First- and Second-order Motion
(clicking on the reference numbers in the text below will bring you to the reference and abstract)
Back to Research
- First- and second-order motion perception in Gabor micropattern stimuli: psychophysics and computational modelling
- Studies of neurological patients
- Anomalous perception of coherence and transparency
- Modelling first- and second-order motion
First- and second-order motion perception in Gabor micropattern stimuli: psychophysics and computational modelling. This paper examines the perception of first- and second-order motion in human vision. In an extension of previous work by Boulton and Baker [2,3], the direction of two-frame apparent motion is measured for stimuli composed of Gabor or Gaussian micropatterns. Three conditions are investigated. Condition 1 is that used by Boulton and Baker, in which motion is defined by the displacement of Gabor micropatterns. In Condition 2, motion is defined by the displacement of Gaussian micropatterns. In Condition 3, the envelopes of Gabor micropatterns are displaced while their carriers remain static. Using sparsely distributed micropatterns, direction judgements in all three conditions are determined by the spacing of the micropatterns. With a dense stimulus, direction judgements vary as a function of displacement in qualitatively different ways for the three conditions. The psychophysical results are predicted by a two-channel computational model. In one channel, motion is calculated directly from stimulus luminance, while in the other it is preceded by a texture-grabbing operation. The relative activities of the two channels dictates which governs direction judgements for any given stimulus.
Studies of neurological patients. We developed novel stimuli for investigating the relationship between local and global first- and second-order (1 and 2). We provided further evidence for separate processing pathways for first- and second-order motion from two neurological patients with unilateral brain damage (case RA and FD). Both patients were submitted to a detailed battery of tests spanning various levels and types of moving and static displays. Both patients presented a double dissociation of deficits between first- and second-order motion. a) FD was selectively impaired on all second-order motion tasks but had normal performance on their first-order motion counterparts and other first-order motion tasks. b) RA was selectively impaired on first-order motion tasks but performed normally on all second-order motion tasks.
Both, RA and FD?s deficits were long lasting (more than 2 years after the lesion occurred) and were confined to stimuli presented in the visual field contralateral to the lesion. Using the Cardview parcellation method we delineated the locus and extent of the patients? brain lesions and compared them anatomically. The parcellation of the lesions can be seen in . Briefly, F.D.'s lesion involved mostly the dorsolateral caudal aspect of the temporal lobe while RA's lesion involved parts of the caudal and superficial cuneus, the superior and caudal part of the lingual gyrus and the rostro-medial occipital cortex. Together, the locus of the lesions suggests that these two types of motion are mediated by different neural circuitry. These results suggest second-order motion may be mediated by a region in the visual cortex lying adjacent and dorsal to area MT which does not directly require cortical areas such as V1 and V3.
Anomalous perception of coherence and transparency. Moving plaid patterns generated by additive superposition of two sine wave gratings oriented at ±30i were viewed through a circular aperture subtending 8° of visual angle. The spatial frequency of one grating remained fixed at 1 cyc/deg while the other was varied between 0.25 and 2.5 cyc/deg in steps of 0.25 cyc/deg. Observers were required to indicate whether the moving pattern appeared "coherent" or "transparent". Data from five neurological patients and six normal subjects has been interpreted in terms of the operation and interaction of parallel first- and second-order motion processing channels. While some patients perceive coherent motion over a much smaller range of spatial frequencies than normal controls, others reported coherence over almost the entire range tested. Our data suggest a motion processing architecture in which linear and non-linear channels are present at multiple scales. At each scale the pairs of channels interact competitively with the outputs from each scale combined in additional stages (Clifford CWG, Vaina LM ?Anomalous Perception of Coherence and transparency in moving plaid Patterns: Evidence for Competing Processes?, Cognitive Brain Research , submitted).
Modeling first- and second -order motion. The clear cut double dissociation of deficits on first- and second-order motion discussed above motivated us to develop a model of first- and second-order motion processing in the normal human visual system. When elements of the model are disabled, performance on either first- or second-order motion can be selectively impaired in line with the neurological data. The model should be able to generate testable predictions on a range of psychophysical tasks involving normal and clinical subjects. We conducted two studies with this aim:a) To establish the nature of the computations necessary to extract first- or second-order motion we examined stimuli which have been used by us and other laboratories to assess first- and second-order motion processing capabilities in neurological patients. We proposed and implemented a computational model consisting of two parallel channels geared toward the perception of first- and second-order motion respectively . The image signal in the second-order channel was processed by a low-pass texture grabbing mechanism prior to motion analysis. In both channels motion was extracted by identical motion energy computations, although the second-order channel operated only at a coarse spatial scale. We investigated the response of the first- and second-order channels to stimuli where motion was defined by a range of attributes and compared the performance of the model with patient data from three recent clinical studies.Top of page
The model was sensitive to motion defined by first-order variations in luminance, and to second-order motion defined by contrast, spatial frequency and flicker. The ability of the model to account for data from recent neurological studies supports the theory that, at the lower levels of processing, two parallel channels underlie the detection and direction discrimination of first- and second-order motion.
b) We designed a novel psychophysical stimulus (a variant of that used by Boulton & Baker) to investigate the interaction of first- and second-order motion information in judgements of motion direction. The performance of normal subjects on the test was compared with the model predictions. The direction of two-frame apparent motion was measured for stimuli composed of Gabor or Gaussian micro-patterns. Three conditions were investigated. In condition 1, used by Boulton and Baker, motion was defined by the displacement of Gabor micro-patterns. In condition 2, motion was defined by the displacement of Gaussian micro-patterns and in condition 3, the envelopes of Gabor micro-patterns were displaced while their carriers remain static. The performance of normal subjects was compared with the predictions of the model for each condition. Using sparsely distributed micro-patterns, direction judgements in all three conditions were determined by the spacing of the micro-patterns. With a dense stimulus, direction judgements varied as a function of displacement in qualitatively different ways for the three conditions which were predicted by the two-channel computational model. In one channel, motion was calculated directly from stimulus luminance, while in the other it was preceded by a texture-grabbing operation. The relative activities of the two channels dictated which governed direction judgements for a given stimulus .
Top of page
- Vaina, L.M., Royden, C., Bienfang, D.C., Makris, N., and Kennedy, D. Normal perception of heading in a patient with impaired structure from motion, Invest. Opth. Vis. Sci., 1996; 37: 4890.
- Vaina, L.M., Burin des Roziers, E., and Belliveau, J.W. Cerebellar activation during direction discrimination of learning in visual motion stimuli: an fMRI study. Society of Neuroscience, 1997; 23: 1401.
- Beardsley, S.A. and Vaina, L.M. Computational modeling of optic flow selectivity in MSTd neurons, Investigative Ophthalmology & Visual Science , 1997; 38(4): S80.
- Vaina, L.M., Belliveau, J.W., Burin des Roziers, E., Zeffiro, T. The relationship between cortical activation and psychophysical performance in humans during fast learning of motion discrimination. NeuroImage, 1997; 5(4): S137.
- Clifford, C.W.G. and Vaina, L.M. A computational model of selective deficits in first- and second-order motion processing. Vision Research, 1999 (in press).
- Clifford, C.W.G. and Vaina, L.M. Anomalous Perception of Coherence and transparency in moving plaid Patterns: Evidence for Competing Processes. Cognitive Brain Research (accepted).
- Vaina, L.M. and Cowey, A. Selective deficits to first or second order order motion in stroke patients provides further evidence for separate mechanisms. Soc. Neurosci. Abstr., 1996; 22(1): 1718.
- Vaina, L.M, Makris, N., and Cowey, A. The neuroanatomical damage producing selective deficits of first or second order motion in stroke patients provides further evidence for separate mechanisms in Functional Mapping of the Human Brain, NeuroImage, 1996, 3(3): 360.
- Clifford, C.W.G. and Vaina, L.M. A computational model of selective deficits in first- and second- order motion perception, Investigat. Opthalmology & Visual Science 1998; 39(4): S1077.
last update: 7/21/00