Article Figures & Data
Figures
- Figure 1.
Schematic description of the experimental time course and multivoxel pattern analysis (MVPA). A, Experimental schedule showing the sequence of task and resting-state (RS) scanning sessions. B, Top, Example time course of a single trial within a task session, consisting of one task block followed by one replay block. Bottom, Illustration of multivoxel activity patterns during task and replay blocks within the task sessions. These were used as task and non-task templates to train the decoder. Activity patterns from three task sessions were used for training, and the decoder was then applied to pre-RS and post-RS data to test for the presence of task-related patterns during rest.
- Figure 2.
Behavioral results of tracking error. Average tracking error (in pixels) between the target and the cursor during the task sessions for the right-hand group (white dots) and the left-hand group (black dots). Error bars denote SEMs.
- Figure 3.
Activated areas during the task block identified via mass-univariate analysis. Brain regions showing significant activation during the task block compared with the replay block for the right-hand group (green) and the left-hand group (red). Overlapping activation between groups is shown in blue, based on a conjunction analysis (logical AND operation). Activation was reported with a threshold of p < 0.05, corrected for multiple comparisons using familywise error correction, with an extent threshold of 15 voxels. MNI coordinates of the activated regions are listed in Table 1. Results are displayed on horizontal brain slices, with Z denoting the slice location in MNI space. L, left hemisphere; R, right hemisphere.
- Figure 4.
Classification results of task and replay block activity across task sessions. Classification accuracy for distinguishing task and replay (observation-only) block activity across the three task sessions in the precentral cortex (left) and postcentral cortex (right) for each group. Red and yellow bars indicate results from the left and right hemispheres, respectively. Gray dots represent individual participant data. Black dotted lines denote the chance level of classification. Error bars denote SEMs. *p < 0.05; **p < 0.005.
- Figure 5.
Percentage of volumes labeled as task patterns during resting-state (RS) sessions. Percentage of volumes classified as task-related patterns in the precentral (A) and postcentral (B) cortices during the pre-task RS (pre-RS, blue) and post-task RS (post-RS, green) sessions for each group. Gray dots represent individual participant data. L, left hemisphere; R, right hemisphere. Error bars denote SEMs. ns, non-significant; +p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.005.
- Figure 6.
Difference in the percentage of volumes labeled as task patterns between pre-task and post-task rest. Difference in the percentage of volumes labeled as (A) or more similar to (B) the task pattern relative to the non-task (replay) pattern during pre-RS and post-RS in the precentral and postcentral cortices. Panel A shows results from the classification analysis; panel B shows results from the RSA. Red and yellow bars indicate the left and right hemispheres, respectively. Gray dots represent individual participant data. Error bars denote SEMs. +p < 0.10; * p < 0.05; **p < 0.01. Extended Data Figures 6-1 and 6-2 are supporting Figure 6.
- Figure 7.
Correlation between MVPA results and behavioral performance. Correlation between the increase in task-pattern labeling (post-RS minus pre-RS) and the reduction in tracking error from pre- to post-task performance. The increase in classification reflects the difference in the percentage of volumes labeled as task patterns in the contralateral precentral and postcentral cortices. L, left hemisphere; R, right hemisphere.
Tables
- Table 1.
Anatomical regions, peak voxel coordinates, and t values of observed activation during task blocks compared with replay blocks for each group
Anatomical region Voxels MNI coordinates t-value x y z Right-hand group R cerebellum 673 21 −49 −22 20.45 L precentral/postcentral gyrus 713 −30 −25 56 16.09 L thalamus 55 −12 −19 5 11.47 L cerebellum 63 −30 −58 −25 5.62 Left hand group L cerebellum 400 −18 −52 −22 16.35 R precentral/postcentral gyrus 845 36 −13 59 15.39 R thalamus 55 12 −19 5 15.08 L cerebellum 62 −21 −58 −46 12.43 R basal ganglia 38 21 −7 −1 5.94 R central operculum 22 42 −1 14 9.66 R cerebellum 15 30 −55 −22 5.30 Activation was reported at a threshold of p < 0.05, corrected for familywise error, with an extent threshold of 15 voxels. MNI, Montreal Neurological Institute; L, left hemisphere; R, right hemisphere.
Figure 6-1
Percentage of volumes labeled as task patterns during resting-state (RS) sessions using classifiers trained on individual task sessions Percentage of volumes in the precentral (A) and postcentral (B) cortices classified as task patterns versus non-task (replay) patterns during the pre-task (pre-RS, blue) and post-task (post-RS, green) RS sessions. Analyses were conducted in the hemisphere contralateral to the hand used. The classifier was independently trained on each of the three task sessions (sessions 1–3). L, left hemisphere; R, right hemisphere. Error bars denote SEMs. Download Figure 6-1, TIF file.
Figure 6-2
Example of MVPA-decoded labels in the left postcentral cortex of the right-hand group Frequency of spatiotemporal patterns labeled as task-related during rest, based on MVPA classification (A) and RSA (B). Black labels indicate volumes classified as (A) or more similar to (B) the task pattern compared with the non-task (replay) pattern. In each panel, the top two plots show the task-labeled volumes for each subject (top) and the percentage of task-labeled volumes across subjects (bottom) during the pre-RS session. The bottom two plots show the same data during the post-RS session. Download Figure 6-2, TIF file.
In this issue
