analysis.macaque-rdm-confidence.source-data-summary
Macaque RDM Confidence Wagering Report
Kiani-Shadlen macaque random-dot motion confidence slice generated from the Nature Communications figure source-data ZIP, with motion-strength summaries for accuracy and sure-target choice.
Source rows174,160
Accuracy rows100,694
Sure-target rows73,466
Subjects2
Strength levels7
Confidence rows13
Source
- Dataset
- dataset.khalvati-kiani-rao-rdm-confidence-source-data
- Protocol
- protocol.macaque-rdm-confidence-wager
- Article
- https://www.nature.com/articles/s41467-021-25419-4
- Source data
- https://static-content.springer.com/esm/art%3A10.1038%2Fs41467-021-25419-4/MediaObjects/41467_2021_25419_MOESM3_ESM.zip
- Source files
- ff04b39db1fae945b56e99d1240a5a7abd3d00690f8957542dd6535b0ace46b6
- Subjects
- kiani-shadlen-m1, kiani-shadlen-m2
Provenance
- Generated
- 2026-04-29T00:32:19.053973+00:00
- Commit
- 3ec74b4
- Git dirty
- true
- Code
- https://github.com/koosha66/POMDP-Confidence
Sure-Target Choice
Accuracy
Sure-Target Rows
| Monkey | Strength | Rows | Sure target | P(sure) | Median duration ms | Mean duration ms |
|---|---|---|---|---|---|---|
| M1 | 0 | 6,163 | 3,380 | 0.5484 | 229 | 272.5 |
| M1 | 1.6 | 6,137 | 3,303 | 0.5382 | 229 | 271.9 |
| M1 | 3.2 | 6,307 | 3,296 | 0.5226 | 217 | 269.4 |
| M1 | 6.4 | 6,109 | 3,034 | 0.4966 | 229 | 270.4 |
| M1 | 12.8 | 6,139 | 2,572 | 0.419 | 229 | 275.7 |
| M1 | 25.6 | 6,214 | 1,550 | 0.2494 | 229 | 273.6 |
| M1 | 51.2 | 6,082 | 620 | 0.1019 | 229 | 275.8 |
| M2 | 0 | 5,002 | 2,719 | 0.5436 | 242 | 289.4 |
| M2 | 3.2 | 5,032 | 2,533 | 0.5034 | 243 | 292 |
| M2 | 6.4 | 5,050 | 2,591 | 0.5131 | 243 | 291.6 |
| M2 | 12.8 | 5,155 | 2,341 | 0.4541 | 243 | 288.1 |
| M2 | 25.6 | 5,054 | 1,547 | 0.3061 | 243 | 288.4 |
| M2 | 51.2 | 5,022 | 729 | 0.1452 | 243 | 286.5 |
Accuracy Rows
| Measure | Monkey | Strength | Rows | Correct | P(correct) | Median duration ms | Mean duration ms |
|---|---|---|---|---|---|---|---|
| accuracy_no_sure_target | M1 | 1.6 | 6,225 | 3,313 | 0.5322 | 229 | 273.5 |
| accuracy_no_sure_target | M1 | 3.2 | 6,063 | 3,476 | 0.5733 | 229 | 271.4 |
| accuracy_no_sure_target | M1 | 6.4 | 6,250 | 3,982 | 0.6371 | 229 | 272.7 |
| accuracy_no_sure_target | M1 | 12.8 | 6,227 | 4,672 | 0.7503 | 229 | 272.3 |
| accuracy_no_sure_target | M1 | 25.6 | 6,158 | 5,654 | 0.9182 | 229 | 272 |
| accuracy_no_sure_target | M1 | 51.2 | 6,286 | 6,227 | 0.9906 | 229 | 276 |
| accuracy_no_sure_target | M2 | 3.2 | 5,091 | 2,682 | 0.5268 | 243 | 287.4 |
| accuracy_no_sure_target | M2 | 6.4 | 5,077 | 2,721 | 0.5359 | 242 | 290.3 |
| accuracy_no_sure_target | M2 | 12.8 | 4,980 | 3,151 | 0.6327 | 243 | 290.2 |
| accuracy_no_sure_target | M2 | 25.6 | 5,071 | 3,977 | 0.7843 | 242 | 290.4 |
| accuracy_no_sure_target | M2 | 51.2 | 5,081 | 4,693 | 0.9236 | 242 | 285.9 |
| accuracy_sure_available_direction_chosen | M1 | 1.6 | 2,834 | 1,567 | 0.5529 | 230 | 280.4 |
| accuracy_sure_available_direction_chosen | M1 | 3.2 | 3,011 | 1,878 | 0.6237 | 230 | 279.7 |
| accuracy_sure_available_direction_chosen | M1 | 6.4 | 3,075 | 2,143 | 0.6969 | 242 | 284 |
| accuracy_sure_available_direction_chosen | M1 | 12.8 | 3,567 | 3,002 | 0.8416 | 243 | 296.1 |
| accuracy_sure_available_direction_chosen | M1 | 25.6 | 4,664 | 4,497 | 0.9642 | 243 | 294 |
| accuracy_sure_available_direction_chosen | M1 | 51.2 | 5,462 | 5,442 | 0.9963 | 236 | 285.9 |
| accuracy_sure_available_direction_chosen | M2 | 3.2 | 2,499 | 1,360 | 0.5442 | 269 | 306.3 |
| accuracy_sure_available_direction_chosen | M2 | 6.4 | 2,459 | 1,399 | 0.5689 | 256 | 305.2 |
| accuracy_sure_available_direction_chosen | M2 | 12.8 | 2,814 | 1,935 | 0.6876 | 257 | 304.5 |
| accuracy_sure_available_direction_chosen | M2 | 25.6 | 3,507 | 2,937 | 0.8375 | 269 | 312 |
| accuracy_sure_available_direction_chosen | M2 | 51.2 | 4,293 | 4,071 | 0.9483 | 256 | 302.3 |
Generated Files
Caveats
- The Nature Communications ZIP provides source-data rows for figure panels, not the complete raw behavioral export. Rows should be interpreted as source-data records and may not be unique raw trial identifiers across panels.
- Motion direction and screen-side direction choice are not preserved in these CSVs. For rows where the source reports direction-choice correctness, the adapter exposes a target-coded choice proxy (right=correct direction choice, left=incorrect direction choice) so psychometric baselines can be fit without claiming raw target side recovery.
- Saccade response time is not preserved in these CSVs. response_time is populated from the source motion-duration column as a stimulus-duration proxy and is flagged in task_variables for model fits that need a chronometric axis.
- White-dot direction-choice accuracy rows are paired after dropping zero-strength rows, because direction correctness is not defined for zero coherence in the duration source table.