Neural representation of newly instructed rule identities during early implementation trials
- Technische Universität Dresden, Germany
- Max-Planck-Institute for Human Cognitive and Brain Sciences, Germany
Figures
Figure 1
Stimulus-response (S–R) learning task used in experiment 1 exemplarily depicted for one of 18 blocks per condition (easy and difficult).
Each block consisted of an instruction phase and an implementation phase. During the instruction phase participants were presented with 4 (easy instruction) or 10 (difficult instruction) pairings between disyllabic nouns and manual responses. The vertical bars framing the nouns indicated the correct response (e.g. Bottle - left). During the subsequent implementation phase (here, selectively shown for the easy condition), each nouns was presented 4 times in random order without the vertical bars and participants had to respond as instructed. Irrespective of S-R rule difficulty (4 vs. 10 nouns in the instruction phase), a constant number of 4 different nouns was presented in the implementation phase. At the end of each block, feedback specifying the percentage of correctly answered trials was displayed.
Figure 2
Stimulus-response (S–R) learning task used in Experiment 2 exemplarily depicted for one of 12 blocks per condition (Intentional learning vs. control).
As in experiment 1, each block consisted of an instruction phase and an implementation phase. The Intentional learning condition was identical to the easy condition of experiment 1 (i.e., four instructed S-R rules) except that each S-R rule needed to be implemented 8 times instead of 4 times. In the control condition the response cues (i.e. the vertical bars) were omitted during the instruction phase and were instead presented during the subsequent implementation phase. At the end of each block, feedback specifying the percentage of correctly answered trials was displayed.
Figure 3
Schematic illustration depicting how identity-specific multi-voxel pattern similarity was computed exemplarily for one implementation stage in one learning block.
For illustrative purposes, only two stimuli (S1 and S2) each occurring twice are considered here (instead of 4 stimuli in reality). Bottom left: For each stimulus occurrence voxel-wise beta estimates (visualized by grayscale values) are arranged in vectors that constitute the basis of multi-voxel pattern correlations. Bottom right: matrix values depict multi-voxel pattern correlations for all combinations of trials. Green cells denote correlations between same stimuli, orange cells denote correlations between different stimuli. Top right: Identity-specific pattern similarity is defined by significantly greater mean correlations in green cells compared to orange cells.
Figure 4
Behavioral performance data for experiment 1 and experiment 2.
Error bars represent 90% confidence intervals.
Figure 5
Summary of the ROI-based MVPA results for experiments 1 and 2.
Error bars represent 90% confidence intervals. Significant differences are indicated by asterisks. (A) Identity-specific pattern similarities in experiment 1 collapsed across implementation stages. (B) Identity-specific pattern similarities in experiment 2 collapsed across implementation stages. (C) Identity-specific pattern similarities collapsed across experiments 1 and 2 broken by implementation stages. Early implementation stage pattern similarities are based on stimulus repetitions 1 and 2 whereas late implementation stage pattern similarities are based on stimulus repetitions 3 and 4. (D) Identity-specific pattern similarities for experiment 2 broken by implementation stages. Early implementation stage pattern similarities are based on aggregated values for stimulus repetitions 1 and 2 and stimulus repetitions 3 and 4. Late implementation stage pattern similarities are based on aggregated values for stimulus repetitions 3 and 4 and stimulus repetitions 7 and 8.
Figure 6
Results of the whole-brain searchlight MVPA testing for overall identity-specific pattern similarity effects.
(A) Horizontal brain slices depicting the findings for the left sensorimotor cortex, the ventro-lateral PFC, and the visual cortex. For display purposes the map shows voxels with p<0.001 uncorrected. (B) Pattern-similarity effects broken by instruction difficulty (exp. 1) or instruction type (exp. 2). In addition to sensorimotor cortices and visual cortices, the white-matter volume is included as a control region to highlight the absence of analysis bias. For a comprehensive summary of ventro-lateral PFC results see Figure 5. Error bars represent 90% confidence intervals.
Figure 7
Summary of the ROI-based mean activity results for experiments 1 and 2.
Error bars represent 90% confidence intervals.
Figure 8
Summary of the functional connectivity analysis results for the left VLPFC seed region based on single-trial beta-series correlations.
The analysis tested for a functional connectivity increase from early implementation trials (stimulus repetitions 1 and 2) to late implementation trials (stimulus repetitions 7 and 8) which was stronger for intentional learning blocks than control blocks. (A) Visualization of the significant effect in the anterior striatum. For display purposes the map shows voxels with p<0.001 uncorrected. (B) The detailed connectivity pattern for the anterior striatum cluster. (C) Mean activations at each stimulus repetition level based on a conventional univariate analysis for the anterior striatum cluster. Error bars represent 90% confidence intervals.
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Neural representation of newly instructed rule identities during early implementation trials
eLife 8:e48293.
https://doi.org/10.7554/eLife.48293
