Content-specific activity in frontoparietal and default-mode networks during prior-guided visual perception
- National Institutes of Health, United States
- Ghent University, Belgium
- New York University Langone Medical Center, United States
Figures
Figure 1
Paradigm and behavioral results.
(A) Task design, and flow of events at trial, block, and fMRI run level. Subjects viewed gray-scale and Mooney images and were instructed to respond to the question ‘Can you recognize and name the …
Figure 2 with 4 supplements
Disambiguation-induced changes in neural activity magnitude and pattern.
(A) GLM contrast related to the disambiguation effect. For each subject, the set of Mooney images that were not recognized in the pre-disambiguation period and recognized in the post-disambiguation …
Figure 2—figure supplement 1
GLM results of the post-disambiguation identified >pre disambiguation not-identified contrast, based on the verbal identification responses.
For each subject, the same set of Mooney images are included in both conditions. Results were corrected for multiple comparisons using a cluster-defining threshold of z > 3.1, cluster size ≥17 …
Figure 2—figure supplement 2
Searchlight decoding of pre-disambiguation not-identified vs post-disambiguation identified Mooney images, selected based on verbal identification responses.
For each subject, the same set of images are included in both conditions. Results were corrected for multiple comparisons using a cluster-based permutation test and thresholded at a p<0.01, …
Figure 2—figure supplement 3
Regions of interest (ROIs) used in RSA.
(A) The FPN included the following ROIs: right frontal (14427 voxels, vx), left frontal (13383 vx), right parietal (3381 vx) and left parietal (3949 vx). The DMN included: MPFC (1135 vx), PCC (2145 …
Figure 2—figure supplement 4
The definition of FPN ROIs from image status decoding results (for details, see Materials and methods, ROI definition).
(A) Searchlight decoding of Mooney image status, reproduced from Figure 2C on inflated cortical surface. (B) The map from A is overlaid with the FPN and DMN ROIs used in this study. These ROIs are …
Figure 3 with 2 supplements
Neural representation format of individual images in visual regions.
(A) Analysis schematic. Dissimilarity (1 – Pearson’s r) between the neural response patterns to pairs of images was computed to construct the representational dissimilarity matrix (RDM) for each …
-
Figure 3—source data 1
RDM for each ROI in each subject. Includes source code to perform statistical analysis and produce Figure 3 and 4.
- https://doi.org/10.7554/eLife.36068.011
Figure 3—figure supplement 1
RSA results for left hemisphere visual regions.
Same as Figure 3B–E, but for left hemisphere ROIs.
Figure 3—figure supplement 2
RSA results for the right and left V4 (left and middle column, respectively), and the corresponding ROI size control analysis (right column).
The size of V4 ROIs (right: 256 vx; left: 198 vx) was much smaller than V1-V3 ROIs. Right column: We merged the left and right V4 ROIs for each subject, which yielded ROIs of approximately 500 …
Figure 4 with 3 supplements
Neural representation format of individual images in frontoparietal regions.
RSA results for FPN and DMN ROIs from the right hemisphere. Format is the same as in Figure 3. Interactive 3-dimensional MDS plots corresponding to first-order RDMs for each ROI can be found at: http…
Figure 4—figure supplement 1
RSA results for left hemisphere FPN and DMN regions.
Same as Figure 4, but for left hemisphere FPN and DMN ROIs, including left parietal and left frontal ROIs from FPN, and left lateral parietal ROI from DMN.
Figure 4—figure supplement 2
DMN results of the ROI size control analysis.
To assess whether ROI size affected dissimilarity, a control analysis was performed in a subset of regions, including DMN and V4 (V4 results presented in Figure 3—figure supplement 2). We …
Figure 4—figure supplement 3
Image category (natural vs.manmade) information.
(A) Analysis schematic. For details, see Materials and methods, RSA. (B–D) Category information in each ROI under each perceptual condition. Bars represent the difference in mean dissimilarity …
Figure 5
Relationship between neural representation format in different ROIs.
(A) Analysis schematic for second-order RSA across perceptual conditions (left) and within each condition (right). The RDMs from different ROIs were averaged across subjects. Then, a dissimilarity …
-
Figure 5—source data 1
RDM for each ROI in each subject. Includes source code to perform second-order RSA and reproduce Figure 5.
- https://doi.org/10.7554/eLife.36068.017
Figure 6
Dimensionality of neural representation space across ROIs and perceptual conditions.
The dimensionality of neural representation (estimated by the number of dimensions needed in the MDS solution to achieve r2 >0.9) for each ROI, in the pre-disambiguation (A), post-disambiguation (B),…
-
Figure 6—source data 1
RDM for each ROI in each subject. Includes source code to perform dimensionality analysis and reproduce Figure 6.
- https://doi.org/10.7554/eLife.36068.019
Figure 7
Significant preservation of neural code is found in visual and FPN regions.
(A) Analysis schematic. For each ROI in each subject, a representational similarity matrix (RSM) is constructed using the set of Mooney images that are not recognized in the pre-disambiguation …
-
Figure 7—source data 1
RSM (as shown in Figure 7A, left) for each ROI in each subject. Includes source code to perform analysis and reproduce Figure 7.
- https://doi.org/10.7554/eLife.36068.021
Author response image 1
Additional files
-
Transparent reporting form
- https://doi.org/10.7554/eLife.36068.022
