Evidence for parallel activation of the pre-supplementary motor area and inferior frontal cortex during response inhibition: a combined MEG and TMS study

Abstract

This pre-registered experiment sought to uncover the temporal relationship between the inferior frontal cortex (IFC) and the pre-supplementary motor area (pre-SMA) during stopping of an ongoing action. Both regions have previously been highlighted as being central to cognitive control of actions, particularly response inhibition. Here we tested which area is activated first during the stopping process using magnetoencephalography, before assessing the relative chronometry of each region using functionally localized transcranial magnetic stimulation. Both lines of evidence pointed towards simultaneous activity across both regions, suggesting that parallel, mutually interdependent processing may form the cortical basis of stopping. Additional exploratory analysis, however, provided weak evidence in support of previous suggestions that the pre-SMA may provide an ongoing drive of activity to the IFC.

Keywords: inferior frontal cortex; magnetoencephalography; pre-supplementary motor area; response inhibition; timing; transcranial magnetic stimulation.

Figure 1.

Illustration of the progression of a single trial where a ‘Stop’ signal is presented at 200 ms post stimulus onset. In this task, participants are asked to respond to the white arrow as quickly as possible with a directional button press, and to inhibit their response if a ‘Stop’ signal occurs (contrast reversal). Stimulus dimensions are indicated in degrees of visual angle. The delay between the onset of the ‘Go’ stimulus and the ‘Stop’ signal (the SSD) was dynamically altered to account for individual differences in performance. Also illustrated is the tROI where it is possible for regions under investigation to exert influence in response to the ‘Stop’ signal. TMS was applied to the pre-SMA and the IFC at 12.5% and 62.5% of each participant's tROI. An approximate stop signal reaction time (approx SSRT) is also shown for illustrative purposes.

Figure 2.

Example of a single participant's activity traces. The time courses of the events during trials are shown, including the end of the fixation period, the ‘Go’ and ‘Stop’ signal onsets, the participant-specific tROI and their mean Go reaction time (GoRT). The thicker dashed lines indicate the z-score criterion level, applied to each activity trace where the time point at which the trace crosses the criterion value is the latency-dependent measure. Also depicted are the mean and maximum level values used in the calculation of the criterion levels, where the absolute criteria levels were incremented in step sizes of 0.5 s.d. until the absolute mean was exceeded and the absolute maximum was at least 0.5 s.d. greater.

Figure 3.

Example of the frequency restriction applied to single participant and aROI γ spectral data. Oscillatory change across a broad frequency range (e.g. 31–130 Hz) was collapsed across the tROI to produce a spectral response as depicted by the blue line. The red line indicates the frequency region resulting from a selection process (range selected is the maximum of a multiplication of the adjusted R² values by the amplitude of the deflection). The fit is represented by the red crosses. The red region, therefore, represents the restricted frequency range used in subsequent analyses applied to this participant's aROI γ response.

Figure 4.

Group evoked (a,b) and induced α (c,d), β (e,f) and γ (g,h) activity traces during stopping. Main traces (a,c,e,g) are based on the (Stop–Go)–Fixation contrasts averaged across participants, following exclusions, from −1.3 to 1.3 s relative to the onset of the stimulus, and bandwidths applied in the induced traces are specified in §2.3.5. Insets (b,d,f,h) correspond to activity during the group tROI where individual participant data have been normalized to cover the group mean duration. Thick lines represent means across participants and thin lines are ±1 s.e. Red crosses indicate the mean point of deflection used in the temporal comparisons, for which the Bayesian comparison revealed concurrent activity across aROIs. Dots indicate the tROI from first onset of the stop signal, to the start and then the end of the critical tROI.

Figure 5.

Behavioural effects of pre-SMA and IFC stimulation on response inhibition and response execution. Bars indicate group mean SSRT(a), proportion stopped (b) and ‘Go’ RT (c) under each of the TMS conditions (sham and active, as well as where TMS is applied to the pre-SMA before the IFC and where TMS is applied to the IFC before the pre-SMA). Individual data points are shown to the left of within-participant standard error bars [63].