The role of conjunctive representations in prioritizing and selecting planned actions
- Department of Cognitive, Linguistic, and Psychological Sciences, Brown University, United States
- RIKEN Center for Brain Science, Japan
- Department of Psychology, University of Oregon, United States
- Carney Institute for Brain Science, Brown University, United States
Figures
Figure 1
Task design and the procedure of decoding analysis.
(A) Sequence of trial events in the rule-selection task with two independent action plans. Test probabilities of each action are assigned randomly every block. (B) Spatial translation of different …
Figure 2
The effect of priority of actions on behavior.
Average response times, error probability, swapping probability, and cutoff time (i.e. trial-to-trial response deadlines) in the low and high test probability conditions. Note that a swapping error …
Figure 3 with 1 supplement
Trajectories of decoded representations of two actions in the different levels of priority.
Average, single-trial t-values associated with each of the basic features (rule, stimulus, and response) and their conjunction derived from the representational similarity analysis (RSA), separately …
Figure 3—figure supplement 1
Decoded representations of two actions using frequency-specific scalp topography.
The representational similarity analysis scores of individual subjects using EEG signals in specific frequency ranges (1–3 Hz for the delta-band, 4–7 Hz for the theta-band, 8–12 Hz for the …
Figure 4 with 1 supplement
Modulation of action representations during selection using stimulus-aligned EEG.
Average, single-trial t-values associated with the stimulus, conjunction, and response derived from the representational similarity analysis (RSA), separately for high-priority (red) action and …
Figure 4—figure supplement 1
Modulation of action representations during selection using response-aligned EEG.
Average, single-trial t-values associated with the stimulus, conjunction, and response derived from the representational similarity analysis (RSA), separately for high-priority (red) action and …
Figure 5
Interference between tested and untested actions with different levels of priority.
The coefficients of the multilevel regression models predicting the variability in trial-to-trial RTs and errors. The model included representational similarity analysis scores of all action …
Author response image 1
Trial-to-trial variability between high and low priority conjunctions, using above median trials.
The coefficients of the multilevel regression model predicting the variability in trial-to-trial highpriority conjunction by low-priority conjunction.
Author response image 2
The number of retained trials of each S-R mapping condition.
The box plot shows the count of number of trials retained after artifact rejection. The number label of each S-R mapping corresponds to Figure 1C in the main text.
Author response image 3
Decoded representations of two actions using frequency-specific scalptopography.
The RSA scores of individual subjects using EEG signals in specific frequency ranges (1-3 Hz for the δ-band, 4-7 Hz for the theta-band, 8-12 Hz for the α- band, 13-30 for the β-band, 31-35 Hz for …
Tables
Table 1
Trial-by-trial representational similarity analysis scores of high- and low-priority actions in each test probability context during the test phase.
| Decoded features | Tested high priority | Tested low priority | ||
|---|---|---|---|---|
| beta (CI) | t-Value | beta (CI) | t-Value | |
| Rule | 0.071 (.05.09) | 6.08*** | 0.069 (.04.09) | 4.76** |
| Stimulus (high priority) | 0.034 (.01.05) | 3.41* | 0.021 (.01.03) | 2.9* |
| Conjunction (high priority) | 0.021 (.01.04) | 3.57* | 0.007 (–.01.02) | 0.86 |
| Response (high priority) | 0.034 (.02.05) | 5.04** | 0.009 (–.01.03) | 1.04 |
| Stimulus (low priority) | 0.011 (.01.02) | 2.20* | 0.014 (–.01.03) | 1.30 |
| Conjunction (low priority) | 0.003 (–.01.01) | 0.61 | 0.001(–.01.02) | 0.12 |
| Response (low priority) | 0.008 (–.01.02) | 1.61 | 0.015(–.01.03) | 1.82 |
-
Note: *, **, ***, and a dot indicate p<0.05, p<0.01, p<.001.
Table 2
Cluster-level statistics for the main effects of the interaction model between test type and priority regressed on the representational similarity analysis (RSA) scores.
| RSA scores | Effect | Time (ms) | Cluster T-value | Cluster p-value |
|---|---|---|---|---|
| Stimulus | Priority | 1566–1598 | 20.11 | <0.01 |
| Priority | 1846–1866 | 12.45 | <0.01 | |
| Conjunction | Test type | 1898–2050 | 98.36 | <0.01 |
| Response | Priority | 1738–1794 | 38.70 | <0.01 |
| Test type | 1858–1894 | 22.94 | <0.01 | |
| Test type | 2014–2110 | 67.02 | <0.01 |
-
Note: The clusters were identified from the model that included the effect of priority and test type and their interaction using a cluster-forming threshold, p<0.05. The interaction effect is shown in Figure 4.
