(A) Task-bracketing index averaged over stages on trials with (empty bars) or without (solid bars) deliberation, for DLS (left) and ILs (right) units. **p < 0.01.
(B) DLS task-bracketing index for trials with (dotted) and without (solid) deliberation across stage blocks. *p < 0.05.
(C) Normalized activity (baseline-subtracted spiking) around run start, instruction cue, and run end for each trial type and site. *p < 0.05.
See also Figure S5.