A comparison of monkey and human motion processing mechanisms

Single-cell recording studies have provided vision scientists with a detailed understanding of motion processing at the neuronal level in non-human primates. However, despite the development of brain imaging techniques, it is not known to what extent the response characteristics of motion-sensitive neurons in monkey brain mirror those of human motion-sensitive neurons. Using a motion adaptation paradigm, the direction aftereffect, we recently provided evidence of a strong resemblance in the response functions of motion-sensitive neurons in monkey and human to moving dot patterns differing in dot density. Here we describe a series of experiments in which measurements of the direction aftereffect are used to infer the response characteristics of human motion-sensitive neurons when viewing transparent motion and moving patterns that differ in their signal-to-noise ratio (motion coherence). In the case of transparent motion stimuli, our data suggest suppressed activity of motion-sensitive neurons similar to that reported for macaque monkey. In the case of motion coherence, our results are indicative of a linear relationship between signal intensity (coherence) and neural activity; a pattern of activity which also bears a striking similarity to macaque neural activity. These findings strongly suggest that monkey and human motion-sensitive neurons exhibit similar response and inhibitory characteristics.     

Global motion processing is not tuned for binocular disparity.

An important goal of the visual system is the segmentation of image features into objects and their backgrounds. A primary cue for this is motion: when a region shares the same pattern of motion it is segregated from its surround. Three experiments were carried out to investigate whether the segmentation of image features on the basis of motion information is facilitated by the addition of binocular disparity. Coherence thresholds were measured for the discrimination of the global direction of motion of random dot kinematograms (RDKs) in which the relative disparity of the signal and noise dots was manipulated. When the signal dots were embedded in a three dimensional cloud of noise dots, coherence thresholds were similar to those measured when signal and noise dots were both presented with zero disparity. However, when the signal dots were separated from the noise dots in depth, global motion processing was strongly facilitated. These results were considered in terms of two models, one in which global motion is processed by disparity tuned mechanisms, the other in which the discrimination of the direction of motion is mediated by an attention-based system. It was concluded that global motion processing is not tuned for binocular disparity and that the facilitation of the discrimination of direction provided by binocular disparity in certain circumstances reflects the r?le of an attention-based system.     

Plasticity of human motion processing mechanisms following surgery for infantile esotropia

Monocular oscillatory-motion visual evoked potentials (VEPs) were measured in prospective and retrospective groups of infantile esotropia patients who had been aligned surgically at different ages. A nasalward-temporal response bias that is present prior to surgery was reduced below pre-surgery levels in the prospective group. Patients in the retrospective group who had been aligned before 2 yr of age showed lower levels of response asymmetry than those who were aligned after age 2. The data imply that binocular motion processing mechanisms in infantile esotropia patients are capable of some degree of recovery, and that this plasticity is restricted to a critical period of visual development.     

Orientation sensitivity in human visual motion processing

Orientation tuning of receptive fields is well documented in the spatial domain, but considerable variability exists amongst published estimates of orientation sensitivity of motion receptive fields. We used a two-frame motion sequence, in which one frame was binary noise and the other was a horizontally displaced and filtered version of the same noise field, to examine the orientation sensitivity of human motion mechanisms. Initially, orientations orthogonal to the direction of motion were removed from each filtered frame. Observers indicated perceived direction of motion in a single interval, binary choice task. D(max) was determined for different amounts of removed orientations, and found to remain constant across the removal of energy up to approximately +/-60 deg from vertical. In a second experiment, the orientations removed were now parallel to the direction of motion of the stimulus. D(max) fell as a cosine function with increasing removal of orientation information, in agreement with off-orientation looking or matched filtering predictions. The two experiments show the presence of mechanisms both broadly tuned and more narrowly tuned for orientation. A control experiment introduced an interstimulus interval between the two frames of our motion sequence. Performance on the direction discrimination task was severely degraded, indicating that the original results are not explicable in terms of a feature-tracking or long-range motion process. The presence of both broadly and narrowly tuned mechanisms implies multiple possible solutions to the processing of coherent plaid motion.     

Deficits to global motion processing in human amblyopia

We investigated global motion processing in a group of adult amblyopes using a method that allows us to factor out any influence of the known contrast sensitivity deficit. We show that there are independent global motion processing deficits in human amblyopia that are unrelated to the contrast sensitivity deficit, and that are more extensive for contrast-defined than for luminance-defined stimuli. We speculate that the site of these deficits must include the extra-striate cortex and in particular the dorsal pathway.     

Pattern specificity of human visual motion processing

Visual motion processing is strongly susceptible to adaptation. A variety of patterns have been used as stimuli in previous studies. Three of these, namely random dots, barcode-like gratings, and sinusoidal gratings, were compared in the present study using motion-onset visual evoked potentials (VEPs). We assessed the effects of the adaptation pattern and the test pattern to which the VEP is recorded. Furthermore, we evaluated the interaction between both, i.e. whether differences between adaptation and test pattern affect the response. Isodirectional and antidirectional adaptation were used to differentiate between the actual motion adaptation and associated flicker adaptation. Motion adaptation was almost 2.5-fold stronger (p < 0.01) if the same rather than different pattern types were used for both adaptation and test. This implies that separate neural populations are involved, suggesting the presence of pattern-tuned motion mechanisms.     

Three-dimensional motion aftereffects reveal distinct direction-selective mechanisms for binocular processing of motion through depth

Motion aftereffects are historically considered evidence for neuronal populations tuned to specific directions of motion. Despite a wealth of motion aftereffect studies investigating 2D (frontoparallel) motion mechanisms, there is a remarkable dearth of psychophysical evidence for neuronal populations selective for the direction of motion through depth (i.e., tuned to 3D motion). We compared the effects of prolonged viewing of unidirectional motion under dichoptic and monocular conditions and found large 3D motion aftereffects that could not be explained by simple inheritance of 2D monocular aftereffects. These results (1) demonstrate the existence of neurons tuned to 3D motion as distinct from monocular 2D mechanisms, (2) show that distinct 3D direction selectivity arises from both interocular velocity differences and changing disparities over time, and (3) provide a straightforward psychophysical tool for further probing 3D motion mechanisms.     

Explicit estimation of visual uncertainty in human motion processing

We examine whether human observers have explicit access to an estimate of their own uncertainty in extrapolating the motion trajectories of moving objects. Objects moved across a display area at constant speed changing direction at short time intervals. Each new direction was obtained by adding a random perturbation to the previous direction. The perturbation distribution was always symmetric with mean zero (no change in direction) but could differ in variability: objects with low directional variability tended to travel in straight lines while objects with high directional variability moved more erratically. Objects eventually disappeared behind the near edge of an occluder. Observers marked a 'capture region' along the far edge of the occluder that they estimated would contain the object when it re-emerged. We varied both occluder width and directional variability across trials and found that observers correctly compensated for these changes. We present a two-stage model of observer performance in which the visual system first estimates the directional variability of the object and then uses this estimate to set a capture region.     

Electrophysiological evidence for independent speed channels in human motion processing

A variety of psychophysical studies suggests that motion perception in humans is mediated by at least two speed-tuned channels. To study the neurophysiological underpinnings of these channels in the human visual cortex, we recorded visual evoked potentials (VEPs) to motion onset. We applied an adaptation paradigm that allowed us (a) to isolate and extract direction-specific cortical responses and (b) to assess cross-adaptation in the speed domain. VEPs resulting from the onset of left- or rightward motion at either low or high speeds were recorded from three occipital recording sites in 11 subjects. For each of these test stimuli, responses were collected after adaptation to one of five different conditions: a static adaptation pattern (baseline), adaptation to low-speed motion (3.5 degrees/s) either in the same or in the opposite direction as the test, or adaptation to high-speed motion (32 degrees/s) either in the same or in the opposite direction as the test. We report considerable direction-specific adaptation for same adaptation and test speeds (by 28-37% of baseline response; p <.002), whereas there was no direction-specific adaptation across speeds. We supplement these electrophysiological data with corresponding psychophysical results. The lack of direction-specific cross-adaptation in the speed domain demonstrated with physiological and psychophysical techniques supports models of at least two speed-tuned channels in the human motion system.     

Differential aging of motion processing mechanisms: evidence against general perceptual decline

While the percentage of older people in our society is steadily increasing, knowledge about perceptual changes during healthy aging is still limited. We investigated age effects on visual motion perception in order to differentiate between general decline and specific vulnerabilities. A total of 119 subjects ranging in age from 20 to 82 years participated in our study. Perceptual thresholds for different types of motion information, including translational motion, expanding radial flow, and biological motion, were determined. Results revealed a substantial increase of thresholds for translational motion with age. Biological motion perception was only moderately affected by age. For both motion types, threshold elevation seemed to develop gradually with age. In contrast, we found stable radial flow analysis across lifespan. There was no evidence that age effects were dependent on gender. Results demonstrate that visual capabilities are not equally prone to age-related decline. Surprisingly, higher motion complexity might not be necessarily associated with more pronounced perceptual constraints. We suggest that differential age effects on the perception of specific motion types might indicate that specialized neuronal processing mechanisms differ in their vulnerability to physiological changes during aging.     

Velocity blindness during shearing motion

When two objects lie at different distances from a moving observer, there is a velocity step at the occluding boundary of the nearer object. When the differential motion is tangential to the boundary, the effect is as if a shearing is taking place. If all texture cues are removed by using similar random dot patterns on each side of the boundary, then 20% of the population cannot use this differential motion to locate the boundary when it is presented to the parafovea. These observers are thus abnormally insensitive to the differential rate of texture flow at boundaries undergoing shearing motion. As no such population differences were observed for differential motion perpendicular to the occluding boundary (occluding motion), we infer that independent mechanisms process shearing and occluding motion.     

Co-operative interactions between first- and second-order mechanisms in the processing of structure from motion

Structure from motion (SFM) is the ability to perceive three-dimensional structure from stimuli containing only two-dimensional motion signals and this ability seems to be a result of high-level cortical processes. It has long been thought that local motion signals defined by second-order cues only weakly contribute to perception of SFM since performance on purely second-order SFM tasks is poor, relative to first-order stimuli. We hypothesized that the mechanisms responsible for deriving SFM were insensitive to low-level stimulus attributes such as the first- or second-order nature of the dots composing the stimulus, in other words: that they were "cue-invariant", but that large differences in sensitivity to local first- and second-order motions were responsible for previous findings. By manipulating the relative strength of first-order dots in an SFM stimulus that combines first- and second-order dots, we show that the two types of motion can separately support SFM and co-operatively interact to produce vivid three-dimensional percepts. This provides strong support that the mechanisms underlying SFM are cue-invariant.     

Binocular influences on global motion processing in the human visual system

This study investigates four key issues concerning the binocular properties of the mechanisms that encode global motion in human vision: (1) the extent of any binocular advantage; (2) the possible site of this binocular summation; (3) whether or not purely monocular inputs exist for global motion perception; (4) the extent of any dichoptic interaction. Global motion coherence thresholds were measured using random-dot-kinematograms as a function of the dot modulation depth (contrast) for translational, radial and circular flow fields. We found a marked binocular advantage of approximately 1.7, comparable for all three types of motion and the performance benefit was due to a contrast rather than a global motion enhancement. In addition, we found no evidence for any purely monocular influences on global motion detection. The results suggest that the site of binocular combination for global motion perception occurs prior to the extra-striate cortex where motion integration occurs. All cells involved are binocular and exhibit dichoptic interactions, suggesting the existence of a neural mechanism that involves more than just simple summation of the two monocular inputs.     

Electrophysiological correlates of vernier and relative motion mechanisms in human visual cortex

Vernier onset/offset thresholds were measured both psychophysically and with the steady-state VEP by introducing a series of horizontal breaks in a vertical square-wave luminance grating. Several diagnostic tests indicated that the first harmonic component of the evoked response generated by periodic modulation of offset gratings taps mechanisms that encode the relative position of spatial features. In the first test, a first harmonic component was only found with targets that contained transitions between collinear and noncollinear states. VEP vernier onset/offset thresholds obtained with foveal viewing were in the range of 15-22 arc sec. Control experiments with transitions between symmetrical, noncollinear patterns (relative motion) did not produce first harmonic components, nor did full-field motion of a collinear grating. A second series of experiments showed that VEP thresholds based on the first harmonic component of the vernier onset/offset response had an eccentricity dependence that was very similar to that found in a psychophysical discrimination task that required a left/right position judgment (vernier acuity). Other recordings showed that the first harmonic of the vernier onset/offset VEP was degraded by the introduction of a gap between stimulus elements, as is the displacement threshold. The vernier onset/offset target also produced a second harmonic component that was virtually identical to the one produced by a relative motion stimulus. Displacement thresholds based on these second harmonic components showed a more gradual decline with retinal eccentricity than did the first harmonic component elicited by vernier offsets. The second harmonic of the vernier onset/offset VEP was relatively unaffected by the introduction of gaps between the stimulus elements. The first and second harmonic components of the vernier onset/offset VEP thus tap different mechanisms, both of which support displacement thresholds that are finer than the resolution limits set by the spacing of the photoreceptors (hyperacuity).     

Asymmetric spatial frequency tuning of motion mechanisms in human vision revealed by masking

PURPOSE: To investigate the spatial frequency selectivity of the human motion system by using the technique of visual masking. METHODS: Modulation-depth thresholds for identifying the direction of a sinusoidal test pattern were measured over a range of spatial frequencies (0.25-4 cyc/deg) in the absence and presence of a temporally jittering mask. RESULTS: At the lowest test frequency (0.25 cyc/deg), maximum masking occurred when the test and mask shared the same spatial frequency, decreasing as the difference in spatial frequency between the test and mask increased. However, as test spatial frequency increased, maximum masking began to shift to when the mask was presented at approximately 1 octave below the test spatial frequency. Control experiments demonstrated that the asymmetric masking functions at higher test spatial frequencies was not affected by mask amplitude nor was it an effect of speed. The results confirmed that the peak at 1 octave from the test still occurred when the potential for off-frequency looking was minimized by presenting two masks positioned equidistant in frequency from the test grating. Control experiments revealed, however, that the peak at 1 octave below the test was mediated by image size and/or the number of cycles presented on screen. CONCLUSIONS: These findings provide support for the notion that motion perception is mediated by band-pass, spatial-frequency-selective mechanisms. Moreover, asymmetric tuning of the masking functions may reflect asymmetric spatial frequency selectivity of the mechanisms in the human visual system that encode motion or inhibition between mechanisms tuned to different spatial frequencies.     

Global motion processing in human color vision: a deficit for second-order stimuli

The investigation of the mechanism of global motion in color vision has been limited because the processing of the first-order chromatic RDK elements, based on low-level linear motion detectors, is impaired. Here we return to this problem by using second-order elements in a global motion stimulus. Second-order RDK elements were circular contrast-modulated (CM) envelopes of a low-pass filtered noise carrier. The stimuli were achromatic or isolated L/M- or S-cone opponent mechanisms. We measured simultaneously detection and motion direction identification thresholds at 100% motion coherence and at different RDK speeds with a 2-AFC paradigm. We found that direction identification thresholds were higher than detection thresholds for both chromatic and achromatic stimuli. The gap between these thresholds was greater for the chromatic than the achromatic stimuli and motion direction thresholds for the chromatic RDK were very high or impossible to obtain. We also measured global motion performance (RDK speed of 4 deg/s) by varying the coherence of limited lifetime RDK stimuli. Global motion thresholds could only be obtained for achromatic stimuli and not for chromatic ones. Within the limits of the present stimulus conditions, we found no global motion processing of second-order chromatic stimuli.     

Structural processing in biological motion perception

To investigate the basis for biological motion perception, structural and motion information were manipulated independently in a dynamic display using a novel stimulus with multiple apertures. Performance was compared in discrimination of global motion (translation and rotation) and biological motion. When structural information in the display was eliminated but motion information was intact, human observers were able to perceive global motion yet were at chance in discriminating walking direction of biological movement. In contrast, when the display provided even noisy and impoverished structural information, walking direction became identifiable. The present findings thus provide direct psychophysical evidence that motion information is insufficient and structural information is necessary for the identification of walking direction in biological movement. These findings imply that computational models must utilize a structural representation of the human body to account for perception of biological movements.     

The sequential processing of visual motion in the human electroretinogram and visual evoked potential

Mechanisms of motion vision in the human have been studied extensively by psychophysical methods but less frequently by electrophysiological techniques. It is the purpose of the present investigation to study electrical potentials of the eye (electroretinogram, ERG) and of the brain (visual evoked potential, VEP) in response to moving regular square-wave stripe patterns spanning a wide range of contrasts, spatial frequencies, and speeds. The results show that ERG amplitudes increase linearly with contrast while VEPs, in agreement with the literature, show an amplitude saturation at low contrast. Furthermore, retinal responses oscillate with the fundamental temporal stimulus frequency of the moving pattern while brain responses do not. In both the retina and the brain, the response amplitudes are tuned to certain speeds which is in agreement with the nonlinear correlation-type motion detector. Along the ascending slopes (which means increasing amplitudes) of the tuning functions, the ERG curves overlap at all spatial frequencies if plotted as a function of temporal stimulation frequency. The ascending slopes of the tuning functions of the VEP overlap if plotted as a function of speed. The descending slopes (which means decreasing amplitudes) of the tuning functions show little (ERG) or no (VEP) overlap and the waveforms at high speeds approach pattern-offset-onset responses. These observations suggest that in the retina motion processing along the ascending slopes of the tuning curves takes place by coding the temporal stimulation frequency which depends on the spatial frequency of the moving pattern. In the brain, however, motion processing is by speed independent of spatial frequency. Simple calculations show that the VEP information is decoded from the ERG signal into a speed signal.     

Linear mechanisms can produce motion sharpening

Human observers are not normally conscious of blur from moving objects [Nature 284 (1980) 164]. Several recent reports have even shown that blurred images appear sharper when drifting than when stationary and have suggested different non-linear mechanisms to explain this phenomenon [Vision Res. 36 (1996) 2729; Vision Res. 38 (1998) 2099]. We demonstrate here that even though distortions of drifting narrow-band sine-wave gratings cannot be explained by linear mechanisms, these mechanisms may have an important role in sharpening of moving edges. We show first that the effective spatial filter for a moving object that is formed by a simple difference-of-Gaussians spatial filter and the typical biphasic temporal impulse response function can be approximated by a combination of Gaussian filters only. When this filter is applied to moving, Gaussian-blurred edges, regions of blurring and sharpening are found over the same ranges of blur widths and velocities where recent experimental findings have shown them to exist. In general, that means that the output of the filter shows blurring in response to small blur widths and sharpening in response to larger blur widths.