Anterior regions of monkey parietal cortex process visual 3D shape

Abstract

The intraparietal cortex is involved in the control of visually guided actions, like reach-to-grasp movements, which require extracting the 3D shape and position of objects from 2D retinal images. Using fMRI in behaving monkeys, we investigated the role of the intraparietal cortex in processing stereoscopic information for recovering the depth structure and the position in depth of objects. We found that while several areas (CIP, LIP, and AIP on the lateral bank; PIP and MIP on the medial bank) are activated by stereoscopic stimuli, AIP and an adjoining portion of LIP are sensitive only to depth structure. Furthermore, only these two regions are sensitive to both the depth structure and the 2D shape of small objects. These results indicate that extracting 3D spatial information from stereo involves several intraparietal areas, among which AIP and anterior LIP are more specifically engaged in extracting the 3D shape of objects.

Figure 1. IPS Regions Processing Stereoscopic Information in Experiment 1

(A) Examples of stimuli for the three conditions (3D structure, 3D position, and No disparity) and schematic representation of the corresponding stereo percepts. (B) Regions sensitive to structural and positional stereoscopic information are projected onto the flattened representations of left and right IPS, color coded in red and in yellow, respectively, with mixed sensitivity indicated in orange (p < 10⁻⁵ uncorrected, masked by all stimulus conditions versus fixation baseline at p < 10⁻³ uncorrected). Blue outlines delineate the regions sensitive to structural depth from motion (same statistical threshold). White dotted lines indicate the AIP/LIP borders derived from the saccade-related activity (same statistical threshold). Gray open symbols demarcate anatomically defined neighboring areas: Opt, PG, PFG, and PF on the inferior parietal lobule (Rozzi et al., 2006), and V6 and V6A on the parieto-occipital sulcus (Luppino et al., 2005).

Figure 2. Path Activity Profiles for the Lateral Bank Activations in Experiment 1

(A) Paths running along postero-anterior axes on the IPS lateral banks, passing through CIP, LIP, and AIP. Black symbols indicate the flat map segments constituting the paths (one per coronal level; from −14 to +2 mm). (B) MR signal change (± standard error of the mean) along the paths for the subtractions 3D structure – 3D position (red) and 3D position –No disparity (yellow). Filled symbols indicate the visually active segments for which MR signal change was significantly above zero (one-tailed t test), both for the group (p < 0.001) and for 2/3 of the individuals (p < 0.01). Note that, while results issued from path activity profiles and those given by SPM99 are very consistent with each other, slight differences can arise due to the use of unsmoothed data and additional statistical requirements (significance for 2/3 of the individuals) in the path activity profiles. (C and D) Same as (B) for the subtraction 3D motion – 2D motion in the 3D motion test (C), and for the subtraction Saccades – Visual control in the Saccade test (D). (n, number of monkeys).

Figure 3. Path Activity Profiles for the Medial Bank Activations in Experiment 1

(A) Paths running along ventro-dorsal axes on the IPS medial banks, from the fundus to the medial lips, and passing through PIP and MIP. Black symbols represent the flat map segments constituting the paths (in adjoining coronal segments at −11 mm, from the fundus to PIP, and at −10 mm, from MIP to the medial lip). (B–D) MR signal change (± SEM) along these paths for the subtractions 3D structure – 3D position (red) and 3D position – No disparity (yellow), in the first stereo experiment (B), for the subtraction 3D motion – 2D motion in the 3D motion test (C), and for the subtraction Saccades – Visual control in the Saccade test (D). Same conventions as Figure 2.

Figure 4. IPS Regions Sensitive to the Curvature and Orientation in Depth of Stereo Surfaces in Experiment 2

Examples of stimuli for the 3D curvature (A) and 3D orientation (B) conditions, and schematic representation of the corresponding stereo percepts. Several curvature and slant profiles specified by binocular disparity were used for testing 3D curvature and 3D orientation processing in interleaved runs. (C) Regions sensitive to curvatures in depth and positions in depth are projected onto the flattened representations of left and right IPS with the same conventions as Figure 1. Pink outlines delineate the regions sensitive to orientations in depth from stereo. (*sensitivity to structural depth, either curvature or orientation, was assessed by the interaction between disparity and depth order).

Figure 5. Path Activity Profiles for the Lateral Bank Activations in Experiment 2

(A) Same paths as Figure 2. (B) MR signal change (± SEM) along the paths for the subtractions 3D curvature –3D position (*after subtraction of their respective No disparity control), in red, and 3D position – No disparity, in yellow, in the 3D curvature test. (C) Same as (B) for the subtraction 3D orientation – 3D position, in pink, and 3D position – No disparity, in yellow, in the 3D orientation test. Same conventions as Figure 2.

Figure 6. IPS Regions Sensitive to 2D Shapes in Experiment 3

(A) Examples of grayscale images and line drawings of objects, presented with their respective scrambled control (Denys et al., 2004; Kourtzi and Kanwisher, 2000). (B) Example of a half-sized (from ~10° to ~5° on average) grayscale image and its scrambled version as used in the control experiment (overall size of the stimuli was identical: 15°). (C) 2D shape-sensitive regions projected onto the flattened representations of left and right IPS. Blue regions, whether pale or dark, indicate a significant main effect of scrambling for large 2D shapes. Dark blue regions exhibit in addition a significant interaction between scrambling and image type in the test with large shapes. Magenta outlines delineate the 2D shape-sensitive regions for the half-sized grayscale images. Same conventions as Figure 1.

Figure 7. Path Activity Profiles for the Lateral Bank Activations in Experiment 3

(A) Same paths as Figure 2. (B) MR signal change (± SEM) along the paths for the subtraction Intact – Scramble for the grayscale images (circles) and for the line drawings (squares) with large 2D shapes. Dark blue asterisks indicate significant difference (two-tailed t test) between these two subtractions, at the group level (p < 0.001) and for at least 2/3 of the individuals (p < 0.01). (C) Same as (B) for the subtraction Intact –Scramble in the test with small grayscale shapes. Black asterisks indicate a significant interaction in the 2-way ANOVA with stimulus size and scrambling as factors (p < 0.05 at the group level and for the two individuals, Bonferroni correction for the number of tests). Same conventions as Figure 2.