Triangulating a cognitive control network using diffusion-weighted magnetic resonance imaging (MRI) and functional MRI

Abstract

The ability to stop motor responses depends critically on the right inferior frontal cortex (IFC) and also engages a midbrain region consistent with the subthalamic nucleus (STN). Here we used diffusion-weighted imaging (DWI) tractography to show that the IFC and the STN region are connected via a white matter tract, which could underlie a "hyperdirect" pathway for basal ganglia control. Using a novel method of "triangulation" analysis of tractography data, we also found that both the IFC and the STN region are connected with the presupplementary motor area (preSMA). We hypothesized that the preSMA could play a conflict detection/resolution role within a network between the preSMA, the IFC, and the STN region. A second experiment tested this idea with functional magnetic resonance imaging (fMRI) using a conditional stop-signal paradigm, enabling examination of behavioral and neural signatures of conflict-induced slowing. The preSMA, IFC, and STN region were significantly activated the greater the conflict-induced slowing. Activation corresponded strongly with spatial foci predicted by the DWI tract analysis, as well as with foci activated by complete response inhibition. The results illustrate how tractography can reveal connections that are verifiable with fMRI. The results also demonstrate a three-way functional-anatomical network in the right hemisphere that could either brake or completely stop responses.

Figures

Figure 1.

Diffusion-weighted tractography results. A, 3-D rendering of the tracts between the right IFC, the right preSMA, and the right STN region. B, Triangulation method for determining the third point in a network from the other two. Tracts originating in one brain area are overlaid on tracts originating from another. The overlap is superimposed on a gray matter mask in standard space. Tracts clearly overlap in the white matter space, but the overlap in gray matter is fairly unique: the preSMA only for tracts originating in the IFC and STN regions; the IFC and anterior prefrontal cortex (not shown) for tracts originating in the preSMA and the STN region and the thalamus only for tracts originating in the preSMA and the IFC.

Figure 2.

The conditional stop-signal paradigm. A, Critical and noncritical Stop trials. On Go trials, the subject has 1 s (the hold period) to press a left or right button in response to a stimulus, using the index or middle finger of the right hand. On a Stop trial, a tone is played at some delay (SSD) after the arrow stimulus. The SSD changes dynamically throughout the experiment. Subjects are instructed in advance whether the critical finger is the index or middle one. If the arrow stimulus is in the critical direction and a tone occurs, then the subject must try to inhibit, but if the arrow is in the noncritical direction and a tone occurs, then the subject must respond anyway. B, The different trial types with descriptions. L, Left; R, right.

Figure 3.

Conflict-induced slowing. A, Regions of significant activation are shown for the contrast of StopRespond − Go (in the noncritical direction) (Z > 2.3; p < 0.05, corrected for multiple comparisons at the whole-brain level). There is significant activation in a predominantly right-lateralized network including the preSMA/anterior cingulate, the IFC, the right middle frontal gyrus, the right STN region (shown overlaid on anatomical location of the STN ROI; see Materials and Methods), the auditory cortex (stop-signal tone), and the right caudate. B, Parametric analysis of brain activation with increasing RT. On noncritical trials, separate regressors were created for trials with a stop signal (StopRespond) and trials without (Go), each parametrically modulated by RT. Shown is the contrast between the following: StopRespond_parametric − Go_parametric (Z >2.3; p < 0.05, whole-brain corrected). Activations were found in preSMA, IFC, and STN regions. Superimposed on the activation map are the triangulation method tract overlaps. Green, The preSMA is the gray matter overlap for tracts originating in the IFC and STN region; blue, the IFC is the overlap for tracts originating in the preSMA and the STN region; red, the STN is the overlap for tracts originating in the preSMA and IFC. C, Confirmatory analysis shows increases of activation with slowing for stop-signal but not for no-signal trials on noncritical trials. The mean activation in each of the IFC, STN, and preSMA regions (anatomically defined; see Materials and Methods) is shown for each of StopRespond and Go for each of three RT bins (fast, medium, and slow, defined on a subject-by-subject basis). D, Conjunction analysis between conflict-induced slowing and outright inhibition. This image shows loci where the parametric increase in activation with increasing RT (B) overlaps with loci activated by outright inhibition (Fig. 4A), in which each individual group map was itself corrected for multiple comparisons. In the bottom right panel, the anatomical locus of the STN (Lucerna et al., 2002) is indicated with a blue region of interest.

Figure 4.

Outright response inhibition: relationships between activation, tracts, and SSRT. A, Regions of significant activation are shown for the contrast of StopInhibit − Go (in the critical direction) (Z > 2.3; p < 0.05, corrected for multiple comparisons at the whole-brain level). Right lateral and medial views, along with slices, show activation in a mainly right-lateralized network including the preSMA/anterior cingulate, the IFC, the STN region, the auditory cortex (auditory ctx; stop-signal tone), and the parietal cortex (parietal ctx). On the slices, blue regions at the IFC and the STN region denote foci where there was a significant correlation between SSRT and activation (small-volume correction for anatomically defined IFC and STN regions). The white matter tract connecting the IFC and the STN region is overlayed in red (see also top right inset). B, Mean peristimulus time courses (and SEs) extracted from each of the right preSMA, IFC, and STN regions for each event type. C, Subjects who inhibit more quickly activate the right IFC and the STN region more. SSRT was regressed against StopInhibit − Go activation in each of right IFC, preSMA and STN regions, anatomically defined, and the results were corrected for multiple comparisons. Activation was significant in the right IFC (MNI: 44, 22, 12; Z = 3.84; p < 0.05, small-volume correction) and the STN region (MNI: 14, −8, −4; Z = 3.02; p < 0.05, small-volume correction) but not in the preSMA. Mean activation was extracted from a sphere (radius, 3 mm) centered on the peak voxel. Activation of the right IFC and the right STN region was also significantly correlated across subjects (r = 0.53; p = 0.04).