De Vries-Rose Research Articles

Light adaptation controls visual sensitivity by adjusting the speed and gain of the response to light.

The range of c. 1012 ambient light levels to which we can be exposed massively exceeds the <103 response range of neurons in the visual system, but we can see well in dim starlight and bright sunlight. This remarkable ability is achieved largely by a speeding up of the visual response as light levels increase, causing characteristic changes in our sensitivity to different rates of flicker. Here, we account for over 65 years of flicker-sensitivity measurements with an elegantly-simple, physiologically-relevant model built from first-order low-pass filters and subtractive inhibition. There are only two intensity-dependent model parameters: one adjusts the speed of the visual response by shortening the time constants of some of the filters in the direct cascade as well as those in the inhibitory stages; the other parameter adjusts the overall gain at higher light levels. After reviewing the physiological literature, we associate the variable gain and three of the variable-speed filters with biochemical processes in cone photoreceptors, and a further variable-speed filter with processes in ganglion cells. The variable-speed but fixed-strength subtractive inhibition is most likely associated with lateral connections in the retina. Additional fixed-speed filters may be more central. The model can explain the important characteristics of human flicker-sensitivity including the approximate dependences of low-frequency sensitivity on contrast (Weber’s law) and of high-frequency sensitivity on amplitude (“high-frequency linearity”), the exponential loss of high-frequency sensitivity with increasing frequency, and the logarithmic increase in temporal acuity with light level (Ferry-Porter law). In the time-domain, the model can account for several characteristics of flash sensitivity including changes in contrast sensitivity with light level (de Vries-Rose and Weber’s laws) and changes in temporal summation (Bloch’s law). The new model provides fundamental insights into the workings of the visual system and gives a simple account of many visual phenomena.

Visual Performance as a Function of Luminance in Glaucoma: The De Vries-Rose, Weber's, and Ferry-Porter's Law.

To determine whether the De Vries-Rose, Weber's, and Ferry-Porter's law, which describe visual performance as a function of luminance, also hold in patients with glaucoma. A case-control study with 19 glaucoma patients and 45 controls, all with normal visual acuity. We measured foveal and peripheral contrast sensitivity (CS) using static perimetry and foveal and peripheral critical fusion frequency (CFF; stimulus diameter 1°) as a function of luminance (0.02 to 200 cd/m2). ANOVA was used to analyze the effect of glaucoma and luminance on CS and CFF; analyses were adjusted for age and sex. Foveally, logCS was proportional to log luminance at lower luminances (de Vries-Rose) and saturated at higher luminances (Weber); glaucoma patients had a 0.4 log unit lower logCS than controls (P < 0.001), independent of luminance. Peripherally, the difference was more pronounced at lower luminances (P = 0.007). CFF was linearly related to log luminance (Ferry-Porter). Glaucoma patients had a lower CFF compared with controls (P < 0.001), with a smaller slope of the CFF versus log luminance curve, for both the fovea (6.8 vs. 8.7 Hz/log unit; P < 0.001) and the periphery (2.5 vs. 3.4 Hz/log unit; P = 0.012). Even in apparently intact areas of the visual field, visual performance is worse in glaucoma patients than in healthy subjects for a wide range of luminances, without a clear luminance dependency that is consistent across the various experiments. This indicates impaired signal processing downstream in the retina and beyond, rather than an impaired light adaptation in the strictest sense.

Internal noise sources limiting contrast sensitivity

Contrast sensitivity varies substantially as a function of spatial frequency and luminance intensity. The variation as a function of luminance intensity is well known and characterized by three laws that can be attributed to the impact of three internal noise sources: early spontaneous neural activity limiting contrast sensitivity at low luminance intensities (i.e. early noise responsible for the linear law), probabilistic photon absorption at intermediate luminance intensities (i.e. photon noise responsible for de Vries-Rose law) and late spontaneous neural activity at high luminance intensities (i.e. late noise responsible for Weber’s law). The aim of this study was to characterize how the impact of these three internal noise sources vary with spatial frequency and determine which one is limiting contrast sensitivity as a function of luminance intensity and spatial frequency. To estimate the impact of the different internal noise sources, the current study used an external noise paradigm to factorize contrast sensitivity into equivalent input noise and calculation efficiency over a wide range of luminance intensities and spatial frequencies. The impact of early and late noise was found to drop linearly with spatial frequency, whereas the impact of photon noise rose with spatial frequency due to ocular factors.

Brightness in human rod vision depends on slow neural adaptation to quantum statistics of light.

In human rod-mediated vision, threshold for small, brief flashes rises in proportion to the square root of adapting luminance at all but the lowest and highest adapting intensities. A classical signal detection theory from Rose (1942, 1948) and de Vries (1943) attributed this rise to the perceptual masking of weak flashes by Poisson fluctuations in photon absorptions from the adapting field. However, previous work by Brown and Rudd (1998) demonstrated that the square-root law also holds for suprathreshold brightness judgments, a finding that supports an alternative explanation of the square-root sensitivity changes as a consequence of physiological adaptation (i.e., neural gain control). Here, we employ a dichoptic matching technique to investigate the properties of this brightness gain control. We show that the brightness gain control: 1) affects the brightness of high-intensity suprathreshold flashes for which assumptions of the de Vries-Rose theory are strongly violated; 2) exhibits a long time course of 100-200 s; and 3) is subject to modulation by temporal contrast noise when the mean adapting luminance is held constant. These findings are consistent with the hypothesis that the square-root law results from a slow neural adaptation to statistical noise in the rod pool. We suggest that such adaptation may function to reduce the probability of spurious ganglion cell spiking activity due to photon fluctuation noise as the ambient illumination level is increased.

Perceptual Uniformity for High-Dynamic-Range Television Systems

High dynamic range (HDR) television promises a qualitatively better viewing experience, but to fully realize its potential and to achieve consistent high quality images, the psychovisual aspects of human vision must be taken into account. This paper reviews the nonlinearities inherent in conventional standard dynamic range (SDR) television and discusses how these are required to accommodate psychovisual effects. The thresholds at which changes in brightness can be detected, embodied in Weber's and De Vries-Rose's laws, are discussed in the context of the need to minimize “banding” artifacts due to quantization. This effect also provides a simple psychovisual definition for the dynamic range of video. The second important psychovisual effect considered is that of surround brightness on the perception of emissive images in dim or dark surroundings. This effect is accommodated by implementing an overall system nonlinearity, the rendering intent, which is typically a gamma (power law) function. The appropriate gamma exponent is considered in special cases and interpolated to provide a general formula. Based on these psychovisual effects, an HDR television system is proposed that supports the highest quality HDR images with a simultaneous dynamic range substantially beyond the limits of the human visual system in a single adaptation state. It is defined by a single opto-electronic transfer function (OETF), with a variable display electro-optic transfer function (EOTF) to accommodate the eye's adaptation. It is also compatible with existing, SDR, displays and infrastructure and requires no metadata.

Dim-Light Sensitivity of Cells in the Awake Cat's Lateral Geniculate and Medial Interlaminar Nuclei: A Correlation With Behavior

Contrast thresholds of cells in the dorsal lateral geniculate (LGNd) and medial interlaminar (MIN) nuclei of awake cats were measured for scotopic and mesopic vision with drifting sine gratings (1/8, 2, and 4 cycles/deg [cpd]; 4-Hz temporal frequency). Thresholds for mean firing rate (F0) and temporally modulated responses (F1) were derived with receiver-operating-characteristic analyses and compared with behavioral measures recently reported by Kang and colleagues. Behavioral sensitivity was predicted by the neural responses of the most sensitive combinations of cell class and response mode: Y-cell F1 responses for 1/8 cpd, X-cell F1 responses for 2 cpd, and Y-cell F0 responses for 4 cpd. All previous estimates of neural scotopic increment thresholds in animal models fell between Weber's law (proportional to retinal illuminance) and the deVries-Rose law (proportional to the square root of illuminance). However, psychophysical experiments suggest that under appropriate conditions human scotopic vision follows the deVries-Rose law. If behavioral sensitivity is assumed to be determined by the most sensitive class of cells, this discrepancy is resolved. Under scotopic conditions, off-center Y cells were the most sensitive and these followed the deVries-Rose law fairly closely. MIN Y cells were, on average, 0.25 log units more sensitive than LGNd Y cells under scotopic conditions, supporting a previous proposal that the MIN is a specialization of the carnivore for dim-light vision. We conclude that both physiologically and behaviorally, cat and human scotopic vision are fundamentally similar, including adherence to the deVries-Rose law for detection of Gabor functions.

Contrast Sensitivity of Cats and Humans in Scotopic and Mesopic Conditions

Human contrast sensitivity in low scotopic conditions is regulated according to the deVries-Rose law. Previous cat behavioral data, as well as cat and mice electrophysiological data, have not confirmed this relationship. To resolve this discrepancy at the behavioral level, we compared sensitivity in dim light for cats and humans in parallel experiments using the same visual stimuli and similar behavioral paradigms. Both species had to detect Gabor functions (SD = 1.5 degrees, spatial frequencies from 0 to 4 cpd, temporal frequency 4 Hz) presented 8 degrees to the right or left of a central fixation point over an 8 log-unit range of adaptation levels spanning scotopic vision and extending well into the mesopic range. Cats had 0.74 log unit greater absolute sensitivity than that of humans for spatial frequencies <or=1/8 cpd. Cats had better contrast sensitivity overall for spatial frequencies <1/2 cpd, whereas humans were more sensitive for spatial frequencies above this. However, most of the cat's sensitivity advantage for low spatial frequencies could be accounted for by the greater light-concentrating abilities of its optics. Contrast sensitivity to 4 cpd was poor or absent in the scotopic range for both species. For both, scotopic increment thresholds were proportional to the square root of retinal illuminance, in accordance with the deVries-Rose law. Overall, cat and human visual systems appear to operate under very similar constraints for rod vision, including the regulation of contrast sensitivity across adaptation levels. A companion paper compares sensitivity of neurons in the lateral geniculate nucleus to these behavioral data.

A new quality metric based on just-noticeable difference, perceptual regions, edge extraction and human vision

Several human visual system (HVS) based quality metrics have been developed in recent years to measure the quality distortions caused by digital image coding techniques. Because of the complicated nature of the HVS characteristics, these metrics do not provide acceptable correlation with perceptual evaluations of the distortion. Recent studies by the Visual Quality Experts Group (VQEG) also show that the current HVS-based techniques do not provide a clear advantage over a mathematically defined technique such as mean square error (MSE). Therefore, this paper proposes a new quality metric based on just-noticeable difference threshold characteristics in the perceptual regions (dark, De Vries Rose, Weber and saturation regions), extraction of visually important edge details and human vision (low-level and high-level vision). Families of perceptual weights that are suitable for digitally compressed images with different distortions are generated, using compressed versions of a gradient image as stimuli and human observers in a subjective test. The simulations show that the proposed metric successfully measures the quality distortion and provides high correlation with both mathematical and HVS-based measures.

Changes in contrast thresholds with mean luminance for chromatic and luminance gratings: A reexamination of the transition from the DeVries–Rose to Weber regions

AbstractWe have examined the influence of the mean luminance level on the detection thresholds for luminance and red–green chromatic gratings for three different spatial frequencies. The changes in detection thresholds according to the mean luminance level reflect the two different regions, the DeVries–Rose and Weber ranges, found in previous studies. The results for luminance gratings suggest that the transition luminance is proportional to the spatial frequency of the grating. Predictions based on the constant‐flux hypothesis indicate, however, that the transition luminance is proportional to the square of the spatial frequency of the grating and so do not describe the distributions of luminance contrast thresholds adequately. For chromatic gratings, we obtained the same transition luminance for the two lowest spatial frequencies, showing that luminance and chromatic mechanisms behave differently as far as the dependence of the transition luminance on spatial frequency is concerned. Our results suggest that the transition luminance is related to the peak spatial frequency of visual mechanisms that respond to luminance and chromatic gratings. © 2004 Wiley Periodicals, Inc. Col Res Appl, 29, 177–182, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/col.20003

Evidence for a noise gain control mechanism in human vision.

For small, brief targets incremental threshold is known to obey the de Vries-Rose law: threshold rises in direct proportion to the square-root of background intensity. We present data demonstrating a square-root law for brightness matching as well. The square-root law for brightness is obtained over the full range of scotopic vision, and the low intensity end of photopic vision. The classic theory of de Vries and Rose explains the square-root law on the basis of increased variability of the photon count as the background increases. Our brightness matching data instead indicates that the mean signal level is reduced by a factor which is inversely proportional to the standard deviation of the photon count. This result is consistent with the idea that in the retina there exists a gain control mechanism that is sensitive to the variance in the photon input, rather than to the mean illuminance. The importance of this idea to the modelling of retinal gain controls is discussed.

A model of weber and noise gain control in the retina of the toad Bufo marinus

We present several variations of a model of gain control in the retina of the toad Bufo marinus, and use them to fit the threshold-vs-intensity data of an actual toad ganglion cell [Donner et al. (1990). Journal of General Physiology, 95, 733-753]. Our models are based on a proposal by Donner et al. that the gain (neural spike per photon ratio) of toad ganglion cells is set by a sequence of two retinal gain control stages. The first stage consists of a Weber gain control mechanism at the level of the red rods. The second is a more proximal "noise gain" stage, which multiplies the (incremental) input signal by a factor that is inversely proportional to the standard deviation of the random ganglion cell input and, under conditions that produce the de Vries-Rose threshold law, is also proportional to the standard deviation of the photon fluctuations within the ganglion cell receptive field. We demonstrate that noise gain control arises naturally from modeling ganglion cell spike generation with either of two common types of spike generation models: integrate-and-fire models or threshold accommodation models. We simulate the process of spike generation in both types of models and show that either model can account for the basic overall shape of the toad t.v.i. curve. However, although integrate-and-fire models appropriately generate noise gain control, they cannot quantitatively fit the threshold data with realistic retinal parameters. Integrate-and-fire models also fail to account for the observed relationship between the generator potential of the ganglion cell and its spiking probability. A threshold accommodation model with realistic retinal parameters, on the other hand, can account for both the threshold data and the generator potential-spike probability relationship. When a Weber gain stage is added to the model at the photoreceptor level, the resulting two-stage gain control model is shown to account quantitatively for the ganglion cell t.v.i. curve of Bufo marinus over the full range of background levels studied by Donner et al.

Mechanisms and Meaning of Devries—Rose Adaptation

‘Square-root’ or ‘deVries — Rose’ light adaptation is observed over a substantial luminance range in human foveal vision. The classical interpretation is that a detector (presumably in the brain) discriminates the neural signal evoked by the stimulus from the neural noise evoked by quantum fluctuations. It is known, however, that the retina may adjust its gain in inverse proportion to the square root of mean luminance, as observed eg in cat retinal ganglion cells under scotopic or mesopic adaptation. This kind of gain change is approximated even by the primary visual cells, the rods and cones, in at least some vertebrate species up to luminances producing 103 – 104 photoisomerisations per photoreceptor cell, per second. Is square-root adaptation in fact mainly an expression of an inverse-square-root gain in retinal cells? We investigated the roles of gain and noise in human foveal detection of 0.25 deg incremental spots presented for 50 ms on 5 deg steady backgrounds ranging from −0.25 to 2.35 log td, by measuring effects of pixel noise added to the stimulus and background. The results were consistent with the hypothesis that square-root adaptation mainly reflects gain changes, whereas the signal is detected against a constant level of neural noise. They were not consistent with the idea that signals proportional to stimulus intensity are detected against a noise that increases in proportion to quantum fluctuations. Thus, they do not support a simple interpretation of the deVries — Rose law. Still, an inverse-square-root retinal gain may in an evolutionary sense be seen as an adaptation to quantum fluctuations, in view of its functional consequences: (1) that output noise stays constant, independent of luminance level; (2) that light signals of constant statistical significance are encoded by visual responses of constant size.

Neural modulation transfer function of the human visual system at various eccentricities

We measured r.m.s. contrast sensitivity as a function of retinal illuminance at various spatial frequencies within 3–37 deg of eccentricity in the nasal visual field. In dim light contrast sensitivity increased in proportion to the square root of retinal illuminance obeying the De Vries-Rose law but in bright light contrast sensitivity was independent of luminance following Weber's law. Critical retinal illuminance ( I c ) marking the transition between the laws was found to be independent of grating area but proportional to the spatial frequency squared at all eccentricities, in agreement with the Van Nes-Bouman law of foveal vision. In addition, the proportionality constant was found to be independent of eccentricity and similar to that of the fovea. According to our contrast detection model of human vision the modulation transfer function ( P MTF ) of the neural visual pathways squared is directly proportional to the critical retinal illuminance. On this basis our result means that P MTF is similar, i.e. equal to spatial frequency across the visual field, thus attenuating low spatial frequencies relatively more than high spatial frequencies. Hence, up to the spatial cut-off frequency determined by the lowest neural sampling density of each retinal location the neural modulation transfer function is independent of visual location.

The dynamics of light adaptation in cat horizontal cell responses

In order to model the dynamic properties of light adaptation processes in cat horizontal (H-) cells, the time course of the gain adjustment following changes in the mean illumination level was studied. H-cell responses were recorded intracellularly in the optically intact, in vivo, eye of the cat. The light stimulus consisted of two spots, a large background spot (8.8 deg diameter) and a concentrically arranged smaller test spot (3.9 deg). The background was either square wave or sine wave modulated in intensity at a frequency of 0.2–1 Hz. The instantaneous value of the response gain was measured with brief flashes (10 msec) of the test spot, generated repetitively at a frequency of 5 or 10 Hz. Modulation of the background intensity, at a constrast of 0.6 and in the photopic range, effectively induces a modulation of the gain. The readjustment of the gain by a stepwise increase or decrease in background illumination is completed within about 200 msec. The amplitude of the gain modulation due to a 0.5 Hz background flicker is quantitatively comparable to that measured between steady illumination levels. Dynamic changes of the gain at low frequency stimuli therefore, have to be taken into account in modelling H-cell responses. For sinusoidal modulations of the background luminance the time course of gain adjustment is quantified by the phase shift of the gain modulation relative to background intensity modulation. The results, together with those described in two preceding papers, are used to test and discuss several light adaptation models that have been proposed previously. It was found that light adaptation in cat H-cells is described more adequately by a de Vries-Rose type of adaptation model than by a Weber type of light adaptation.

Temporal and spatial summation in the human rod visual system.

1. Absolute and increment thresholds were measured in a retinal region 12 deg temporal from the fovea with 520 nm targets of varying size and duration. Measurements were made under rod-isolation conditions in two normal observers and in a typical, complete achromat observer who has no cone-mediated vision. The purpose of these experiments was to determine how the temporal and spatial summation of rod-mediated vision changes with light adaptation. 2. The absolute threshold and the rise in increment threshold with background intensity depend upon target size and duration, but the psychophysically estimated dark light of the eye (the hypothetical light assumed to be equivalent to photoreceptor noise) does not. 3. The rise in increment threshold for tiny (10 min of arc), brief (10 ms) targets approaches the de Vries-Rose square-root law, varying according to the quantal fluctuations of the background light. The slope of the rod increment threshold versus background intensity (TVI) curves in logarithmic co-ordinates is about 0.56 +/- 0.04 (when cones are not influencing rod field adaptation). For large (6 deg) and long (200 ms) targets, a maximum slope of about 0.77 +/- 0.03 is attained. 4. The steeper slopes of the rod-detected TVI curves for large, long targets implies some reduction in temporal or spatial summation. In fact, the change in summation area is much more critical: under conditions where only the rod system is active the TVI curve slope is independent of target duration, suggesting that temporal summation is practically independent of background intensity. 5. The rise in threshold also depends on the wavelength of the background field in the normal observer but not in the achromat, confirming reports that the field adaptation of the rods is not independent of the quantal absorptions in the cones. The cone influence is most conspicuous on long-wavelength backgrounds and is found for all target sizes and durations, but is greater for large and long targets than for the other conditions.

Spatiotemporal contrast sensitivity of early vision

Based on the spatial and temporal statistics of natural images, a theory is developed that specifies spatiotemporal filters that maximize the flow of information through noisy channels of limited dynamic range. Sensitivities resulting from these spatiotemporal filters are very similar to the human spatiotemporal contrast sensitivity, including the dependence on ambient light intensity. The theory predicts several psychophysical laws: Ferry-Porter's law, the de Vries-Rose law, Weber's law, Bloch's law, Ricco's law, and Piper's law.

What Does the Retina Know about Natural Scenes?

By examining the experimental data on the statistical properties of natural scenes together with (retinal) contrast sensitivity data, we arrive at a first principle, theoretical hypothesis for the purpose of retinal processing and its relationship to an animal's environment. We argue that the retinal goal is to transform the visual input as much as possible into a statistically independent basis as the first step in creating a redundancy reduced representation in the cortex, as suggested by Barlow. The extent of this whitening of the input is limited, however, by the need to suppress input noise. Our explicit theoretical solutions for the retinal filters also show a simple dependence on mean stimulus luminance: they predict an approximate Weber law at low spatial frequencies and a De Vries-Rose law at high frequencies. Assuming that the dominant source of noise is quantum, we generate a family of contrast sensitivity curves as a function of mean luminance. This family is compared to psychophysical data.

Edge detection based on human visual response

An algorithm based on the characteristics of the human visual system is presented by which it is possible to select automatically (without human intervention) the thresholds for detecting the significant edges as perceived by human beings. The threshold value changes with the background intensity according to the criterion governed by the characteristic of one of the De Vries–Rose, Weber, and saturated regions. The effect of background size and splitting image (dynamic thresholding), and a provision for reducing the computation time are also included in the study. The algorithm is found to provide a satisfactory improvement in the performance over the conventional edge detection process for a wide range of input image.

Thresholding for edge detection using human psychovisual phenomena

An algorithm based on the facts of the human visual system is presented here whereby it is possible to select automatically (without human intervention) the thresholds for detecting the significant edges as perceived by human beings. The threshold value adapts with the background intensity according to the criterion governed by a characteristics of one of the De Vries-Rose, Weber's and saturated regions. The algorithm is found to provide a satisfactory improvement in the performance in the conventional edge detection process.

Psychophysical experiments on spatial summation at threshold level of the human peripheral retina

Experiments on spatial summation were done on the temporal retina between 4° and 40° eccentricity. The stimuli were circular discs between 7′ and 280′ of arc dia and 0.01 sec exposure at backgrounds from 0 to 10 −8 W/deg 2. The wavelength of the light was 510 nm for both background and test stimulus. The diameter of “Ricco's region” increases between 4° and 15° eccentricity and has an almost constant value of 100′ between 15° and 40°. For the smallest stimuli, deviations from Ricco's law were found. It appeared in increased thresholds between 15° and 40° eccentricity. We called this phenomenon the non-summation effect. The dependence of retinal sensitivity on eccentricity differs for different sizes of test stimulus and for absolute, scotopic and photopic background intensities. At photopic conditions increment thresholds are governed by Weber's law, the Weber fraction: increment intensity/background intensity ( Δ/LL) being dependent on retinal location and test stimulus size. For scotopic and mesopic adaptation levels at 7° eccentricity and small stimulus sizes the De Vries-Rose law: Δ/LL 0.50= constant holds. The transition from these power functions into Weber's law Δ/LL= c occurs all over the retina within the same narrow range of background intensities. Only the small stimuli that are subject to the non-summation effect also show a different behaviour in this respect.