that saccadic eye movements participate in calibrating color experience across the visual field: each time we make a saccade towards an object in the periphery, the particular change in sensory input that is produced would be learnt and consequently compensated. Such process would participate in perceptual stability across eye movements, ensuring that only relevant color changes are actually perceived. This hypothesis is made plausible by previous findings showing that correlations between eye movements and color changes can be learnt and compensated for by our visual system (Bompas & O'Regan 2006, Perception 35(1), 65-78; Bompas & O'Regan 2006, Journal of Vision 6(2), 145-153).

9:50 AM: Discussion

Close Window

Between the Eyes and the Cortex: Active and Passive Filtering in the Geniculate

The classic view of the lateral geniculate nucleus (LGN) as a passive relay station has recently been challenged by findings demonstrating its active role in visual information processing. The speakers will describe research performed in animals or humans to study the effects of attention and alertness on LGN responses. Some of their results differ markedly from previous studies of the role of LGN, and the speakers will be invited to speculate on the possible reasons for this difference.

Moderator: Peter Lennie, University of Rochester

11:00 AM: Sabine Kastner, Princeton University

Neural correlates of visual attention and awareness in the human LGN

The LGN is the thalamic station in the projection of the visual pathway from retina to visual cortex and has been traditionally viewed as a gateway for sensory information. Its topographic organization and neuronal response properties have been extensively studied in nonhuman primates, but are poorly understood in humans. I will review a series of studies aimed at elucidating functional roles of the human LGN in perception and cognition using fMRI. Together, these studies indicate the need to revise the traditional view and to consider the LGN as an early gatekeeper in controlling attention and conscious perception.

Functional LGN topography [1] was studied by presenting periodic flickering checkerboard stimuli that evoked a traveling wave of activity. We found that the contralateral visual hemifield was represented with the lower field in the medial-superior portion and the upper field in the lateral inferior portion of each LGN. The fovea was represented in posterior and superior portions, with increasing eccentricities represented more anteriorly. This topography is strikingly similar to that of the macaque.

Selective attention has been shown to modulate neural activity in both extrastriate and striate cortex. We studied the poorly understood role of earlier, subcortical structures in attentional processing and found that attention modulated neural responses in the human LGN in several ways: it enhanced neural responses to attended stimuli, attenuated responses to ignored stimuli and increased baseline activity in the absence of visual stimulation, suggesting a role as a gatekeeper in controlling attentional response gain.

Finally, we studied the role of the human LGN in conscious perception by investigating the level at which competing inputs to the eyes, as perceived in binocular rivalry, can be resolved [3]. Neural activity in the LGN correlated strongly with the subjects' reported percepts, suggesting a mechanism by which LGN layers that process the input from one particular eye are selectively enhanced or suppressed. These finding suggest a role for the LGN as an early gatekeeper of visual awareness.

11:20 AM: Geraint Rees, University College London

Functional MRI of the human LGN and subcortical pathways

Subcortical nuclei, particularly those in the thalamus, may play an important role in human brain function. They provide challenging targets for functional neuroimaging because of their small size and deep locations. In this talk I will discuss recent studies from our group investigating subcortical nuclei in the retinotectal and geniculostriate visual pathways. Focusing particularly on the lateral geniculate nucleus (LGN) of the thalamus, I will present data suggesting that signals in the LGN are correlated with conscious visual perception of brightness and during binocular rivalry. Such observations present challenges to conventional views of thalamic function but also raise important new questions for further investigation. Preliminary data suggest that novel approaches using high spatial resolution fMRI show great promise in permitting decoding of local spatial patterns of thalamic activity to reveal the selectivities of neuronal populations in the human LGN.

11:40 PM: Henry Alitto, University of California-Berkeley

Spatial attention and visual processing in the lateral geniculate nucleus

Several studies have shown that spatial attention increases the responsiveness of neurons throughout striate and extrastriate cortical areas. Given the robust feedback pathway from striate cortex to the lateral geniculate nucleus (LGN) of the thalamus, it is possible that the cortical effects of attention may influence visual processing in the LGN. Because corticogeniculate axons provide direct excitatory input as well as polysynaptic inhibitory input onto LGN neurons, the effects of attention could be complex. Indeed, fMRI studies report increased blood flow to the LGN with attention, while 2-deoxy-glucose studies report decreased metabolic activity in the magnocellular layers of the LGN with attention.

To study the effects of spatial attention on the spiking activity of neurons in the macaque LGN, we trained two animals to perform a spatial attention task. In this task, animals fixated on a central location while two sine-wave grating patches were presented peripherally. One grating was presented over the receptive field of a recorded LGN neuron, while the second grating was presented at a different location. Animals were instructed to attend to one or the other grating in block format based on the color of the fixation spot. At a random delay after the onset of fixation, the contrast of the attended grating either increased or decreased. Animals were rewarded for correctly reporting the nature of the contrast change. Catch trials demonstrate that animals were under proper behavioral control. Our results show that attention decreases the activity of magnocellular LGN neurons when distracter stimuli are located in the same visual hemifield as the receptive fields of recorded neurons, but not when the distracters are in the opposite visual hemifields. Given this result, we suggest an interaction between spatial attention and the extraclassical suppressive field of LGN neurons.

12:00 PM: Jose-Manuel Alonso, SUNY College of Optometry

Receptive field dynamics and response gain in visual thalamus

The Lateral Geniculate Nucleus (LGN) of the thalamus is the main entrance of visual information to the cerebral cortex and its functional role has been frequently compared with a simple gate that regulates information transfer. This concept of a passive gate, which implies that there is no major receptive field transformation from retina to LGN, is supported by two well-accepted findings. First, the LGN receptive fields are very similar to the receptive fields from their retinal inputs. And second, LGN visual responses are strongly dominated by the input from a single retinal afferent. In the first part of the talk, I will challenge the concept of a passive gate by showing that, while the fundamental structure of individual receptive fields (center-surround) is the same in LGN and retina, the tiling of the receptive field arrays is very different. I will argue that LGN receptive field arrays are more spatiotemporally diverse and that this enhanced diversity results from the divergence and convergence patterns of retinogeniculate connections (Alonso et al., 2006).

The notion that the LGN functions as a passive gate has been challenged in the past by the finding that most excitatory synapses in LGN originate in the brain stem and visual cortex but not the retina. In the second part of the talk, I will show that changes in the level of alertness in awake animals (likely mediated by the brain stem), can double the amplitude of visual responses and dramatically increase the bandwidth of temporal frequency tuning in LGN (Cano et al., 2006; Bezdudnaya et al., 2006). Remarkably, these pronounced changes in response amplitude and temporal tuning can be accomplished without altering the contrast sensitivity and receptive field size of LGN cells. Therefore, these results demonstrate that the abundant excitatory synapses from the brain stem can have a powerful effect in LGN function without necessarily altering the basic receptive field structure of LGN neurons.

12:20 PM: Discussion

Close Window

Measuring Population Activity in Visual Cortex

This symposium focuses on recent application of neuroimaging techniques aimed at decoding population responses in the visual system. The aim is to present a broad scope of research concerning the recording and interpreting data from neural populations. A specific aim of this symposium is to tie together general analysis techniques from different modalities, for example, the use of support vector machines.

Moderator: Greg Appelbaum, Duke University

2:15 PM: Adam Kohn, Albert Einstein College of Medicine

Neural correlation in V1 and its effect on coding

Cortical responses to repeated presentations of a stimulus are variable and this variability is correlated between cells. Theoretical work has shown that correlated variability can strongly affect the ability of neuronal populations to encode sensory information, although the impact depends critically on both the structure of correlation and the way in which responses are decoded. Ultimately, however, the impact of correlated variability in a local network rests in how it affects downstream networks. To evaluate how correlated variability propagates through the visual system, we have recorded simultaneously from a population of neurons in macaque primary visual cortex (V1, using implanted arrays of 100 microelectrodes) and their downstream targets in the input layers of area V2. We measured responses to repeated presentations of drifting gratings and evaluated how well we could predict trial-to-trial fluctuations in V2 responsiveness by monitoring population activity in V1. We fit a generalized linear model (GLM) to a subset of the trials, and tested its ability to predict responses on novel data. We found we could predict a substantial portion of trial-to-trial fluctuations in V2 by monitoring V1 responses, with a performance level that was similar to our ability to predict a V1 neuron's response by monitoring its nearby neighbors; that is, it was as if V1 and V2 neurons were embedded in a single network with shared noise. The V1 neurons weighted most heavily in the GLM were those with spatial receptive fields most similar to that of the target V2 cell; relative orientation preference of the V1 and V2 cells played relatively little role. Our results suggest that a substantial portion of V2 variability can be explained by fluctuations in the response of V1. Cortical variability in V1 thus has a substantial impact on downstream networks.

2:35 PM: David Fitzpatrick, Duke University

Distortions in perceived direction of motion predicted by population response dynamics in primary visual cortex

Encoding the trajectory of a moving stimulus that abruptly changes its direction of motion provides a particularly vivid example of the challenges inherent in using cortical circuits to represent rapidly changing stimulus features that are ubiquitous in visual scenes. Instantaneous changes in the properties of a visual stimulus are accompanied by changes in the activity of cortical circuits that may outlast the stimulus event by 100's of milliseconds, but how these complex circuit dynamics impact population coding mechanisms remains unclear. Here we employed in vivo voltage sensitive dye (VSD) imaging to explore how abrupt changes in the trajectory of a moving stimulus impact the population coding of motion direction in ferret primary visual cortex (V1). For motion in a constant direction, the peak of the cortical population response reliably signaled the stimulus trajectory; but for abrupt changes in motion direction, the peak of the population response departed significantly from the stimulus trajectory in a fashion that depended on the size of the direction deviation. For small direction deviation angles, the peak of the active population shifted from values consistent with the initial direction of motion to those consistent with the final direction of motion by progressing smoothly through intermediate directions not present in the stimulus. In contrast, for large direction deviation angles, peak values consistent with the initial motion direction were followed by: a small deviation away from the final motion direction, a rapid 180 degree jump, and a gradual shift to the final direction. While the basic qualities of the population response to motion transitions are consistent with linear summation of the response to each component, direct tests of linearity reveal the presence of additional nonlinear mechanisms that shape the dynamics of population response both during and after motion transitions. These departures of the population response from the actual trajectory of the stimulus predict specific misperceptions of motion direction that were confirmed by human psychophysical experiments. Small angular deviations in direction of motion are perceived as smoother than the actual stimulus change, while large angular deviations are perceived as sharper than the actual stimulus change. We conclude that cortical dynamics and population coding mechanisms combine to place constraints on the accuracy with which abrupt changes in direction of motion can be represented by cortical circuits. The smoothing of small direction deviations and the sharpening of larger deviations could serve to enhance the discrimination of continuous and discontinuous motion trajectories. But even if these perceptual distortions are not beneficial, their existence must represent an acceptable tradeoff that balances the need for accuracy with the innumerable advantages that are afforded by distributed coding mechanisms.

2:55 PM: Justin Gardner, New York University

Inferring population responses in human visual cortex with classification analysis

How the joint activity of many neurons with different selectivities represent visual stimuli is of key importance to understanding visual perception and visually guided behavior, yet most tools for probing these response properties in human visual cortex have been severely limited due to their coarse spatial-resolution. Conventional sized functional magnetic resonance imaging (fMRI) voxels (3x3x3 mm) are thought to encompass many dozens of cortical columns containing neurons with different specificities for basic visual stimulus properties like orientation and direction. However classification techniques have successfully shown that by combining responses across many voxels with small, noisy biases for orientation and direction, the stimulus a subject is viewing can be correctly classified. We used high-spatial resolution fMRI and classification analysis, and found that robust orientation and direction selectivity can be observed in large draining veins that would be expected to drain over an area of cortex comprising many cortical columns. Thus the success of classification techniques in fMRI may be based on sampling from easily measurable signals from large draining veins rather than biased sampling of well localized signals from cortical maps. Large veins may have specificity because they amplify a biased representation inherent to the cortical architecture or alternatively, because they might be organized to drain specifically from cortical columns that are functionally active together. Either way, our results suggest that inferring population response in human visual cortex can be made possible by measuring valid, if indirect, signals represented in large draining veins.

3:15 PM: Serge Dumoulin, Stanford University and Utrecht University

Quantitative population receptive field estimates in human visual cortex

We introduce a functional magnetic resonance imaging (fMRI) method for estimating the neuronal population receptive field (pRF) in human visual cortex. This pRF method builds on conventional visual field mapping techniques but extracts additional information beyond the maps and generalizes both the traveling wave method and alternative mapping techniques. The pRF method computes a model of the pRF from responses to a wide range of stimuli. It produces estimates of the visual field map as well as other neuronal population properties, such as the pRF size ipsi and contralateral extent (laterality). The pRF method decouples the visual field map measurements from the conventional ring and wedge stimuli, thereby eliminating some of the difficulties with the conventional techniques. We describe four results with this method. First, we show that visual field maps obtained with the pRF method are more accurate than those obtained using conventional visual field mapping. Second, we delineate the visual field maps to the center of the foveal representation (foveal confluence). Third, we describe two visual field maps within human motion-sensitive cortex (MT+), located on the lateral surface at the temporal-occipital (TO) boundary. We adopt the neutral terms TO-1 and TO-2 for these maps because of primate homology uncertainties but they likely correspond to macaque areas MT and MST. TO-1/2 abut the LO-1/2 visual field maps and are part of a continuous sequence of visual field maps extending from medial to lateral occipital cortex. Fourth, we report quantitative estimates of pRF size in medial, lateral and ventral occipital regions of human visual cortex. Human pRF sizes vary systematically across visual field maps with pRF sizes about 5x larger in TO-1/2 than V1. Within each visual field map, the pRF sizes increase with increasing eccentricity. The human pRF size estimates in V1/2/3 agree well with monkey and human electrophysiological receptive field measurements in corresponding locations. The pRF method is non-invasive and can be applied to a wide range of conditions when it is useful to link fMRI signals in the visual pathways to neuronal population receptive fields.

3:35 PM: Discussion

Close Window

Thursday (10/23/08) – Held at the OSA Annual Meeting, Frontiers in Optics (FiO)

Optical Models of the Eye

With the advent of accurate measurements of the optical quality of the eye, new models are now being developed of normal and abnormal eye growth. Models are important to understanding the normal development of the eye and optical changes in the development of refractive error. In addition, optical models are important to the design of optical corrections to the eye including intraocular lens implants.

Larry Thibos, Indiana University

Optical Models of the Eye

Pablo Artal, Campus de Espinardo, Univ. de Murcia, Spain

The Eye as an Aplantic Design

Ronald Kröger, Lund Univ., Sweden

Gradient Refractive Index Eye Models

Patricia Piers, AMO, Netherlands

Designing IOLs with Eye Models

Close Window

Virtual Displays and Natural Tasks

A central issue in vision science is how various visual mechanisms operate and combine their outputs during the performance of natural tasks. In recent years, improvements in computational power and in virtual display systems have made it possible to conduct highly controlled experiments while subjects are performing complex naturalistic tasks. This session will describe some of the empirical findings that are emerging from recent advances in virtual display technology.

Martin Banks, Univ. of California at Berkeley, USA

Mary Hayhoe, Univ. of Texas at Austin, USA

Michele Rucci, Boston Univ., USA

Bill Warren, Dept. of Psychology, Brown Univ., USA

Close Window

The Stiles-Crawford Effects of the First and Second Kind, 75 Years of Scientific Achievements

In 1933 Drs. Walter Stanley Stiles, F.R.S., O.B.E., and Brian H. Crawford of the National Physical Laboratory at Teddington, Middlesex, England, described a new phenomenon, which became known as "the directional sensitivity of the retina", later as the Stiles-Crawford Effect of the first kind (SCE-I). These able scientists had designed a pupilometer, simple in concept, but important in the way it failed to achieve their rather straight-forward and intended goals. Properly interpreting the failure of their device and experiment, Stiles and Crawford deduced that rays entering the eye pupil off-center were less effective at stimulating vision. In later years, it was found that there were distinct differences in photopic and scotopic vision relative to these properties; that the effects noted largely occurred at the retina; and there were wavelength dependent aspects to the SCE-I. Also, they noted that varying angle of incidence of light at the retina altered perceived hue and saturation of visual stimuli, now known as SCE-II. These effects could be largely attributed to the fiber optics and waveguide properties of retinal receptors; the photoreceptors tended throughout life to point in close proximity to the center of the exit pupil of the eye (potentially the result of an active feedback mechanism on their orientations); characteristic anomalies of receptor orientation were recorded in certain disorders and diseases; and receptor alignments could be altered/affected and often could recover. Alterations in photoreceptor alignments could be induced by marked accommodation, by tractional strains (various), by use of a displaced pupil for a finite period of time, etc. In turn, these physical factors affected measurements of visual functions, and affected the absorption of radiant energy within the visual spectrum in producing a visual response (light). Today, we celebrate the 75th anniversary of the discovery of these effects, the scientists who have contributed to and defined these properties, and of our understanding of the major factors which result in these now well-known phenomena.

Jay Enoch, University of California, Berkeley

The Stiles-Crawford Effects, 75 years: A Brief History, and Experiences at the National Physical Laboratory, Teddington, England, With W.S. Stiles and B.H. Crawford

Vasudevan Lakshminarayanan, University of Waterloo

Photometric and Radiometric Issues Associated With Measurements of the Integrated Stiles-Crawford Effect I, and Specification of the Visual Stimulus

Austin Roorda, University of California, Berkeley

Optical Properties of Human Cone Photoreceptors Revealed with Adaptive Optics

Brian Vohnsen, University College Dublin

Waveguide Models and the Stiles Crawford Effects

Stacey S. Choi, University of California, Davis

Studies of the Stiles -Crawford Effect of the First Kind in Myopic Conditions

Close Window