Effects of luminance contrast and phase difference on motion assimilation for sinusoidal gratings

Visual discomfort from flicker: Effects of mean light level and contrast

Uncomfortable images generally have a particular spatial structure, which deviates from a reciprocal relationship between amplitude and spatial frequency (f) in the Fourier domain (1/f). Although flickering patterns with similar temporal structure also appear uncomfortable, the discomfort is affected by not only the amplitude spectrum but also the phase spectrum. Here we examined how discomfort from flicker with differing temporal profiles also varies as a function of the mean light level and luminance contrast of the stimulus. Participants were asked to rate discomfort for a 17° flickering uniform field at different light levels from scotopic to photopic. The flicker waveform was varied with a square wave or random phase spectrum and filtered by modulating the slope of the amplitude spectrum relative to 1/f. At photopic levels, the 1/f square wave flicker appeared most comfortable, whereas the discomfort from the random flicker increased monotonically as the slope of the amplitude spectrum decreased. This special status for the 1/f square wave condition was limited to photopic light levels. At the lower mesopic or scotopic levels, the effect of phase spectrum on the discomfort was diminished, with both phase spectra showing a monotonic change with the slope of the amplitude spectrum. We show that these changes cannot be accounted for by changes in the effective luminance contrast of the stimuli or by the responses from a linear model based on the temporal impulse responses under different light levels. However, discomfort from flicker is robustly correlated with judgments of the perceived naturalness of flicker across different contrasts and mean luminance levels.

Investigation of center-surround interaction in motion with reaction time for direction discrimination

How motion onset asynchrony (MOA) alters the effects of stimulus size on reaction time (RT) for direction discrimination of a drifting grating was examined. MOA is a delay from the stimulus onset to the onset of motion. Without MOA, RTs were found to increase as the stimulus size was increased at high contrast, but decrease with it at low contrast or at high noise levels. With MOA, however, RTs did not increase as the stimulus size increased even at high contrast. These results suggest that sudden stimulus onset evokes the increase of RTs with the increase of stimulus size at high contrast. RTs for direction discrimination of a drifting Gabor patch (the target) surrounded by a different drifting or a static grating as well as RTs for the target that was not surrounded by an additional grating were measured. The RTs for the target moving in the same or opposite direction as the motion of the surrounding grating were larger than those for the target with the static grating or no additional grating at moderate or high contrast. There was no significant difference between the RTs for the target moving in the same direction as the surrounding grating and the RTs for the target moving in the opposite direction. At low contrast and without MOA, however, the RTs for the target moving in the same direction as the surrounding grating were larger than those for the target moving in the opposite direction. These results suggest surround suppression at low contrast under some conditions. They also suggest that the decrease of RTs for discriminating motion direction of a drifting single Gabor patch with the increase of stimulus size at low contrast does not necessarily mean the absence of surround suppression.

Differential effect of luminance contrast reduction and noise on motion induction

Motion perception in a region is affected by motion in the surround regions. When a physically static or flickering stimulus surrounded by moving stimuli appears to move in the direction opposite to that of the surround motion, it is referred to as motion contrast. When the centre appears to move in the same direction, it is referred to as motion assimilation. We investigated how noise and luminance contrast affect motion induction by employing static and dynamic counterphase flickering targets. The tendency of motion assimilation was found to be stronger at a high noise level than at a low noise level for both static and dynamic targets. On the other hand, a decrease of luminance contrast tended to strengthen the tendency of motion contrast. However, the addition of noise and the decrease of luminance contrast decreased the visibility of motion comparably. These results suggest that the visual system changes the mode of motion induction according to the noise level, but not the visibility.

Effects of the noise level on induced motion

Motion in a part of the field induces motion in an adjoining region. In this study, it was investigated how the noise level affects induced motion of a counterphase flickering (target) grating due to adjacent drifting (inducer) gratings. It was shown that at low noise levels, motion contrast occurred, and at high noise levels, motion assimilation occurred. When the noise level was randomly set for each trial, the adaptive change with the noise level was also observed. The result suggests that the adaptive change occurs for a short period. It was also found that noise for the target as well as noise for the inducers contributes to the effect of noise on motion induction. It suggests that the overall noise level is crucial for the effect. The study provided evidence that motion integration changes from a spatially band-pass operation to a low-pass operation as the signal-to-noise ratio (SNR) decreases.

Perceptual switching, eye movements, and the bus paradox

According to a widely cited finding by Ellis and Stark (1978 Perception 7 575-581), the duration of eye fixations is longer at the instant of perceptual reversal of an ambiguous figure than before or after the reversal. However, long fixations are more likely to include samples of an independent random event than are short fixations. This sampling bias would produce the pattern of results also when no correlation exists between fixation duration and perceptual reversals. When an appropriate correction is applied to the measurement of fixation durations, the effect disappears. In fact, there are fewer actual button-presses during the long intervals than would be expected by chance. Moving-window analyses performed on eye-fixation data reveal that no unique eye event is associated with switching behaviour. However, several indicators, such as blink frequency, saccade frequency, and the direction of the saccade, are each differentially sensitive to perceptual and response-related aspects of the switching process. The time course of these indicators depicts switching behaviour as a process of cascaded stages.

Motion capture depends on signal strength

When flickering dots are superimposed onto a drifting grating, the dots appear to move coherently with the grating. In this study we examine: (i) how the perceived direction of a compound stimulus composed of superimposed grating and dots, moving in opposite directions with equal speeds, is influenced by the relative strength of the motion signals; (ii) how the perceived speed of a compound stimulus composed of superimposed grating and dots, moving in the same direction but at different speeds, is influenced by the relative strength of the motion signals; and (iii) whether this stimulus is discriminable from its metameric speed match. Dot signal strength was manipulated by using different proportions of signal dots in noise and different dot lifetimes. Both the perceived direction and speed of these compound stimuli depended upon the relative motion-signal strengths of the grating and the dots. Those compound stimuli that appeared coherent were not discriminable from the speed-matched metameric compound stimuli. When the signals were completely integrated into a coherent compound stimulus, the local motion signals were no longer perceptually available, though both contributed to the global percept. These data strongly support a weighted-combination model where the relative weights depend on signal strength, instead of a winner-takes-all model.

Perceptual dimorphism in visual motion from stationary patterns

Fraser and Wilcox [1979 Nature (London) 281 565-566] devised a series of complex stationary patterns that provoked episodes of compelling illusory motion, but only in about two-thirds of people tested. Using simplified versions of their stimuli, we have confirmed their claim of perceptual dimorphism. We show that the strength of the illusory motion depends upon stimulus duration, eccentricity, and contrast. The illusory motion does not require fluctuations in accommodation, as has been suggested for some other forms of illusory motion. Finally, we consider the relation of Fraser-type motion to other forms of illusory motion.

Summation between nearby motion signals and facilitative/inhibitory interactions between distant motion signals

To explain the finding that motion assimilation was dominant between nearby motion signals while motion contrast between distant ones, a center-surround antagonistic mechanism was proposed [Nawrot & Sekuler (1990). Vision Research, 30, 1439-1451]. However, motion assimilation occurred not only between nearby signals but also between distant ones, suggesting the existence of a center-surround non-antagonistic mechanism [Ido. Ohtani & Ejima (1997). Vision Research, 37, 1565-1574]. The present study was designed to provide direct evidence for the non-antagonistic mechanism, and to examine further the motion interactions which operate in different spatial scales. The nature of motion interaction between the test and the inducer was examined by varying the size, the number of frames, the frame duration and the inter-frame displacement of random-dot kinematograms. The results were consistent with the notion that there are three types of interactions in human motion processing; one is a summation process effective within nearby regions, and the other two are facilitative and inhibitory induction processes operating over larger spatial scales. Analysis of the results in terms of the Fourier components suggests that the facilitative and the inhibitory induction processes may be sensitive, respectively to the lower and the higher temporal frequency components of the stimulus.

Motion assimilation for expansion/contraction and rotation and its spatial properties

In a two-frame apparent motion display, a test grating was displaced horizontally or vertically in the presence of an inducer of which component gratings made up expanding/contracting or rotational motion as a whole. In the first experiment, we demonstrated that motion assimilation did occur for the test accompanied by the two-dimensional motion of the inducer. In the second experiment, we showed that the spatial limit of motion assimilation for expansion/contraction or rotation was large, extending over at least a visual angle of 14-21 deg in diameter, but spatial summation did not occur within the limit. The results were discussed in terms of the interaction between local motion detectors and higher-order detectors which monitor global motion of the whole stimulus pattern.

Dependencies of motion assimilation and motion contrast on spatial properties of stimuli: spatial-frequency nonselective and selective interactions between local motion detectors

Two sets of experiments were carried out to examine dependencies of two types of induced motion (motion assimilation and motion contrast) on spatial properties of stimuli in terms of spatial-frequency tuning of local motion detectors. In the first set, the magnitudes of motion assimilation and motion contrast for a sinusoidal grating were measured at a function of the spatial frequency of the inducing gratings, with the spatial frequency of the test grating as a parameter. In the second set, the magnitudes were measured as a function of the height of the inducing gratings with the spatial frequencies of the test and the inducing gratings as parameters. For motion assimilation, the magnitude was characterized by a low-pass function of the spatial frequency of the inducing gratings, and the critical height of the inducing gratings, which demarcates the extent of the spatial pooling, varied systematically depending on the spatial frequency of the inducing gratings. For motion contrast, on the other hand, the magnitude was characterized by a hand-pass function, and the critical height depended on the frequency of the test grating. These results suggest that motion assimilation is mediated by the spatial-frequency nonselective interaction between the local detectors, in which the motion signals of the detectors tuned to different spatial frequencies are integrated with each other. Motion contrast is mediated by the spatial-frequency selective interaction, in which the motion signals of the local detectors tuned to the same or similar spatial frequencies are compared and differentiated.

Anisotropy for direction discrimination in a two-frame apparent motion display

Direction discrimination (upward/downward or left/right) for a Gabor patch in a two-frame motion display was measured as a function of the inter-frame displacement size of the component grating with the stimulus position (center, left, right, upper and lower visual fields) as a parameter. The results showed that, for vertical motion in the center, left, right and lower visual fields, the observers saw downward motion more frequently than upward motion, whereas for vertical motion in the upper field and for horizontal motion, no preference for one of the two opposite directions was obtained. Human motion vision is anisotropic in the lower half of the visual field.

Simultaneous motion contrast across space: involvement of second-order motion?

A static or counterphase (target) grating surrounded by drifting (inducer) gratings is perceived to move in the direction opposite that of the inducers. We compared the relative magnitudes of these simultaneous motion contrasts generated by both first-order and second-order stimuli. The first-order stimuli were sinusoidal luminance-modulations of a uniform field, and the second-order stimuli were sinusoidal contrast-modulations of a random-dot field. When the target was a static grating, the second-order stimuli induced little motion contrast, while the first-order stimuli of the same effective contrast produced clear motion contrast. When the target was a counterphase grating, both first- and second-order stimuli produced clear motion contrast. These results are discussed in relation to the involvement of second-order motion pathways in the relative-motion processing, and the two types of motion aftereffects obtained with static and dynamic test stimuli.