Adding mean insula high-gamma after movement increased mean error in linear confidence prediction across trial blocks in 14 SEEG patients

We tested whether mean insula high-gamma after first mouse movement improves confidence prediction. We used public MetaRDK SEEG data from OpenNeuro ds006253 version 1.0.3. A fixed linear base model contained behavior and mean insula high-gamma before movement. Both contacts of each selected bipolar channel had to share a left or right insula parcel. The sample had 14 patients, 93 selected channels, and 2,868 trials. Five contiguous trial blocks served as test sets. Adding the after-movement mean increased equal-patient mean relative squared error by 1.75%. The 95% patient bootstrap interval for this increase was 0.31% to 3.22%. Three patients improved. The exact two-sided sign-flip p value was 0.04114, under independent-patient and sign-symmetry assumptions. Correct-trial and fixed-penalty checks also had negative mean gains. This is a secondary analysis of the same cohort as AutoSci 041a699f-7342-4252-93c6-9f64132fe0fb. It does not establish an absence of confidence information in the insula. The package preserves anatomy decisions, compact data, source versions and hashes, fitted models, predictions, code, and reproduction instructions.

insula · high-gamma · confidence prediction · stereoelectroencephalography · SEEG · MetaRDK · OpenNeuro · linear models · cross-validation