Figure 3.

(A) SPM{t} map thresholded at FDR P < 0.05 for choose versus specified rule trials in experiment 1, averaging across color and height modalities and types of second cue (see Table 1) (B) SPM{t} map thresholded at FDR P < 0.05 for action-selection versus specification in experiment 1 (choose action vs. specified action trials in Table 1) (C) SPM{t} emphasizing that rule-selection in (A) and action-selection in (B) did not differ from each other by t-tests at either FDR P < 0.05 or at the liberal threshold of P < 0.01 uncorrected. (D) SPM{t} at uncorrected threshold P < 0.001 for rule-selection versus specification from experiment 2, averaging across all types of second cue (see Table 1). Correction to P < 0.05 within a frontal ROI is indicated in Table 2. (E) SPM{t} FDR P < 0.05 for action-selection versus specification from experiment 3 (cf. choose action vs. specified action trials in Table 1) (F) SPM{t} FDR P < 0.05 for the application of a current rule to the selection among 2 or 4 targets versus a single target in experiment 1 (apply rule vs. irrelevant to specified response in Table 1). Note the absence of prefrontal cortical activations in comparison to free-selection of actions in (B). The SPM{t}s are rendered on the SPM5 canonical T₁ brain volume in MNI space. Yellow lines are overlaid to assist comparison of the dorsal- and ventral-prefrontal activations in (A), (B), (D), and (E). The sagittal slice is at x = −5 mm.