Brightness and lightness

Articulation effects of 3D backgrounds in the simultaneous lightness contrast illusion

Galina Menshikova, Elizaveta Luniakova, Denis Zakharkin and Maxim Pestun

Articulation effects can influence the strength of lightness illusions (Gilchrist et al., 1999, Psychological Review, 4 (109), 795-834). We investigated the role of 3D articulated backgrounds in the perception of the simultaneous lightness contrast (SLC) illusion. The 2D articulated version of the SLC illusion was used to construct different 3D displays. They consisted of test squares which were moved out of the backgrounds and three different types of 3D backgrounds: 1) 2D patches 2) cubes 3) balls. The luminance range was 1:230. The 2D articulated version of the SLC illusion and three types of 3D displays were presented using the CAVE system. Twenty five observers (age range 17-30) were tested. The method of constant stimuli was used to measure the illusion strength. The results showed that the illusion strength decreased for all 3D displays relative to the 2D articulated version. There were no significant differences between different types of 3D displays. It seems that articulation effects in the perception of 3D SLC illusion weakly depend on the type of 3D backgrounds. Supported by RFBR grant 12-07-00146a

Embedding color perception into a variational framework

Edoardo Provenzi and Vicent Caselles

Color perception is a fascinating research field that is receiving more and more attention due to its intrinsical interdisciplinary nature. In the literature, one can find many different models of color perception that focus on various human visual system properties and try to implement them in distinct ways. We will present a general variational framework to model color perception and show that many existing models (notably the famous Retinex theory) can be embedded in this framework. The energy functionals that we consider are composed by two opposite terms: one that induces a contrast intensification in accordance with Weber-Fechner's law of contrast perception and the other that tries to push the intensity values towards the middle gray. The balance between these opponent actions realizes a perceptually-inspired color correction that strongly reduces the effects of color shifts due to illumination changes and notably enhances under and over-exposed images. Variational principles first appeared in Physics with Fermat's principle and then extended throughout the science, providing, as in the present case, a view from above on complex problems.

Induced curvature and size change by a series of luminance ramps

Mariann Hudak and János Geier

A geometrical illusion is presented, induced by a series of luminance ramps progressing from black to white back and forth. If this grating is displayed on a grey background, surrounded by two thin lines (1 pixel) at the upper and lower boundaries, the lines appear to be curved, while the grating seems thinner where its luminance is low compared to the line, and thicker, where the contrast is high. The illusion also works when several rows of the gratings and the lines are placed under each other in counterphase, but not, if the rows are not in counterphase. The illusion is strongest when the lines are black or white, and weak in case of grey lines. These laws are entirely the opposite to that of the Café-Wall illusion. Another difference is that while in case of our illusion, one row of luminance ramp is sufficient to elicit the illusory curvature, at least two rows of tiles are necessary in case of the Café-Wall. One common feature is that if the lines are thicker, both illusions cease. If the rows are not separated by lines, the boundary edges themselves are not distorted. The illusion is also elicited by chromatic ramps.

Is the tremor of an eye involved in the decomposition of a complex image into simpler fragments over time?

Dalius Matuzevicius and Henrikas Vaitkevicius

Although, eye movements have been extensively investigated, functions of the tiny eye movements, known as tremor, and its impact on vision are not understood yet. Generally, a cancellation process of the fading effect, information reduction and hyperacuity phenomenon are associated with eye micromovements. We suggest that eye tremor can be related to the calculations of the brightness gradient along the path of the eye movement. Numerical simulations of tremor and its effect on image formation on the retina were performed in Matlab environment. Taking into account the results of the modelling, we conclude that (i) eye tremor can serve as a way to decompose complex image into time-separated simple segments; such procedure simplifies segmentation process of images that satisfy condition of similarity and continuity, and provides positions of imaginary contours; (ii) moreover, owing to the tremor, edges may be specified with sub-photoreceptor accuracy that should increase the acuity of vision; (iii) further analysis of the complex image may be carried out as a processing of the sequence of feature-like images which are simplified fragments of the complex image. [Postdoctoral fellowship is being funded by European Union Structural Funds project 'Postdoctoral Fellowship Implementation in Lithuania']

Brightness variation in function of stimulus exposure time

Simone Gori, Luca Ronconi, Tiziano Agostini and Andrea Facoetti

Brightness contrast can be the result of an interaction process with the luminance of the immediate surround (local interactions) or a consequence of perceptual organization (global factors). It is already known that a display with competing local and global factors without exposure time limitations results in a perceived brightness contrast induced by global factors. In this study we used reverse contrast Necker cube type displays with the relative controls. The two experimental stimuli were two medium-gray dashed cubes. The first one, having dark inducer corners, was placed on a light inducer background and the second one, having light inducer corners, was placed on a dark inducer background. The control stimuli were identical to the two experimental displays except that the inducer corners were medium gray. The stimuli were presented for 200, 300, 500 and 1000 ms. We asked observers to match the brightness of the dashed lines of each cube to the brightness of a gray scale. Our results show that the brightness change in function of the exposure time: local factors produced the brightness contrast effect at short stimulus exposure time while at longer exposure time the global factors win the competition and become crucial for brightness perception.

Effect of stimulus intensity on LRP latency, RT in simple and choice tasks

Agnieszka Nowik, Jerzy Moczko and Ewa Marzec

Van der Molen and Keuss [1979, Quarterly Journal of Experimental Psychology, 31, 95-102; 1981, Quarterly Journal of Experimental Psychology, 33, 177-184] reported U-shaped relationship between reaction time (RT) and loudness in difficult tasks requiring choice responses. This effect was replicated by Jaśkowski and Włodarczyk [2006, International Journal of Psychophysiology, 61, 98-112] for ultrabright and large visual stimuli. In the current study, we used ERP to investigate the locus of this paradoxical elongation of RTs for extremely bright and large stimuli. The luminance of stimuli was manipulated. Same we also tested a different group of participants with two disparate auditory tons and a five different loudness conditions, task simple and choice reaction. The RT-luminance relationship was monotonic for simple responses and U-shaped for choice responses. Notably, LRP-R was independent of stimulus intensity for both tasks. S-LRP latency changed with brightness similarly to RTs. These results support Van der Molen and Keuss'proposal that it is the response selection stage that is affected by very strong stimuli. Our study clearly indicates that response selection is influenced by intensity changes irrespective of whether visual or auditory stimuli are used, resulting in a U-shaped relationship between RT and intensity when the task is difficult.

Filling-in versus multiscale filtering: Measuring the magnitude of induction as a function of the distance from the test/inducing field edge

Barbara Blakeslee and Mark McCourt

Brightness induction predictions for a filling-in versus a multiscale filtering mechanism were tested by comparing them to quadrature-phase motion cancellation and static matching measures of induction magnitude as a function of distance from the test/inducing field edge. Filling-in predicts that brightness information derived from the test/inducing field edge fills-in to produce a homogeneous test field percept. Multiscale filtering, however, predicts that induction magnitude will decrease with distance from the test/inducing field edge. In the motion cancellation experiments the 0.0625 c/d inducing grating was counterphased at a temporal frequency of 1 Hz for test field heights of 3o and 9o. A 0.25o quadrature probe grating was added to the test field at various distances from the test/inducing field edge and observers made left/right motion judgments of the induced+quad probe compound. In the static matching experiments subjects made point-by-point matches to test fields of 3o, 6o, and 9o using a small matching patch. The vertical position of the patch cued the vertical location within the display to be matched. The results of both studies show that the magnitude of induction decreases with increasing distance from the test/inducing field edge. These results are consistent with multiscale filtering but not with a filling-in process.

Spatial Structure of Contextual Modulation

Colin Clifford and Isabelle Mareschal

Contextual effects are ubiquitous in vision and may reveal fundamental principles of sensory processing. Here, we used a reverse correlation technique to investigate the spatial determinants of contextual modulation in orientation (the Tilt illusion) and lightness (White's Illusion). The surround consisted of windows each revealing one of two opposite inducing stimuli. The borders between windows were defined by the zero-crossings of band-pass filtered noise. Subjects had to make a forced-choice judgment on a fixed test stimulus. We were thus able to build up classification images showing which regions of the surround had the greatest influence on the perception of the test. The spatial structure in classification images for the Tilt Illusion showed considerable inter-observer variability. However, comparing each observer's results for horizontal and vertical test orientations revealed a high degree of consistency. This lack of dependence on test orientation argues against accounts of the tilt illusion that place particular reliance on collinear interactions. The classification images for White's Illusion contained structure localized along the flanks of the test bar, indicating that the primary effect on perceived lightness in our experiment is assimilation. However, we cannot rule out at least some additional contribution of contrast with the collinear end regions.

Switching off brightness induction through induction-reversed images

Olivier Penacchio, Laura Dempere-Marco and Xavier Otazu

Brightness induction is the modulation of the perceived intensity of an area by the luminance of surrounding areas. Although V1 is traditionally regarded as an area mostly responsive to retinal information, neurophysiological evidence suggests that it may explicitly represent brightness information. In this work, we investigate possible neural mechanisms underlying brightness induction. To this end, we consider the model by Z.Li (Comput Neural Syst, 10 (1999): 187-212) which is constrained by neurophysiological data and focuses on the part of V1 responsible for contextual influences. This model, which has proven to account for phenomena such as contour detection and preattentive segmentation, shares with brightness induction the relevant effect of contextual influences. Importantly, the input to our network model derives from a complete multiscale and multiorientation wavelet decomposition, which makes it possible to recover an image reflecting the perceived luminance and successfully accounts for well known psychophysical effects for both static and dynamic contexts. By further considering inverse problem techniques we define induction-reversed images: given a target image, we build an image whose perceived luminance matches the actual luminance of the original stimulus, thus effectively canceling out brightness induction effects. We suggest that induction-reversed images may help remove undesired perceptual effects and can find potential applications in fields such as radiological image interpretation.

Effect of relative total area on perceived lightness

Huseyin Boyaci, Mahru Kobal and Ece Subasi

In a simple two dimensional (2D) display composed of two uniform contiguous surfaces of different luminances, perceived lightness of the surfaces varies as a function of their relative areas while their luminance remains unchanged. Despite empirical evidence dating back to the first half of the 20th century, there is still no clear explanation for this 'area rule' of lightness [Gilchrist and Radonjic, 2009, Journal of Vision, 9(9):13, 1-10]. While contiguous surfaces have been used in classical studies, here we investigate the effect using stimuli with non-contagious light and dark regions. Stimuli were computer generated 2D disks, which varied in relative area of light and dark regions with 9 dark wedges of 11 to 354 degrees. The task was to estimate the lightness of dark wedges by adjusting the luminance of an adjustable patch. In line with classical results, we found that the lightness of the darker region increased with its relative area. In the second experiment we used the same stimulus except now the darker regions were divided into two equal parts. Surprisingly, we found that their lightness is affected by not only their individual areas but also the sum of their areas.