非表示:
キーワード:
-
要旨:
Interactions in our dynamic environment require that the visual system processes both the shape
and motion of objects. Different cortical areas have been proposed to be involved in the processing
of visual motion (hMT+/V5 = human middle temporal homologue), kinetic boundaries
(KO = kinetic occipital) and object shape (LOC = lateral occipital complex). The goal of this
study was to investigate whether these areas are involved in the perception of shapes dened
by motion coherence. To this end, we used human event-related fMRI and employed stimuli
in which the shape was dened by the relative motion of random dots in the shape and the
background. We manipulated the perception of these shapes by independently varying the motion
coherence of the dots in the shape and the background. Increased motion coherence in
either the shape or the background improved the behavioral performance of the observers in a
shape categorization task. FMRI responses in the LOC and KO were consistent with the behavioral
performance; that is, enhanced fMRI responses were observed with increased motion
coherence in either the shape or the background. Interestingly, hMT+/V5 showed activation
patterns similar to the LOC, suggesting strong interactions between ventral (LOC) and dorsal
(hMT+/V5) visual areas in the perception of shape from motion. To further investigate shape
representations from motion in the different visual areas we tested for selectivity for shape and
motion direction information. To this end, we used an fMRI adaptation paradigm in which
lower fMRI responses are observed for two identical than two different stimuli presented consecutively
in a trial. Recovery from adaptation was observed across changes in shape in the
LOC, KO, and hMT+/V5, but not in early visual areas. In a third study, we tested whether
shape and motion related areas are selective for the 3D structure of shapes. To this end, we
employed an fMRI adaptation paradigm and similar stimuli as in the two previous studies. In
addition, these stimuli were rendered with 3D structure using horizontal disparity cues. Recovery
from adaptation was observed across changes in 3D structure in the LOC, KO, and
hMT+/V5. In summary, these ndings suggest that not only object (LOC) but also motionrelated
areas (hMT+/V5) are involved in the selective representation and perception of shapes
dened by coherent motion.
