Research Articles, Behavioral/Cognitive

Proactive Interhemispheric Disinhibition Supports Response Preparation during Selective Stopping

Corey G. Wadsley, John Cirillo, Arne Nieuwenhuys and Winston D. Byblow
Journal of Neuroscience 8 February 2023, 43 (6) 1008-1017; DOI: https://doi.org/10.1523/JNEUROSCI.1712-22.2022

Abstract

Response inhibition is essential for terminating inappropriate actions. A substantial response delay may occur in the nonstopped effector when only part of a multieffector action is terminated. This stopping-interference effect has been attributed to nonselective response inhibition processes and can be reduced with proactive cuing. This study aimed to elucidate the role of interhemispheric primary motor cortex (M1–M1) influences during selective stopping with proactive cuing. We hypothesized that stopping-interference would be reduced as stopping certainty increased because of proactive recruitment of interhemispheric facilitation or inhibition when cued to respond or stop, respectively. Twenty-three healthy human participants of either sex performed a bimanual anticipatory response inhibition paradigm with cues signaling the likelihood of a stop-signal occurring. Dual-coil transcranial magnetic stimulation was used to determine corticomotor excitability (CME), interhemispheric inhibition (IHI), and interhemispheric facilitation (IHF) in the left hand at rest and during response preparation. Response times slowed and stopping-interference decreased with increased stopping certainty. Proactive response inhibition was marked by a reduced rate of rise and faster cancel time in electromyographical bursts during stopping. There was a nonselective release of IHI but not CME from rest to in-task response preparation, whereas IHF was not observed in either context. An effector-specific reduction in CME but no reinstatement of IHI was observed when the left hand was cued to stop. These findings indicate that stopping speed and selectivity are better with proactive cueing and that interhemispheric M1–M1 channels modulate inhibitory tone during response preparation to support going but not proactive response inhibition.

SIGNIFICANCE STATEMENT Response inhibition is essential for terminating inappropriate actions and, in some cases, may be required for only part of a multieffector action. The present study examined interhemispheric influences between the primary motor cortices during selective stopping with proactive cuing. Stopping selectivity was greater with increased stopping certainty and was marked by proactive adjustments to the hand cued to stop and hand cued to respond separately. Inhibitory interhemispheric influences were released during response preparation but were not directly involved in proactive response inhibition. These findings indicate that between-hand stopping can be selective with proactive cuing, but cue-related improvements are unlikely to reflect the advance engagement of interhemispheric influences between primary motor cortices.

Introduction

Actions often require the coordination of multiple effectors. For example, steering, accelerating, and braking require precise coordination while driving. Response-selective stopping (hereafter referred to as selective stopping) refers to scenarios where only a subcomponent of a multicomponent action must be terminated (Wadsley et al., 2022b). Terminating an inappropriate action, response inhibition, supports selective stopping. However, a substantial delay occurs in the response of nonstopped effectors during selective stopping (Coxon et al., 2007; Aron and Verbruggen, 2008). This stopping-interference effect is a demonstrable constraint of selective stopping, which may arise from fast-acting global response inhibition (Raud et al., 2020; Wadsley et al., 2022a).

Selective stopping can be required with or without foreknowledge. For example, suddenly needing to cancel a lane change is more likely on a busy motorway than on a quiet street. Proactive response inhibition reflects inhibitory processes that occur in anticipation of a need to stop (Verbruggen and Logan, 2009). The influence of foreknowledge during selective stopping can be assessed with uninformative (reactive) and informative (proactive) warning cues (Raud and Huster, 2017; Cirillo et al., 2018). Informative cues allow for stopping to be prepared for a specified subcomponent, and others can respond with certainty (e.g., maybe stop-left cue), whereas the forthcoming stopping requirement is ambiguous during uninformative cues (e.g., maybe stop either cue). In the present study, we hypothesized that stopping would occur more quickly with increased stopping certainty, because of proactive response inhibition.

The stopping-interference effect is reduced but not abolished with informative cues. Greater selectivity is driven, in part, by proactive response inhibition during action preparation (Majid et al., 2012). Behaviorally, response times are slowed in effectors cued to stop. Transcranial magnetic stimulation (TMS) of primary motor cortex (M1) showed a concomitant suppression of corticomotor excitability (CME) during behavioral slowing (Cai et al., 2011; Majid et al., 2012). Paired-pulse TMS studies have shown a decrease in GABA receptor-mediated intracortical inhibition toward effectors more likely to respond (Cowie et al., 2016; Cirillo et al., 2018). Thus, improvements in selective stopping are driven through proactive control over M1 output during action preparation. Therefore, we hypothesized that stopping selectivity would improve with increased stopping certainty in the present study.

Interhemispheric M1–M1 influences support action preparation and can be investigated with dual-coil TMS. Interhemispheric inhibition (IHI) is produced through transcallosal activation of inhibitory interneurons and is mediated by GABA-B postsynaptic receptors at long interstimulus intervals (Daskalakis et al., 2002; Irlbacher et al., 2007). Interhemispheric facilitation (IHF) is mediated by corticocortical interactions when probed at short (4–8 ms) interstimulus intervals (Bäumer et al., 2006; Koch et al., 2006). Modulation of IHF and IHI occurs during action preparation and is influenced by task dynamics (Neige et al., 2021). In the present study, we hypothesized that IHI would decrease, and IHF would increase from rest to in-task contexts because of between-hemisphere sensorimotor disinhibition during response preparation.

It is unclear how facilitatory and inhibitory M1–M1 mechanisms influence proactive response inhibition. An upregulation of GABA-B receptor-mediated inhibition examined within M1 indicates a general role in setting an inhibitory tone that can influence the selectivity of stopping (Cowie et al., 2016; Cirillo et al., 2018). IHI is increased in the stopping effector and released in the responding effector during reactive selective stopping (MacDonald et al., 2021). Although IHF may contribute to action reprogramming (Mars et al., 2009), it remains to be determined whether IHF plays a role in selective stopping. We hypothesized that during informative selective stopping, IHF and IHI would be upregulated in the responding and stopping hand, respectively, because of cue-dependent modulation for proactive response inhibition.

In summary, the present study uses dual-coil TMS positioned over bilateral M1s to investigate M1–M1 interhemispheric influences for selective stopping with and without proactive cuing in a bimanual anticipatory response inhibition (ARI) paradigm. Additional measures were obtained from electromyography to determine the specificity of proactive response inhibition to the stopping and responding hand during selective stopping.

Materials and Methods

Participants

Twenty-four healthy adults volunteered to participate. One participant was excluded because of left-handedness. The remaining 23 participants (13 female and 10 male, mean age 27.2 years, range 22–39 years) were all right-handed (mean laterality quotient 0.83, range 0.25–1; Veale, 2014). The target sample size was selected based on similar dual-coil TMS studies that have investigated selective stopping (MacDonald et al., 2021). The study was approved by the University of Auckland Human Participants Ethics Committee (Ref. UAHPEC22709).

Task protocol

A multicomponent ARI paradigm was used to assess selective stopping (Wadsley et al., 2022b). The task was programmed in PsychoPy, version 2020.2.4 (Peirce et al., 2019), and interfaced with a custom Arduino Leonardo response board. Participants were seated comfortably in front of an LG 24GL600F-B monitor (144 Hz refresh rate, ∼60 cm viewing distance). The left and right index fingers rested on blocks with mechanical switches positioned ∼1 cm above so that responses required sagittal index finger abduction (Fig. 1A). Switch height was adjusted to minimize postural muscle activity observed from electromyography (EMG) of first dorsal interosseous (FDI) muscles bilaterally. The display consisted of two white bars (15 cm high, 1.5 cm wide) on a gray background. A black horizontal target line was positioned behind each bar at 80% of its total height. Trial onset occurred when the bars appeared to fill (i.e., gradually turning black from bottom to top). Pressing the left or right switch during a trial (1.5 s) would cause the corresponding bar to cease filling.

Figure 1.
Figure 1.

Bimanual anticipatory response inhibition paradigm. A, Responses were made using the left and right index fingers. The TS and CS coils targeted the left and right hand, respectively. B, Timeline from trial onset. All trials began with two empty bars. The objective during GG trials was to press the switches when the indicators reached the target (800 ms). During partial-stop trials, either the left (SG) or right (GS) bar automatically stopped filling before the target, thus requiring the response in the corresponding hand to be cancelled. C, Cues were presented before trial onset to assess responses in rest, certain-go, uninformative (unif.), low-informative (low-inf.), and high-informative (high-inf.) contexts. In-task stimulation was delivered 550 ms from trial onset (TMS). Stop cues were equally distributed across left and right (stop-left informative cues depicted in the figure).

The objective of most trials was to cease the bars from filling as close as possible to the target lines [go-left go-right (GG)]. Each bar took 1 s to fill completely; thus, a target response time (RT) of 800 ms was cued during GG trials (Fig. 1B). A subset of partial-stop trials was included to assess selective stopping. During partial-stop trials, either the left [stop-left go-right (SG)] or right [go-left stop-right (GS)] indicator automatically stopped filling before the target. The objective of partial-stop trials was to withhold the response on the stopped side while still responding in time with the target on the nonstopped side (hereafter referred to as stop-hand and respond-hand, respectively). The stop-signal delay (SSD) was initially set so that the filling ceased 250 ms before the target and was then adjusted in steps of 28 ms (∼4 frames) across partial-stop trial types independently. The SSD was increased (i.e., less time before target) after successful stopping and decreased (i.e., more time before target) after unsuccessful stopping to obtain an average stopping success of ∼50%. Points were awarded based on RT or stop success for each response and were signaled during the intertrial interval (1–2 s) by changing the color of the target lines to encourage accurate responding (green, <25 ms or successful stop; yellow, 26–50 ms; orange, 51–75 ms; red, >75 ms or failed stop).

Cues for stopping certainty were presented on every trial for 2 s before trial onset. Stop cues were embedded into the target line color. A black target line indicated a no-stop (0%) chance. Cyan and magenta indicated a low- (33%) or high-stop (66%) certainty (color counterbalanced across participants). Cues were presented over five primary trial types (Fig. 1C). A no-stop cue was presented for both hands to assess responses without an expectation of stopping (certain-go trials). A low-stop cue was presented for both hands to assess reactive selective stopping (uninformative trials). A low-stop or high-stop cue was presented for only one hand, and a no-stop cue was presented for the other to assess proactive selective stopping (low-informative and high-informative trials, respectively). GS and SG partial-stop trials were equally distributed for all stop cues. Finally, no target lines were presented in a small number of trials to assess TMS without the influence of response preparation (restin-task trials).

The experiment was split into behavioral and TMS sessions (mean separation 6.6 d). The behavioral session was always completed first. Initial SSD values for the TMS session were set to the participant's averages obtained during the behavioral session. Task instructions were given at the start of each session. Participants completed a practice block of certain-go trials and then uninformative trials for familiarization. Participants were informed that their primary goal was to earn as many points as possible. The block and total score were updated and displayed at the end of each block. The task protocol for each session consisted of 690 trials, split into 15 blocks of 46 trials with a random trial order (Table 1, detailed trial numbers). On average, the behavioral and TMS sessions lasted 90 min and 150 min, respectively.

View this table:
Table 1.

Trial numbers during anticipatory response inhibition task

Electromyography

Surface EMG was collected from the task-relevant left and right FDI using Ag-AgCl surface electrodes (CONMED) arranged in a belly-tendon montage (Fig. 1A). The activity of task-irrelevant left abductor pollicis brevis (APB) was also recorded during the TMS session. A shared ground electrode was positioned on the posterior surface of the left hand. EMG activity was amplified (×1000), bandpass filtered (10–1000 Hz), and sampled at 2000 Hz with a CED interface system (MICRO1401mkii, Cambridge Electronic Design). EMG collection was recorded from trial onset (−500–1500 ms) during the task and stimulation onset (−150–850 ms) during TMS at rest for later off-line analysis.

Transcranial magnetic stimulation

Two Magstim 2002 stimulators (Magstim) were used to deliver test and conditioning pulses through 70 and 50 mm figure-of-eight coils (100 µs pulse width, monophasic waveform) positioned over right and left M1, respectively. The coils were held by separate experimenters and oriented to induce a posterior-to-anterior flowing current in the underlying cortical tissue. The optimal coil position for eliciting a motor evoked potential (MEP) in the left and then right FDI was assessed and marked on the scalp.

Motor thresholds were determined using a maximum-likelihood parameter estimation by sequential testing strategy (https://www.clinicalresearcher.org/software.htm). Rest motor threshold was determined for right FDI and defined as the minimum stimulation intensity required to elicit an MEP amplitude of at least 0.05 mV. Active motor threshold was also determined for right FDI and defined as the minimum stimulation intensity required to elicit an MEP amplitude of at least 0.2 mV during low-intensity voluntary contraction. The test stimulus (TS) intensity was determined for left FDI and defined as the minimum stimulation intensity required to elicit an MEP amplitude of at least 0.5 mV at rest. For IHF, the interstimulus interval and conditioning stimulus (CS) intensity were set to 6 ms and 60% active motor threshold, respectively (Bäumer et al., 2006). For IHI, the interstimulus interval and CS intensity were set to 40 ms and 130% rest motor threshold, respectively (Harris-Love et al., 2007).

TMS was performed before and during the task. A total of 24 nonconditioned (i.e., TS only), 24 IHF, and 24 IHI trials were collected for each condition. TMS measures before the task (restpre-task) were collected over two blocks of 36 trials. Stimulated trials were intermixed with nonstimulated trials during standard blocks in the TMS session. The TS during stimulated trials was delivered 550 ms after trial onset to avoid contamination by EMG (Cirillo et al., 2018). The Arduino Leonardo response box controlled in-task stimulation timing to ensure synchronization between task software and TMS equipment. The SSD during stimulated trials was fixed to 550 ms so that stimulation never occurred after a stop signal. Thus, modulation by proactive response inhibition could be isolated from reactive processes.

Dependent measures

Data were processed using custom scripts in MATLAB (R2021b, version 9.11, MathWorks).

Behavioral data

Task data were analyzed from the behavioral session only to avoid the influence of TMS on performance. Responses were coded as errors and removed from further analyses if RTs were >400 ms from the target (average 0.27% of trials). Relative RT (RTrel) was calculated by subtracting the target time from the mean response time for each hand and trial type. Negative and positive RTrel values indicate early and late responses, respectively. Omissions were calculated as the percentage of GG trials where the response in the hand cued to stop was not enacted within the trial period. Response delay effect (RDE) provides a measure of proactive response inhibition at a behavioral level (i.e., response slowing). RDE was calculated by subtracting RTrel during certain-go trials from RTrel of the hand cued to stop for each stop-cue type. Stopping-interference provides a measure of stopping selectivity at the behavioral level, where greater values indicate larger interference. The stopping-interference effect was calculated by subtracting the mean RTrel during GG trials from the mean RTrel of the respond-hand during successful partial-stop trials. Mean SSD was also calculated for each stop-cue type and made relative to the target, where values closer to zero indicate less time for stopping. Stop-signal reaction time was not calculated in the current study as assumptions of the independent race model are often violated in selective stopping contexts (Bissett and Logan, 2014).

EMG data

EMG data were preprocessed as per Raud et al. (2022). Data epochs were bandpass filtered (20–250 Hz) using a second-order Butterworth filter and then resampled to 500 Hz. Filtered data were smoothed by taking root mean square over a 50 ms sliding window and then normalized to baseline (200–400 ms after trial onset). Last, data epochs were z-scored for each hand and participant separately. An EMG burst was identified if activity above threshold (1.2 z) was found between 0.4 and 1.2 s after trial onset. Burst onset and offset were determined by taking the time of the first data point in a consecutive group of five (i.e., 10 ms) that was below the threshold when working backward (onset) and forward (offset) from the peak. The peak rate of rise of the EMG burst (i.e., z/s) was determined as the maximum value of the differentiated signal between burst onset and peak.

EMG-dependent measures included burst-onset, burst-amplitude, burst-rise from successful partial-stop trials where an EMG burst was present in both the respond-hand, and stop-hand (average 33.9% of partial-stop trials). Cancel time was calculated by subtracting SSD from EMG burst offset in the stop-hand trial-by-trial (Jana et al., 2020). In this case, smaller cancel time values reflect faster stopping or faster cessation of going processes relative to stop signal onset (Raud et al., 2022). Δburst-onset was calculated by subtracting the EMG burst onset of the stop-hand from the respond-hand. Δburst-onset is an EMG proxy of stopping-interference that can be quantified on a trial-by-trial basis, where values >0 indicate nonselective stopping (Raud et al., 2020).

TMS data

Peak-to-peak MEP amplitude was calculated for the left FDI and APB between 10 and 50 ms after stimulation. MEPs were excluded when pretrigger root mean square (RMS) EMG activity exceeded 20 µV in a −100 to −50 ms pretrigger window (average 1.15% of trials). The top and bottom 10% of MEP amplitudes were trimmed for each condition before calculating the mean MEP amplitude (Wilcox, 2010), provided that at least 10 MEPs were available. Corticomotor excitability (CME) was determined as the mean of nonconditioned (NC) MEP amplitude for each condition. The magnitude of IHF and IHI were calculated for each condition as follows: IHF(%facilitation)=[C−NCNC]×100% IHI(%inhibition)=[NC−CNC]×100%, where C and NC refer to each participant's mean conditioned and nonconditioned MEP amplitude. ΔCME, ΔIHF, and ΔIHI were computed to determine modulation by proactive cuing by subtracting the mean value during certain-go trials from the mean value observed for each informative stop-cue type. Values >0 represent an upregulation or reinstatement of a particular measure relative to a certain-go response context.

Statistical analyses

Data were analyzed with Bayesian repeated-measures ANOVA using the BayesFactor package (https://cran.rproject.org/web/packages/BayesFactor/index.html) in R software, version 4.1.2 (https://www.R-project.org/). All models included random slopes and were fitted across 100,000 iterations with participant modeled as a random intercept (van den Bergh et al., 2022). Normality of data and model-averaged residual plots were visually inspected before ANOVA (van den Bergh et al., 2020). Logarithmic transformations were used for non-normal data. Interaction effects were determined by comparing models with the interaction term against matched models without the term (van den Bergh et al., 2020). Evidence for main effects and interactions were determined using Bayes factor (BF) in favor of the alternative hypothesis (BF10 ± percentage error), where values >1 indicate support for the alternative hypothesis, and values <1 support the null hypothesis. The strength of evidence was determined using a standard BF10 classification table (BF10 < 0.3, moderate evidence for null hypothesis; 0.3 ≤ BF10 ≤ 3, inconclusive evidence; BF10 > 3, moderate evidence for alternative hypothesis; van Doorn et al., 2021). Post hoc pairwise comparisons were performed using Bayesian paired t tests when a main effect or interaction was found. Corrected posterior odds (Opost.) were calculated by multiplying the uncorrected BF10 by the adjusted prior odds (van den Bergh et al., 2020). Prior odds were adjusted using the Westfall multiple comparisons approach (Westfall, 1997). Data are presented as nontransformed means ± SD unless otherwise specified.

Behavioral and electromyography data were assessed using two-way ANOVAs with the factors of Cue (uninformative, low-informative, high-informative) and Hand (left, right) to test the first and second hypothesis that stopping will be faster and stopping-interference less as stopping certainty increased. Proactive response slowing (RDE) was used to verify that participants were preparing and not withholding responses with proactive cuing. EMG cancel time was used to determine stopping speed as an alternative to stop-signal reaction time (Raud et al., 2022). Stopping-interference and Δburst-onset were used to determine stopping selectivity. To determine how proactive response inhibition affected the respond-hand and stop-hand separately, EMG burst onset, amplitude, and rise were modeled with one-way ANOVAs with the factor of Cue (uninformative, low-informative, high-informative). In this case, the analyses were performed for successful partial-stop trials where an EMG burst was observed in both hands.

IHF and IHI were analyzed with two-way ANOVAs with the factor of Context (restpre-task, restin-task, certain-go) and Muscle (FDI, APB) to test our third hypothesis that IHI will be released and IHF elicited from rest to in-task response preparation. RMS and CME were used to verify that modulation of IHI and IHF were not caused by changes in background excitability. ΔCME, ΔIHF, and ΔIHI were modeled with two-way ANOVAs with the factors of Cue (low-informative, high-informative) and Muscle (FDI, APB) to test our fourth hypothesis that an effector-specific upregulation of IHF and IHI would occur during informative cues in the respond-hand and stop-hand, respectively. In this case, the analyses were performed when the left side was cued to stop (i.e., stop-left cues) and cued to go (i.e., stop-right cues).

Results

Behavioral data are shown in Table 2 and Figure 2. For RDE, there was a main effect of Cue (BF10 = 7.65 × 1014 ± 0.01%). RDE for uninformative trials (9.1 ± 11.3 ms) was less than low-informative (22.8 ± 17.9 ms; Opost. = 6.36 × 104) and high-informative trials (50.4 ± 41.4 ms; Opost. = 2.10 × 106), which also differed from each other (Opost. = 95.04). There was a null main effect of Hand (BF10 = 0.18 ± 0.02%) and a Cue × Hand interaction (BF10 = 0.14 ± 0.03%). One participant was excluded from EMG analyses because of a technical issue during data collection, leaving 22 participants available for EMG analyses.

View this table:
Table 2.

Results for go and partial-stop trials during the behavioral session

Figure 2.
Figure 2.

Behavioral measures from partial-stop trials. A, RDE collapsed across left hand (LH) and right hand (RH). RDE calculated as the difference in response times between stop-cued and certain-go trials, where values >0 reflect response slowing. B, SSD collapsed across LH and RH. SSD is expressed as time relative to target response time, where values closer to zero reflect stop signals closer to the target. C, Stopping-interference (SI) calculated as the difference between mean response times during go trials from the response time of the respond-hand during successful partial-stop trials. Point ranges represent means with 95% bootstrap confidence intervals. Posterior odds** > 10, *** > 100; #Posterior odds > 3 for LH versus RH.

Better stopping with increased stopping certainty

For SSD, there was a main effect of Cue (BF10 = 2.42 × 1033 ± 0.01%). SSD for uninformative trials (−248.0 ± 25.9 ms) was less than low-informative (−216.1 ± 25.7 ms; Opost. = 1.71 × 106) and high-informative trials (−171.7 ± 30.6 ms; Opost. = 3.21 × 1018), which also differed from each other (Opost. = 1.84 × 109). There was a null main effect of Hand (BF10 = 0.18 ± 0.01%), whereas the Cue × Hand interaction was inconclusive (BF10 = 0.52 ± 0.02%). Post hoc Bayesian correlations indicated a moderate association between go RTrel and SSD for uninformative trials (r = 0.56, Opost. = 4.81), inconclusive evidence for low-informative trials (r = 0.49, Opost. = 2.01), and a null association for high-informative trials (r = 0.21, Opost. = 0.23). Responses slowed and the required time for stopping decreased as stopping certainty increased. For cancel time (Fig. 3A), there was a main effect of Cue (BF10 = 4.54 × 1017 ± 0.04%). Cancel time for uninformative trials (279.4 ± 38.4 ms) was greater than low-informative (251.8 ± 48.8 ms; Opost. = 174.95) and high-informative trials (204.3 ± 46.9 ms; Opost. = 9.05 × 1011), which also differed from each other (Opost. = 3.21 × 104). There was a null main effect of Hand (BF10 = 0.19 ± 0.01%) and a Cue × Hand interaction (BF10 = 0.14 ± 0.04%). Stopping occurred more quickly for both hands with increased stopping certainty.

Figure 3.
Figure 3.

EMG characteristics during successful partial-stop trials where a burst was observed in both the stop-hand and respond-hand. A, Modulation of stopping speed (cancel time) and selectivity (Δburst-onset) by stopping certainty (uninformative, low-informative, high-informative). Cancel time calculated as the difference in time between stop-signal and EMG burst-offset in the stop-hand, where greater values indicate slower stopping. Δburst-onset calculated as the difference in onset time between the respond-hand and stop-hand, where values above zero indicate within-trial stopping-interference. B, Stop-locked grand average EMG traces from successful partial-stop trials. Values reflect means with 95% confidence interval bands. Certain-go trials are time locked to 550 ms (equivalent to average stop time for uninformative partial-stop trials). Dashed vertical lines indicate the average RT of the respond-hand for each stop condition. C, Modulation of EMG burst-onset and burst-rise by stopping certainty. Point ranges represent means with 95% bootstrap confidence intervals. Posterior odds* > 3, ** > 10, *** > 100.

Better selectivity of stopping with proactive cueing

For the stopping-interference effect, there was a moderate Cue × Hand interaction (BF10 = 4.00 ± 0.04%). Left hand stopping-interference for uninformative trials (63.4 ± 23.9 ms) was less than low-informative (34.5 ± 23.5 ms; Opost. = 2.46 × 103) and high-informative trials (5.7 ± 15.5 ms; Opost. = 6.35 × 106), which also differed from each other (Opost. = 1.73 × 103). Right hand stopping-interference for uninformative trials (47.7 ± 15.5 ms) was less than low-informative (27.4 ± 15.8 ms; Opost. = 442.32) and high-informative trials (7.7 ± 17.9 ms; Opost. = 2.49 × 104), which also differed from each other (Opost. = 17.31). There was an asynchrony in stopping-interference between the left and right hand for uninformative (Opost. = 24.67) but not low-informative (Opost. = 0.86) or high-informative (Opost. = 0.24) trials. One-sample t tests against zero indicated a stopping-interference effect for uninformative (Opost. = 9.92 × 109) and low-informative (Opost. = 3.07 × 105) partial-stop trials, but inconclusive evidence for high-informative partial-stop trials (Opost. = 2.35). For Δburst-onset (Fig. 3A), there was a main effect of Cue (BF10 = 8.33 × 1013 ± 0.03%). Δburst-onset for uninformative trials (28.1 ± 30.1 ms) was greater than low-informative (−5.7 ± 35.1 ms; Opost. = 3.80 × 105) and high-informative trials (−30.7 ± 37.6 ms; Opost. = 4.05 × 106), which also differed from each other (Opost. = 23.39). There was a null main effect of Hand (BF10 = 0.20 ± 0.03%) and a Cue × Hand interaction (BF10 = 0.28 ± 0.04%). There was no observable interference in EMG when informative stop-cues were presented. The selectivity of stopping improved at a behavioral level as stopping certainty increased.

Grand average EMG bursts from the respond-hand and stop-hand during successful partial-stop trials are shown in Figure 3B. For the respond-hand, there was a main effect of Cue on burst-onset (BF10 = 2.06 × 109 ± 0.01%). Burst-onset occurred earlier for high (710.4 ± 38.6 ms) compared with low-informative (734.2 ± 43.7 ms; Opost. = 1.92 × 103) and uninformative partial-stop trials (764.8 ± 48.9 ms; Opost. = 5.27 × 104), which also differed from each other (Opost. = 2.16 × 103). There was a main effect of Cue on burst-amplitude (BF10 = 5.02 ± 0.03%), however, post hoc tests did not provide conclusive evidence (all Opost. < 3). The main effect of Cue on burst-rise was inconclusive (BF10 = 0.95 ± 0.03%). For the stop-hand, there was a null main effect of Cue on burst-onset (BF10 = 0.18 ± 0.03%) and inconclusive evidence for burst-amplitude (BF10 = 0.33 ± 0.04%). There was a main effect of Cue on burst-rise (BF10 = 627.07 ± 0.01%). Burst-rise was smaller for high-informative trials (1909.7.9 ± 772.6 z/s) compared with low-informative (2346.3 ± 533.1 z/s; Opost. = 3.55) and uninformative partial-stop trials (2568.9 ± 464.7 z/s; Opost. = 61.26), which did not differ from each other (Opost. = 0.63). Successful partial-stop trials were marked by a smaller EMG burst-onset and burst-rise in the respond-hand and stop-hand, respectively (Fig. 3C).

Context-dependent modulation of IHI but not IHF during response preparation

Five participants were excluded from the TMS session because of high resting FDI resting motor thresholds, which caused coil overheating during task blocks. The high motor thresholds were likely because of the small coil size required for dual-hemisphere stimulation. After exclusions, data from 18 participants were available for TMS analyses. The main findings did not change if participants with incomplete datasets were excluded from behavioral analyses. The TS intensity for the remaining participants was 53.5 ± 10.6% of maximum stimulator output (MSO). The CS intensity for the IHF and IHI protocols was 29.6 ± 6.2% and 75.7 ± 16.9% MSO, respectively. On average, 23.7 ± 0.9 trials (range 16–24 trials) were available for each TMS trial type. One-sample t tests on the differences between the behavioral session and nonstimulated trials in the TMS session indicated that RDE (Opost. = 0.89) and stopping-interference (Opost.= 0.31) were similar, whereas mean SSD was slightly longer (9.1 ± 10.4 ms; Opost. = 17.85).

Context-dependent TMS data are shown in Figure 4. For IHF, there was a null main effect of Muscle (BF10 = 0.20 ± 0.01%) and Context (BF10 = 0.21 ± 0.01%), as well as a null Context × Muscle interaction (BF10 = 0.09 ± 0.02%). One-sample t tests against zero indicated no IHF across response contexts (all Opost. values < 0.3). For IHI, there was a null main effect of Muscle (BF10 = 0.19 ± 0.01%) and a null Context × Muscle interaction (BF10 = 0.19 ± 0.01%). There was a main effect of Context (BF10 = 3.59 × 1029 ± 0.02%). IHI during restpre-task (45.4 ± 22.9%) was greater than restin-task (28.1 ± 23.3%; Opost. = 111.23), certain-go (−16.1 ± 27.5%; Opost. = 5.01 × 109), and uninformative contexts (−13.7 ± 28.0%; Opost. = 3.97 × 1010). Restin-task was also greater than certain-go (Opost. = 6.46 × 105) and uninformative contexts (Opost. = 1.18 × 106). There was no difference between certain-go and uninformative contexts (Opost. = 0.05). One-sample t tests against zero indicated IHI during restpre-task (Opost. = 6.35 × 104) and restin-task (Opost. = 170.62) contexts, whereas IHI during certain-go (Opost. = 2.55) and uninformative contexts (Opost. = 0.81) was inconclusive. A post hoc one-way ANOVA on average MEP amplitudes elicited in the right FDI from the CS during IHI trials indicated a null main effect of Cue (BF10 = 0.13 ± 0.02%). IHI was modulated by context for task-relevant and irrelevant muscles.

Figure 4.
Figure 4.

Transcranial magnetic stimulation was used to examine CME, IHF, and IHI. A, Electromyography traces with MEPs in the task-relevant left first dorsal interosseous muscle during pretask rest. Each trace is the average of 24 trials from a representative participant. For IHF, a subthreshold CS was delivered 6 ms before the TS. For IHI, a suprathreshold CS was delivered 40 ms before the TS. B, CME calculated as mean peak-to-peak motor MEP amplitude. C, IHF calculated as percentage facilitation, where greater values indicate a larger conditioned relative to nonconditioned MEP amplitude (i.e., more facilitation). D, IHI calculated as percentage inhibition, where greater values indicate a smaller conditioned relative to nonconditioned MEP amplitude (i.e., more inhibition). †Posterior odds > 3 from one-sample t test against zero. Point ranges represent means with 95% bootstrap confidence intervals. ***Posterior odds > 100, #Posterior odds < 0.03.

For CME, there was a main effect of Muscle (BF10 = 5.55 × 106 ± 0.01%). CME was less in FDI (0.66 ± 0.41 mV) compared with APB (0.67 ± 0.89 mV; Opost. = 10.17) across response contexts. There was a null Context × Muscle interaction (BF = 0.16 ± 0.00%), whereas the main effect of Context was inconclusive (BF10 = 1.69 ± 0.00%). For RMS, there was a main effect of Muscle (BF10 = 56.82 ± 0.02%), however, post hoc comparisons were inconclusive (Opost. = 1.89). There was inconclusive evidence for a main effect of Context (BF10 = 1.26 ± 0.00%) and a Context × Muscle interaction (BF10 = 0.45 ± 0.03%). Modulation of IHI occurred in the absence of shifts in excitability and background EMG activity.

Cue-dependent modulation of CME but not IHI or IHF during informative stop-left cues

Cue-dependent modulation of motor evoked potentials are shown in Figure 5. For ΔCME there was a main effect of Muscle (BF10 = 15.89 ± 0.01%) when the left side was cued to stop. ΔCME was larger in FDI (−0.07 ± 0.14 mV) compared with APB (0.00 ± 0.09 mV; Opost. = 5.30). There was a null main effect of Cue (BF10 = 0.28 ± 0.00%), whereas the Cue × Hand interaction was inconclusive (BF10 = 0.49 ± 0.02%). When the left side was cued to Go (stop-right) the main effect of Muscle (BF10 = 0.78 ± 0.01%) and Cue (BF10 = 0.41 ± 0.01%) was inconclusive. There was a Cue × Hand interaction (BF10 = 5.23 ± 0.02%); however, post hoc comparisons were inconclusive (all Opost. values < 3). CME was suppressed in the task-relevant left FDI muscle during informative stop-left cues.

Figure 5.
Figure 5.

Modulation of transcranial magnetic stimulation measures by informative partial-stop cues in the task-relevant left FDI and task-irrelevant APB muscles. All measures have been calculated as the difference from certain go trials (dashed horizontal line), where positive and negative values indicate upregulation and downregulation, respectively. A, CME calculated as mean peak-to-peak MEP amplitude. B, IHF calculated as percentage facilitation, where greater values indicate larger conditioned relative to nonconditioned MEP amplitude (i.e., more facilitation). C, IHI calculated as percentage inhibition, where greater values indicate smaller conditioned relative to nonconditioned MEP amplitude (i.e., more inhibition). Point ranges represent means with 95% bootstrap confidence intervals. **Posterior odds > 10.

For ΔIHF, when cued to stop the main effect of Muscle (BF10 = 0.32 ± 0.01%), Cue (BF10 = 0.78 ± 0.01%) and a Cue × Hand interaction (BF10 = 0.33 ± 0.02%) was inconclusive. When cued to Go (stop-right), the main effect of Muscle (BF10 = 0.41 ± 0.01%), Cue (BF10 = 1.04 ± 0.01%), and a Cue × Hand interaction (BF10 = 0.34 ± 0.02%) was inconclusive. For ΔIHI when cued to stop, there was a null main effect of Muscle (BF10 = 0.28 ± 0.04%) and Cue (BF10 = 0.28 ± 0.02%), but the Cue × Hand interaction was inconclusive (BF10 = 0.43 ± 0.03%). For ΔIHI when cued to go (stop-right cues), there was a null main effect of Muscle (BF10 = 0.27 ± 0.06%). The main effect of Cue (BF10 = 0.52 ± 0.16%) and Cue × Hand interaction (BF10 = 0.35 ± 0.04%) was inconclusive. There was no modulation of IHF or IHI from certain-go to informative stop contexts.

Discussion

Faster stopping with increased stopping certainty

In support of the first hypothesis, selective stopping was faster with proactive cuing. Response slowing (RDE) was specific to the cued hand and increased with stopping certainty. Although omissions also increased with stopping certainty, responses were enacted on most trials, and signatures of response preparation (EMG bursts and IHI) during high-informative cues were more like the certain-go than the restin-task (no response preparation) context. As such, response slowing likely reflected the behavioral manifestation of proactive response inhibition to improve stopping (Verbruggen and Logan, 2009). Stopping performance improved from uninformative to informative partial-stop trials as evident by shorter SSDs and EMG cancel times. The shorter cancel times indicate that stopping, or the time taken to disengage excitatory processes, was faster during informative partial-stop trials (Raud et al., 2022). Interestingly, the cancel time in the reactive (uninformative) context was ∼100 ms longer than reported during unimanual stopping (Jana et al., 2020; Raud et al., 2022). Visual inspection of the EMG traces indicates that the restart process may have extended to both hands during successful partial-stopping. The cancel times may have been elevated partly by functional coupling in a selective stopping context. Although nonselective stopping was not explicitly assessed, the cancel times in the present study provide evidence of slower stopping in selective compared with nonselective stopping contexts (Smittenaar et al., 2013).

Better selective stopping through proactive response inhibition

In support of the second hypothesis, the stopping-interference effect was smaller during low-informative compared with uninformative partial-stop trials (Raud and Huster, 2017). Stopping-interference was smallest during high-informative partial-stop trials and larger in the left hand than the right hand during uninformative but not informative partial-stop trials. A between-hand discrepancy may result from hand dominance and coupling. The functional coupling account suggests that part of the stopping-interference effect arises from decoupling the respond-hand from the stop-hand (Wadsley et al., 2022b). Stopping-interference may be larger in the nondominant hand as it is more stringently coupled to the dominant hand than vice versa (Byblow et al., 2000). The discrepancy may be absent during informative partial-stop trials because proactive control can target the cued hand regardless of dominance.

No stopping-interference was observed during informative partial-stop trials with a trial-by-trial measure of response delays (Δburst-onset). Indeed, EMG burst-onsets were only delayed in the respond-hand relative to the stop-hand during uninformative partial-stop trials. Greater stopping selectivity during informative partial-stop trials may have occurred through separate adjustments to the respond-hand and stop-hand. For the respond-hand, EMG burst-onset but not peak rate of rise decreased with stopping certainty and indicates that an equivalent go process was planned but released earlier (Coxon et al., 2007). In contrast, EMG burst-onset was unchanged, but the peak rate of rise was smaller in the stop-hand with greater stopping certainty. A smaller rate of increase is indicative of proactive response inhibition of the stop-cued hand (MacDonald et al., 2017). Proactive response inhibition may have effectively decoupled the hands to some degree during response preparation. Indeed, stopping-interference is smaller when selective stopping is assessed in the context of decoupled responses (Wadsley et al., 2019, 2022a). Therefore, better selective stopping with proactive cuing occurs through suppression of the stop-hand and an earlier release of the go process in the respond-hand.

Modulation of IHI but not IHF from rest to in-task contexts

In opposition to our hypothesis, there was a null effect of the IHF protocol in both rest and in-task response preparation contexts. Reports of M1–M1 facilitation are limited (Bäumer et al., 2006) and may be observed at longer interstimulus intervals (Fiori et al., 2017). Alternatively, IHF may be driven by premotor–M1 interactions (Neige et al., 2021). IHF at similarly short interstimulus intervals has been observed during task switching when conditioning the premotor cortex (Mars et al., 2009). In summary, there was no evidence for task-dependent modulation of M1–M1 IHF.

In support of the third hypothesis, there was a release of IHI from rest to in-task response preparation. A similar decrease in short-interval IHI was observed previously (MacDonald et al., 2021) and may reflect disinhibition for action selection. It could be advantageous for IHI to be maintained in task-irrelevant effectors as a potential center-surround inhibitory mechanism (Hinder et al., 2018). However, in the current study, a release of IHI was also observed in the task-irrelevant APB muscle. This nonspecific release of IHI may reflect a wide aperture of sensorimotor disinhibition and the functional similarity of FDI and APB (Labruna et al., 2019). Interestingly, IHI was present but smaller during in-task rest than pretask rest. It is unlikely that the observed modulation of IHI was the consequence of general shifts in excitability as CME did not differ across rest and in-task contexts. Therefore, IHI is released during response preparation and reduced in contexts that necessitate frequent responding.

IHI was not modulated between cues signaling certain-go and uninformative trials. An absence of upregulation from a pure go context to one where stopping may be required indicates IHI may not be involved in proactive response inhibition. The pattern of IHI modulation contrasts with that observed during within-hemisphere probes of GABA-B receptor-mediated inhibition, which were elevated for uninformative trials and positively associated with the magnitude of stopping-interference (Cowie et al., 2016). The discrepancy may be a consequence of the certain-go context being assessed with a trial-by-trial rather than a block-by-block design as GABA-B is insensitive to trialwise modulation (Cirillo et al., 2018). The pattern of IHI modulation indicates a role in setting a general inhibitory tone that is not engaged for proactive response inhibition.

Effector-specific CME but not IHI modulation during informative cues

The fourth hypothesis was not supported because proactive response inhibition was marked by effector-specific CME but not IHI modulation. CME suppression occurred in the task-relevant effector when cued about stopping (informative stop-left cues). This suppression corroborates previous investigations of proactive selective stopping and indicates that CME suppression may be a general feature of response preparation in contexts that may require stopping (Raud et al., 2020). IHI was released across effectors regardless of whether the probed hand was cued to stop. As such, trial-by-trial suppression of CME was not supported by a reinstatement of IHI. Proactive response inhibition during informative cues may instead be driven through separate neural mechanisms such as proactive engagement of within-hemisphere intracortical networks (Cirillo et al., 2018) or the indirect cortico–basal ganglia pathway (Majid et al., 2013). In summary, IHI was reduced between resting and response preparation states but did not contribute to cue-dependent CME suppression.

Preparatory influences on response inhibition during selective stopping

There were signatures of nonselective response inhibition regardless of the magnitude of stopping-interference. Global motor suppression was evident in EMG activity that was ceased momentarily in the stopped and nonstopped hand after the stop signal. The suppression was present during high-informative conditions where stopping-interference was minimal. Converging evidence indicates that preparation for going can influence stopping selectivity (Muralidharan et al., 2019), for example, through the timing of the go response (Raud et al., 2020) or by manipulating response strategy (Xu et al., 2015; Wadsley et al., 2022a). The present findings advance our understanding of response inhibition by demonstrating that proactive adjustments for selective stopping manifest separately in the respond-hand and stop-hand. Returning to our example of changing lanes on a busy motorway, a vigilant driver may proactively suppress the steering effectors that may suddenly have to cancel their response while at the same time maintaining output over the responding effectors (foot on the accelerator). Such proactive control can make the responding effectors resistant to nonselective response inhibition. Overall, the magnitude of stopping-interference depends as much on how an individual prepares to go as it does on how one prepares to stop.

Limitations

The present study has some limitations. Although trial order was randomized across many blocks, sequence effects may have influenced trial-to-trial variability of responses. TMS was applied at only one time point that corresponded to the typical SSD required for a 50% partial-stop success rate in healthy young adults (Wadsley et al., 2022a). Proactive modulation of IHI or IHF may have occurred closer or farther from the time of responding. However, a reliable number of stimulated trials could be collected by limiting the time points for in-task TMS (Goldsworthy et al., 2016). Another limitation relates to the parameterization of TMS for investigating IHF, particularly in terms of localizing the conditioning pulse. The current study examined only M1–M1 interactions. Future studies might target corticocortical interhemispheric interactions between prefrontal and M1 regions during selective stopping (Friehs et al., 2021). Alternatively, future studies may opt to use neurophysiological methods that are better suited to trial-to-trial brain-behavior correlations than TMS (Wessel, 2020).

Conclusion

The present study provides novel insight into proactive response inhibition and selective stopping. Behaviorally, stopping became faster and more selective with proactive cuing. Interhemispheric inhibition reduced from pretask rest to in-task contexts to support bimanual responding but did not increase when cued to stop. Improved stopping selectivity may instead have occurred via targeted suppression of the stop-hand and an earlier release of the go process in the respond-hand during preparation for going. These findings indicate that stopping can become more selective with proactive cuing, although cue-related improvements are unlikely to reflect proactive engagement of facilitatory or inhibitory M1–M1 influences for response inhibition.

Footnotes

  • We thank April Ren and Kelly Tay for assistance with data collection.

  • The authors declare no competing financial interests.

  • Correspondence should be addressed to Winston D. Byblow at w.byblow{at}auckland.ac.nz

References

  1. ↵
    2. Aron AR,
    3. Verbruggen F
    (2008) Stop the presses: dissociating a selective from a global mechanism for stopping. Psychol Sci 19:1146–1153. doi:10.1111/j.1467-9280.2008.02216.x pmid:19076487
    OpenUrlCrossRefPubMed
  2. ↵
    2. Bäumer T,
    3. Bock F,
    4. Koch G,
    5. Lange R,
    6. Rothwell JC,
    7. Siebner HR,
    8. Münchau A
    (2006) Magnetic stimulation of human premotor or motor cortex produces interhemispheric facilitation through distinct pathways. J Physiol 572:857–868. doi:10.1113/jphysiol.2006.104901 pmid:16497712
    OpenUrlCrossRefPubMed
  3. ↵
    2. Bissett PG,
    3. Logan GD
    (2014) Selective stopping? Maybe not. J Exp Psychol Gen 143:455–472. doi:10.1037/a0032122 pmid:23477668
    OpenUrlCrossRefPubMed
  4. ↵
    2. Byblow WD,
    3. Lewis GN,
    4. Stinear JW,
    5. Austin NJ,
    6. Lynch M
    (2000) The subdominant hand increases in the efficacy of voluntary alterations in bimanual coordination. Exp Brain Res 131:366–374. doi:10.1007/s002219900271 pmid:10789951
    OpenUrlCrossRefPubMed
  5. ↵
    2. Cai W,
    3. Oldenkamp CL,
    4. Aron AR
    (2011) A proactive mechanism for selective suppression of response tendencies. J Neurosci 31:5965–5969. doi:10.1523/JNEUROSCI.6292-10.2011 pmid:21508221
    OpenUrlAbstract/FREE Full Text
  6. ↵
    2. Cirillo J,
    3. Cowie MJ,
    4. MacDonald HJ,
    5. Byblow WD
    (2018) Response inhibition activates distinct motor cortical inhibitory processes. J Neurophysiol 119:877–886. doi:10.1152/jn.00784.2017 pmid:29212923
    OpenUrlCrossRefPubMed
  7. ↵
    2. Cowie MJ,
    3. MacDonald HJ,
    4. Cirillo J,
    5. Byblow WD
    (2016) Proactive modulation of long-interval intracortical inhibition during response inhibition. J Neurophysiol 116:859–867. doi:10.1152/jn.00144.2016 pmid:27281744
    OpenUrlCrossRefPubMed
  8. ↵
    2. Coxon JP,
    3. Stinear CM,
    4. Byblow WD
    (2007) Selective inhibition of movement. J Neurophysiol 97:2480–2489. doi:10.1152/jn.01284.2006 pmid:17251361
    OpenUrlCrossRefPubMed
  9. ↵
    2. Daskalakis ZJ,
    3. Christensen BK,
    4. Fitzgerald PB,
    5. Roshan L,
    6. Chen R
    (2002) The mechanisms of interhemispheric inhibition in the human motor cortex. J Physiol 543:317–326. doi:10.1113/jphysiol.2002.017673 pmid:12181302
    OpenUrlCrossRefPubMed
  10. ↵
    2. Fiori F,
    3. Chiappini E,
    4. Candidi M,
    5. Romei V,
    6. Borgomaneri S,
    7. Avenanti A
    (2017) Long-latency interhemispheric interactions between motor-related areas and the primary motor cortex: a dual site TMS study. Sci Rep 7:14936. pmid:29097700
    OpenUrlPubMed
  11. ↵
    2. Friehs MA,
    3. Frings C,
    4. Hartwigsen G
    (2021) Effects of single-session transcranial direct current stimulation on reactive response inhibition. Neurosci Biobehav Rev 128:749–765. doi:10.1016/j.neubiorev.2021.07.013 pmid:34271027
    OpenUrlCrossRefPubMed
  12. ↵
    2. Goldsworthy MR,
    3. Hordacre B,
    4. Ridding MC
    (2016) Minimum number of trials required for within- and between-session reliability of TMS measures of corticospinal excitability. Neuroscience 320:205–209. doi:10.1016/j.neuroscience.2016.02.012 pmid:26872998
    OpenUrlCrossRefPubMed
  13. ↵
    2. Harris-Love ML,
    3. Perez MA,
    4. Chen R,
    5. Cohen LG
    (2007) Interhemispheric inhibition in distal and proximal arm representations in the primary motor cortex. J Neurophysiol 97:2511–2515. doi:10.1152/jn.01331.2006 pmid:17215494
    OpenUrlCrossRefPubMed
  14. ↵
    2. Hinder MR,
    3. Puri R,
    4. Kemp S,
    5. Waitzer S,
    6. Reissig P,
    7. Stöckel T,
    8. Fujiyama H
    (2018) Distinct modulation of interhemispheric inhibitory mechanisms during movement preparation reveals the influence of cognition on action control. Cortex 99:13–29. doi:10.1016/j.cortex.2017.10.002 pmid:29121485
    OpenUrlCrossRefPubMed
  15. ↵
    2. Irlbacher K,
    3. Brocke J,
    4. Mechow JV,
    5. Brandt SA
    (2007) Effects of GABA(A) and GABA(B) agonists on interhemispheric inhibition in man. Clin Neurophysiol 118:308–316. doi:10.1016/j.clinph.2006.09.023 pmid:17174150
    OpenUrlCrossRefPubMed
  16. ↵
    2. Jana S,
    3. Hannah R,
    4. Muralidharan V,
    5. Aron AR
    (2020) Temporal cascade of frontal, motor and muscle processes underlying human action-stopping. Elife 9:e50371. doi:10.7554/eLife.50371
    OpenUrlCrossRef
  17. ↵
    2. Koch G,
    3. Franca M,
    4. Del Olmo MF,
    5. Cheeran B,
    6. Milton R,
    7. Alvarez Sauco M,
    8. Rothwell JC
    (2006) Time course of functional connectivity between dorsal premotor and contralateral motor cortex during movement selection. J Neurosci 26:7452–7459. doi:10.1523/JNEUROSCI.1158-06.2006 pmid:16837593
    OpenUrlAbstract/FREE Full Text
  18. ↵
    2. Labruna L,
    3. Tischler C,
    4. Cazares C,
    5. Greenhouse I,
    6. Duque J,
    7. Lebon F,
    8. Ivry RB
    (2019) Planning face, hand, and leg movements: anatomical constraints on preparatory inhibition. J Neurophysiol 121:1609–1620. doi:10.1152/jn.00711.2018 pmid:30785815
    OpenUrlCrossRefPubMed
  19. ↵
    2. MacDonald HJ,
    3. Laksanaphuk C,
    4. Day A,
    5. Byblow WD,
    6. Jenkinson N
    (2021) The role of interhemispheric communication during complete and partial cancellation of bimanual responses. J Neurophysiol 125:875–886. doi:10.1152/jn.00688.2020 pmid:33567982
    OpenUrlCrossRefPubMed
  20. ↵
    2. MacDonald HJ,
    3. McMorland AJ,
    4. Stinear CM,
    5. Coxon JP,
    6. Byblow WD
    (2017) An activation threshold model for response inhibition. PLoS One 12:e0169320. doi:10.1371/journal.pone.0169320 pmid:28085907
    OpenUrlCrossRefPubMed
  21. ↵
    2. Majid DS,
    3. Cai W,
    4. Corey-Bloom J,
    5. Aron AR
    (2013) Proactive selective response suppression is implemented via the basal ganglia. J Neurosci 33:13259–13269. doi:10.1523/JNEUROSCI.5651-12.2013 pmid:23946385
    OpenUrlAbstract/FREE Full Text
  22. ↵
    2. Majid DS,
    3. Cai W,
    4. George JS,
    5. Verbruggen F,
    6. Aron AR
    (2012) Transcranial magnetic stimulation reveals dissociable mechanisms for global versus selective corticomotor suppression underlying the stopping of action. Cereb Cortex 22:363–371. doi:10.1093/cercor/bhr112 pmid:21666129
    OpenUrlCrossRefPubMed
  23. ↵
    2. Mars RB,
    3. Klein MC,
    4. Neubert FX,
    5. Olivier E,
    6. Buch ER,
    7. Boorman ED,
    8. Rushworth MF
    (2009) Short-latency influence of medial frontal cortex on primary motor cortex during action selection under conflict. J Neurosci 29:6926–6931. doi:10.1523/JNEUROSCI.1396-09.2009 pmid:19474319
    OpenUrlAbstract/FREE Full Text
  24. ↵
    2. Muralidharan V,
    3. Yu X,
    4. Cohen MX,
    5. Aron AR
    (2019) Preparing to stop action increases beta band power in contralateral sensorimotor cortex. J Cogn Neurosci 31:657–668. doi:10.1162/jocn_a_01373 pmid:30633601
    OpenUrlCrossRefPubMed
  25. ↵
    2. Neige C,
    3. Rannaud Monany D,
    4. Lebon F
    (2021) Exploring cortico-cortical interactions during action preparation by means of dual-coil transcranial magnetic stimulation: a systematic review. Neurosci Biobehav Rev 128:678–692. doi:10.1016/j.neubiorev.2021.07.018 pmid:34274404
    OpenUrlCrossRefPubMed
  26. ↵
    2. Peirce J,
    3. Gray JR,
    4. Simpson S,
    5. MacAskill M,
    6. Höchenberger R,
    7. Sogo H,
    8. Kastman E,
    9. Lindeløv JK
    (2019) PsychoPy2: experiments in behavior made easy. Behav Res Methods 51:195–203. doi:10.3758/s13428-018-01193-y pmid:30734206
    OpenUrlCrossRefPubMed
  27. ↵
    2. Raud L,
    3. Huster RJ
    (2017) The temporal dynamics of response inhibition and their modulation by cognitive control. Brain Topogr 30:486–501. doi:10.1007/s10548-017-0566-y pmid:28456867
    OpenUrlCrossRefPubMed
  28. ↵
    2. Raud L,
    3. Huster RJ,
    4. Ivry RB,
    5. Labruna L,
    6. Messel MS,
    7. Greenhouse I
    (2020) A single mechanism for global and selective response inhibition under the influence of motor preparation. J Neurosci 40:7921–7935. doi:10.1523/JNEUROSCI.0607-20.2020 pmid:32928884
    OpenUrlAbstract/FREE Full Text
  29. ↵
    2. Raud L,
    3. Thunberg C,
    4. Huster RJ
    (2022) Partial response electromyography as a marker of action stopping. Elife 11:e70332. doi:10.7554/eLife.70332
    OpenUrlCrossRef
  30. ↵
    2. Smittenaar P,
    3. Guitart-Masip M,
    4. Lutti A,
    5. Dolan RJ
    (2013) Preparing for selective inhibition within frontostriatal loops. J Neurosci 33:18087–18097. doi:10.1523/JNEUROSCI.2167-13.2013 pmid:24227719
    OpenUrlAbstract/FREE Full Text
  31. ↵
    2. van den Bergh D,
    3. Van Doorn J,
    4. Marsman M,
    5. Draws T,
    6. Van Kesteren E-J,
    7. Derks K,
    8. Dablander F,
    9. Gronau QF,
    10. Kucharský Š,
    11. Gupta ARKN
    (2020) A tutorial on conducting and interpreting a Bayesian ANOVA in JASP. L'Année Psychologique 120:73–96.
    OpenUrl
  32. ↵
    2. van den Bergh D,
    3. Wagenmakers E-J,
    4. Aust F
    (2022) Bayesian repeated-measures ANOVA: an updated methodology implemented in JASP. PsyArXiv Advance online publication. Retrieved December 27, 2022. doi:https://doi.org/10.31234/osf.io/fb8zn.
    OpenUrlCrossRef
  33. ↵
    2. van Doorn J,
    3. van den Bergh D,
    4. Böhm U,
    5. Dablander F,
    6. Derks K,
    7. Draws T,
    8. Etz A,
    9. Evans NJ,
    10. Gronau QF,
    11. Haaf JM,
    12. Hinne M,
    13. Kucharský Š,
    14. Ly A,
    15. Marsman M,
    16. Matzke D,
    17. Gupta ARKN,
    18. Sarafoglou A,
    19. Stefan A,
    20. Voelkel JG,
    21. Wagenmakers E-J
    (2021) The JASP guidelines for conducting and reporting a Bayesian analysis. Psychon Bull Rev 28:813–826. doi:10.3758/s13423-020-01798-5 pmid:33037582
    OpenUrlCrossRefPubMed
  34. ↵
    2. Veale JF
    (2014) Edinburgh Handedness Inventory—Short Form: a revised version based on confirmatory factor analysis. Laterality 19:164–177. doi:10.1080/1357650X.2013.783045 pmid:23659650
    OpenUrlCrossRefPubMed
  35. ↵
    2. Verbruggen F,
    3. Logan GD
    (2009) Proactive adjustments of response strategies in the stop-signal paradigm. J Exp Psychol Hum Percept Perform 35:835–854. doi:10.1037/a0012726 pmid:19485695
    OpenUrlCrossRefPubMed
  36. ↵
    2. Wadsley CG,
    3. Cirillo J,
    4. Byblow WD
    (2019) Between-hand coupling during response inhibition. J Neurophysiol 122:1357–1366. doi:10.1152/jn.00310.2019 pmid:31339791
    OpenUrlCrossRefPubMed
  37. ↵
    2. Wadsley CG,
    3. Cirillo J,
    4. Nieuwenhuys A,
    5. Byblow WD
    (2022a) Decoupling countermands nonselective response inhibition during selective stopping. J Neurophysiol 127:188–203. doi:10.1152/jn.00495.2021 pmid:34936517
    OpenUrlCrossRefPubMed
  38. ↵
    2. Wadsley CG,
    3. Cirillo J,
    4. Nieuwenhuys A,
    5. Byblow WD
    (2022b) Stopping interference in response inhibition: behavioral and neural signatures of selective stopping. J Neurosci 42:156–165. doi:10.1523/JNEUROSCI.0668-21.2021 pmid:35022327
    OpenUrlAbstract/FREE Full Text
  39. ↵
    2. Wessel JR
    (2020) β-Bursts reveal the trial-to-trial dynamics of movement initiation and cancellation. J Neurosci 40:411–423. doi:10.1523/JNEUROSCI.1887-19.2019 pmid:31748375
    OpenUrlAbstract/FREE Full Text
  40. ↵
    2. Westfall PH
    (1997) Multiple testing of general contrasts using logical constraints and correlations. J Am Stat Assoc 92:299–306. doi:10.1080/01621459.1997.10473627
    OpenUrlCrossRef
  41. ↵
    2. Wilcox RR
    (2010) Fundamentals of Modern Statistical Methods (2nd Ed.). Springer-Verlag New York.
  42. ↵
    2. Xu J,
    3. Westrick Z,
    4. Ivry RB
    (2015) Selective inhibition of a multicomponent response can be achieved without cost. J Neurophysiol 113:455–465. doi:10.1152/jn.00101.2014 pmid:25339712
    OpenUrlCrossRefPubMed
Back to top

In this issue

Email
Print
View Full Page PDF
Citation Tools
Respond to this article
Proactive Interhemispheric Disinhibition Supports Response Preparation during Selective Stopping
Corey G. Wadsley, John Cirillo, Arne Nieuwenhuys, Winston D. Byblow
Journal of Neuroscience 8 February 2023, 43 (6) 1008-1017; DOI: 10.1523/JNEUROSCI.1712-22.2022
del.icio.us logo Digg logo Reddit logo Twitter logo Facebook logo Google logo Mendeley logo

Jump to section

Keywords

Responses to this article

Respond to this article

Jump to comment:

No eLetters have been published for this article.