Inhibitory motor control based on complex stopping goals relies on the same brain network as simple stopping
Introduction
The ability to stop ongoing behaviors after they have been initiated is a cognitive mechanism that is part of everyday life. Much research has used the stop-signal task (SST; Logan et al., 1984, Verbruggen and Logan, 2009) to investigate the factors that affect stopping, and how stopping is implemented in the brain. Stopping in the standard SST recruits an interconnected network of fronto-subcortical brain regions (the ‘stopping-network’) including the pre-supplementary motor area (pre-SMA), the right inferior frontal cortex (rIFC), and the basal-ganglia, with downstream effects on M1 (for reviews, see: Aron et al., 2014, Bari and Robbins, 2013, Chambers et al., 2009, Ridderinkhof et al., 2011, Stinear et al., 2009, Wiecki and Frank, 2013). Activity within this stopping network has been found across several brain imaging modalities. In the human scalp electroencephalogram (EEG), time–frequency analyses show a signature of successful action-stopping at fronto-central scalp sites, specifically within the theta- (5–8 Hz) and delta-frequency bands (1–4 Hz) (Lavallee et al., 2014, Nigbur et al., 2011, Schmiedt-Fehr and Basar-Eroglu, 2011, Wessel and Aron, 2013, Yamanaka and Yamamoto, 2010).
Yet it is important to ask whether this ‘stopping network’ for the standard SST generalizes to stopping in more realistic scenarios. Arguably, instances in which behaviors need to be canceled following explicit stop-signals (as in the standard SST) are relatively rare in the real world. Instead, stopping must be exerted in more complex situations such as the one given in the abstract, in which someone has to stop their step into the street when a car is bearing down. The stopping goal in that situation presumably consists of a complex template of features, which include the size of an object, its trajectory, velocity, and its distance. This stopping-template is presumably represented in working memory, and the stopping system is turned on if all or many features of a given situation match it.
Here, we developed a new behavioral paradigm that models action-stopping to more complex, realistic, stopping goals. In this task, participants had to quickly respond to arrow stimuli, just like in the standard SST. However, unlike the standard SST, we now used arrow-stimuli that differed perceptually along five different dimensions: color, position, number of arrows, arrow style, and print (outline or bold). Before every sequence of stimuli, a unique combination of these five features was presented to the participants as a ‘stopping-template’, which they had to maintain in memory. Participants then had to respond as quickly as possible to a sequence of arrow-stimuli, unless all five dimensions of the current stimulus matched the dimensions of the stopping-template. In that case, the action had to be stopped.
We hereafter refer to this new task as the ‘complex-stopping task’ (CST). Note that while the task is more akin to a go/nogo task (where the signal to nogo occurs at the same time as the go stimulus) than a classic stop-signal test (where the signal to stop occurs later than the go stimulus), our task is set up to also elicit a clear-cut stopping situation similar to the standard SST. This was done by creating a highly prepotent go-response on all trials, through having relatively few stop/nogo-trials, and by requiring relatively fast reaction times on go-trials. The prepotency of the go-response was measured by the number of failed stop/nogo-trials that is clearly attributable to failed motor inhibition (see below). Note also that this task is clearly more ecologically valid than the SST. This is because participants now have a more complex, multidimensional stopping goal in mind. As they are about to respond, they must match the features of the stimulus (a proxy for context) to their stopping goal. A partial match does not constitute a stopping scenario. This is similar to the situation in which a car is bearing down on a pedestrian with the correct trajectory to be potentially stopping-relevant, but is not moving fast enough to necessitate a stop. Of course, the CST is again a laboratory-based model of control that involves sequential trials with relatively simple stimuli, but it is clearly a closer model of realistic situations than the standard SST.
In a behavioral pilot (Experiment 1), we first established that the go response did have prepotency (similar to the standard SST): participants often failed to successfully stop, despite recognizing that stopping was needed. Interestingly, we further observed that partial matches between the go-stimulus and the stopping-template lead to slowed responding: when some (but not all) of the features of the go-stimulus matched the stopping template, go RT was increased. While the slowing could relate to many potential factors (Jahfari et al., 2010), we hypothesized that it could reflect partial recruitment of the stopping system, something we have referred to elsewhere as ‘braking’ (Swann et al., 2012b, Wessel et al., 2013).
In the main study (Experiment 2), we used EEG to test whether the observed stopping and ‘braking’ in the CST is subserved by the same motor inhibition network that explains stopping to explicit stop-signals in the standard SST. We recorded scalp EEG during the CST (the main task of interest) and also for the SST (which was used as a functional localizer for the stopping-system). We used independent component analysis (ICA, Jutten and Herault, 1991) to decompose each participant's observed scalp EEG signal mixture into its underlying temporally independent source signals (independent components, IC). As done previously (Wessel and Aron, 2013), we identified ICs in each subject that represented a typical EEG signature of successful stopping from the SST. We then tested whether this independent network showed increased activity during outright stopping and/or braking in the CST. We predicted that activity within the stopping-ICs identified in the SST should be increased following action-stopping in the CST (stopping hypothesis). Furthermore, if the RT slowing on partial feature match trials is explained by partial recruitment of the brain's motor inhibition network (i.e., ‘braking’), then the activity within the stopping-ICs should increase when partial matches induce increased RT slowing (braking hypothesis).
Section snippets
Experiment 1
17 right-handed participants (mean age: 21 y, sem: .37, range: 18–24; 12 female) performed the task in exchange for course credit. They provided written informed consent according to a local ethics protocol. Data from two participants were excluded, one due to high error rates (pressed wrong buttons on 46% of trials), and one due to high miss rates (did not respond to go-stimuli on 16% of trials), leaving a sample of 15 participants.
Experiment 2
11 right-handed participants (mean age: 20.9 y, sem: .87, range:
Experiment 1
Average RT for correct go-responses was 436 ms (sem: 8.1). Error- and miss rates were low (3.1% and 4.3%, respectively). Following MSTOP-stimuli, responses were withheld successfully 70.0% of the time (sem: 5.4). Of the remaining MSTOP-stimuli on which responses could not be withheld, 66.5% (sem: 7.8) were due to failed stopping (participants pressed the third button to indicate that they recognized that they were supposed to stop but failed to do so). An outlier contaminated this measurement,
Discussion
We designed a novel paradigm to investigate whether motor inhibition is subserved by the same brain network, regardless of whether stopping is triggered by an explicit stop signal (as in the standard SST) or by a more complex process. Here, that process involved matching the properties of a go-stimulus with the features of a multidimensional stopping-template. In a behavioral pilot (experiment 1), we found that this new task was successful in creating a race-like situation between a prepotent
Conclusions
We developed a novel paradigm that allowed us to compare stopping following simple, salient stop signals to stopping following a more complex matching process that involves a multidimensional stopping-goal. Using ICA of EEG data, we showed that the same brain network for successful stopping in the SST also explains stopping in the more complex CST. Furthermore, in the CST, we observed behavioral slowing on stimuli whose features only partially matched the stopping-template, which was also
Acknowledgments
This work was supported by grants from the Kavli Institute for Brain & Mind (2012-022) (to J.R.W.), the National Institutes of Health (R01DA026452) and the James S McDonnell Foundation (to A.R.A.).
References (69)
- et al.
Inhibition and the right inferior frontal cortex: one decade on
Trends Cogn. Sci.
(2014) - et al.
Inhibition and impulsivity: behavioral and neural basis of response control
Prog. Neurobiol.
(2013) - et al.
Neural correlates of blink suppression and the buildup of a natural bodily urge
NeuroImage
(2012) - et al.
Insights into the neural basis of response inhibition from cognitive and clinical neuroscience
Neurosci. Biobehav. Rev.
(2009) A neural microcircuit for cognitive conflict detection and signaling
Trends Neurosci.
(2014)- et al.
Engagement of large-scale networks is related to individual differences in inhibitory control
NeuroImage
(2010) - et al.
EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis
J. Neurosci. Methods
(2004) - et al.
Activation in ventral prefrontal cortex is sensitive to genetic vulnerability for attention-deficit hyperactivity disorder
Biol. Psychiatry
(2006) - et al.
Conflict and inhibition differentially affect the N200/P300 complex in a combined go/nogo and stop-signal task
NeuroImage
(2010) - et al.
“Shut up!” An electrophysiological study investigating the neural correlates of vocal inhibition
Neuropsychologia
(2012)
Real and fictive outcomes are processed differently but converge on a common adaptive mechanism
Neuron
Dissociable medial frontal negativities from a common monitoring system for self- and externally caused failure of goal achievement
NeuroImage
Electroencephalography of response inhibition tasks: functional networks and cognitive contributions
Int. J. Psychophysiol.
The development of stop-signal and go/nogo response inhibition in children aged 7–12 years: performance and event-related potential indices
Int. J. Psychophysiol.
Blind separation of sources. 1. An adaptive algorithm based on neuromimetic architecture
Signal Process.
Theta power as a marker for cognitive interference
Clin. Neurophysiol.
Imaging human EEG dynamics using independent component analysis
Neurosci. Biobehav. Rev.
Inhibitory control in children with attention-deficit/hyperactivity disorder: event-related potentials identify the processing component and timing of an impaired right-frontal response-inhibition mechanism
Biol. Psychiatry
The role of the subthalamic nucleus in response inhibition: evidence from local field potential recordings in the human subthalamic nucleus
NeuroImage
Rostral cingulate zone and correct response monitoring: ICA and source localization evidences for the unicity of correct- and error-negativities
NeuroImage
Mapping motor inhibition: conjunctive brain activations across different versions of go/no-go and stop tasks
NeuroImage
Current advances and pressing problems in studies of stopping
Curr. Opin. Neurobiol.
Electrophysiological activity underlying inhibitory control processes in normal adults
Neuropsychologia
Event-related delta and theta brain oscillations reflect age-related changes in both a general and a specific neuronal inhibitory mechanism
Clin. Neurophysiol.
Primary motor cortex and movement prevention: where stop meets go
Neurosci. Biobehav. Rev.
Roles for the pre-supplementary motor area and the right inferior frontal gyrus in stopping action: electrophysiological responses and functional and structural connectivity
NeuroImage
Models of response inhibition in the stop-signal and stop-change paradigms
Neurosci. Biobehav. Rev.
Attentional modulation of oscillatory activity in human visual cortex
NeuroImage
Cortical and subcortical contributions to stop signal response inhibition: role of the subthalamic nucleus
J. Neurosci.
Converging evidence for a fronto-basal-ganglia network for inhibitory control of action and cognition
J. Neurosci.
An information-maximization approach to blind separation and blind deconvolution
Neural Comput.
Controlling the false discovery rate — a practical and powerful approach to multiple testing
J. R. Stat. Soc. Ser. B Methodol.
In the trenches of real-world self-control: neural correlates of breaking the link between craving and smoking
Psychol. Sci.
Inhibitory control in mind and brain: an interactive race model of countermanding saccades
Psychol. Rev.
Cited by (41)
The effect of task complexity on the neural network for response inhibition: An ALE meta-analysis
2024, Neuroscience and Biobehavioral ReviewsTowards a unified neural mechanism for reactive adaptive behaviour
2021, Progress in NeurobiologyShocking advantage! Improving digital game performance using non-invasive brain stimulation
2021, International Journal of Human Computer StudiesDelta resting-state functional connectivity in the cognitive control network as a prognostic factor for maintaining abstinence: An eLORETA preliminary study
2021, Drug and Alcohol DependenceCitation Excerpt :Moreover, recent research supports a substantial link between delta synchronization and cognition, particularly when large-scale cortical networks, including the fronto-parietal network (FPN), coordinate their neural activity during cognitive tasks (Fleck et al., 2016; Gulbinaite et al., 2014; Imperatori et al., 2014; Liu et al., 2019; Nácher et al., 2013; Tóth et al., 2012). Finally, a number of studies have demonstrated that delta oscillations play a crucial and definite role in inhibitory control processes (Harper et al., 2016; Knyazev, 2007; Lavallee et al., 2014; Wessel and Aron, 2013, 2014) and reduced delta oscillatory responses was linked to neurofunctional deficits during response inhibition in alcoholism (Colrain et al., 2011; Kamarajan et al., 2004, 2006; López-Caneda et al., 2017; Pandey et al., 2016). Likewise, studies suggest that the inhibitory function of delta oscillations in cognitive processes is not limited only to inhibitory responses but could also be attributed more generally to suppression of interferences that may affect the cognitive performance (Dimitriadis et al., 2010; Harmony, 2013).
Proactive control without midfrontal control signals? The role of midfrontal oscillations in preparatory conflict adjustments
2019, Biological PsychologyCitation Excerpt :However, further research is needed to clarify the functional significance of preparatory beta modulations for inhibition and action switching that we observed here. For example, while the current study focused on the implementation of successful motor adjustments, a future experiment could compare successful with unsuccessful attempts to resolve motor conflicts (e.g. via the stop-signal paradigm, cf. Swann et al., 2009; Wessel & Aron, 2014), in order to test if the degree of proactive motor beta modulation predicts an increased likelihood of successfully adjusting one’s motor response during conflict tasks. Note that our analysis of the theta, beta, and alpha band activity was restricted to a-priori selected brain regions and is therefore not exhaustive concerning the function of these frequency bands for motor conflict tasks.