Inhibition during response preparation is sensitive to response complexity

Ian Greenhouse, Dylan Saks, Timothy Hoang, Richard B Ivry

Abstract
Excitability of the motor system is transiently suppressed during the preparation of movement. Two
mechanisms have been hypothesized to produce this preparatory inhibition, one to facilitate
response selection and a second to prevent premature response initiation. We used transcranial
magnetic stimulation to probe changes in motor system excitability to determine whether
preparatory inhibition is sensitive to response complexity. Inhibition was greater during the
preparation of complex responses comprised of coordinated movements of two effectors that also
involved a choice between the two hands compared to when choosing between easy, single-
effector movements or when no choice was involved. In a second experiment, inhibition was
reduced when the complex responses involved sequential movements of the two effectors, and
this effect also tended to be greater in the context of a choice. The results indicate that preparatory
inhibition is sensitive to the complexity of the responses being prepared, particularly when there is
a choice between competing representations, and this may reflect the engagement of a specific
brain mechanism involved in interlimb coordination and action selection.


Introduction
Response preparation entails the transient inhibition of the motor system. Signatures of this
inhibition have been observed in studies that used transcranial magnetic stimulation (TMS) to elicit
motor evoked potentials (MEPs) during the preparatory period of delayed response tasks
(Hasbroucq et al., 1997; 1999; Duque and Ivry, 2009; Duque et al., 2010; 2012; Labruna et al.,
2013). MEP amplitudes are consistently reduced during this preparatory period relative to resting
baseline measurements. This effect has been attributed to two, distinct mechanisms (Duque et al.,
2010). The first mechanism is hypothesized to facilitate the selection of appropriate responses by
suppressing the representations of undesired responses, a process that has been referred to as
competition resolution. The second mechanism is hypothesized to reflect the suppression of the
selected response to prevent premature movement initiation, a process that has been referred to
as impulse control. Here, we set out to test whether the complexity of a response influences the
level of inhibition associated with one or both of these mechanisms.
A separate line of neuroimaging research has examined how functional patterns of laterality
within motor cortex relate to the complexity of a movement, and this line of work suggests that
inhibition associated with the competition resolution mechanism may be especially sensitive to the
level of response complexity. These studies have consistently shown that activation across many
areas of the cortex, including sensorimotor cortex, increases as a function of movement complexity
(Rao et al., 1993; Shibasaki et al., 1993; Wexler et al., 1997; Verstynen et al., 2005). Moreover, the
influence of complexity is not restricted to contralateral motor cortex, but is especially prominent in
ipsilateral motor cortex (Shibasaki et al., 1993; Hackley and Miller, 1995; Verstynen et al., 2005;
Verstynen and Ivry, 2011a). The greater recruitment of ipsilateral motor cortex associated with
complex responses may increase competition when choosing between contralateral homologous
limbs. Preliminary TMS evidence further supports this prediction. For example, MEP amplitude
increased in homologous muscles of the resting hand when the other hand was used to produce a
complex sequence of movements (Verstynen and Ivry, 2011b) or a movement in coordination with
the foot (van den Berg et al., 2011). The increased recruitment of ipsilateral motor cortex may
result from interhemispheric connections between homologous muscles (Kanouchi et al., 1997;
Kobayashi et al., 2003) or the spill-over of bi-hemispheric planning processes (Shibasaki et al.,
1993; Cramer et al., 1999; Hanakawa et al., 2005; Verstynen et al., 2005). As the complexity of
movement execution increases, greater inhibition may be required to uncouple the two hands
(Meyer-Lindenberg et al., 2002), with failures of this inhibition underlying the manifestation of mirror
movements (Verstynen and Ivry, 2011a).
Given the pronounced effects of movement complexity on cortical dynamics, both within
and between hemispheres, we set out to examine the relevance of complexity on inhibitory
mechanisms that are operative during movement preparation. Participants were required, in
separate blocks, to prepare either a simple, single-effector response or a more complex response
requiring coordinated gestures of two effectors. We used TMS to probe motor excitability during
the preparation of each type of response and compared trials in which the targeted muscle was
either involved or not involved in the planned response. We hypothesized that competition
resolution would be sensitive to response complexity, with greater inhibition observed in the non-
selected effector when the task required a more complex movement. This prediction was based on
the assumption that more inhibition would be required to negate the bilateral recruitment of the
motor pathways for complex movements. In contrast, we did not expect impulse control to be
sensitive to response complexity, and thus predicted that inhibition of the selected effector would
be similar for easy and complex responses. This prediction was based on the assumption that
restraints on movement initiation would be comparable.
In addition to our basic choice RT tasks where the required movements varied across trials
within a block, we also included simple RT blocks in which the same movement was repeated on
every trial. If inhibition of the non-selected effector reflects the operation of a process involved in
resolving response competition, then we would not expect to see diminished inhibition in this
simple RT condition, independent of response complexity.

Methods
Participants
Twenty-four healthy young adults participated in the study, 12 in Experiment 1 (22.4 3.0 years of
age, 3 female, 2 left handed) and a different group of 12 in Experiment 2 (21.4 1.9 years of age,
1 female, all right handed). Participants provided informed consent prior to the start of the study
under a protocol approved by the IRB of the University of California, Berkeley.

Task
We used a delayed response task similar to that employed in previous studies (Duque; Labruna;
see Figure 1a). Participants were seated comfortably in front of a computer monitor with their
hands palm-down on a pillow in their laps. Each trial began with the presentation of a fixation
stimulus for 100 ms followed by a blank screen for 600 ms. A cue was then presented, a bracket
that opened to either the left ) or right (, indicating that the forthcoming response would either be
performed with the left or right hand, respectively. The cue remained visible for 900 ms, and
participants were instructed to prepare the cued response during the preparatory period. At the end
of this period, an imperative stimulus was added to the display. On 81% of the trials, this stimulus
was an O inside the bracket and signaled to the participant to produce the prepared response. On
the remaining 19% of trials, the imperative was an X; for these trials, the participant was
instructed to withhold their planned response. These catch trials were included to limit premature
responding before the presentation of the imperative stimulus. The display was blanked 300 ms
after the onset of the imperative stimulus and initiated a variable intertrial interval (3000-3500 ms,
uniform distribution).
In each experiment participants completed eight blocks of 42 trials each. Four of the blocks
required making a choice between the two hands. The other four blocks did not involve a choice,
with the response hand fixed for the entire block (simple RT: two blocks left hand, two right hand).
The blocks were further divided by whether responses were easy or complex. Easy responses in
both experiments consisted of a lateral movement of the specified index finger towards the midline.
Complex responses in Experiment 1 required that the lateral movement of the index finger be
performed simultaneously with a downward movement of the pinky finger on the same hand
(Figure 1b). In Experiment 2, the complex condition required a sequence of these two gestures
rather than their simultaneous production. That is, the participant first produced the lateral
movement with the index finger and then produced the downward movement with the pinky finger.
Note that in Experiment 2, the initial gesture is essentially identical in the easy and complex
conditions. Thus, each participant completed two complex choice blocks, two complex simple
blocks (one with each hand), two easy choice blocks, and two easy simple blocks (one with each
hand).
The order of conditions was randomized across participants with the two choice blocks
within each complexity level (easy or complex) administered consecutively. Before each block,
participants were instructed how to execute the desired responses and completed ten trials of
practice without any TMS. Participants were instructed to keep their hands at rest when not
responding.

Transcranial Magnetic Stimulation
TMS was administered using a Magstim 200-2 (Magstim, Whitland, Dyfed, UK) stimulator with a 7-
cm diameter figure-of-eight coil. Stimulation was targeted at the right primary motor cortex (M1) in
order to elicit motor evoked potentials (MEPs) from the left first dorsal interosseous (FDI) muscle,
the agonist for the lateral index finger movement. EMG was also recorded from the right FDI and
both abductor digiti minimi (ADM) muscles.
To target the M1 representation of the left FDI, TMS intensity was first set to 30% of
maximum stimulator output and the isocenter of the coil was positioned approximately 5 cm lateral
and 2 cm anterior to the vertex, with the coil orientation approximately 45 degrees off of the
midline. A single TMS pulse was administered every 3 s while stimulation intensity was gradually
increased and the coil was repositioned until reliable MEPs were observed. A marker was used to
record the optimal position on the participants scalp. The resting motor threshold (RMT) was then
determined by adjusting the TMS intensity until 5 out of 10 MEPs with peak-to-peak amplitude
greater than 5 mV were observed at the optimal target location. TMS intensity during the task was
set to 115% of the RMT. RMT was 44 7 % maximum stimulator output for Experiment 1 and 45
7 % maximum stimulator output for Experiment 2.
In both experiments, TMS was either administered at fixation onset (32 trials, baseline, four
per task block) or 800 ms after the cue onset (192 trials, delay period, 24 per task block). Each
block included 14 trials without TMS. These trials were included to measure EMG onset times in
the absence of TMS. In Experiment 2, we also obtained 40 MEP measurements at rest, 20 before
the first experimental block and 20 after the last experimental block. These pre- and post-task
baseline measurements were included to assess changes in resting motor excitability during the
experimental session.

Data Analysis
EMG data were analyzed offline with Matlab using automated routines and visual inspection for the
detection of artifacts. To measure corticospinal excitability, we calculated the peak-to-peak
amplitude of the MEPs elicited by the TMS pulses. RT was defined as the interval from the onset of
the imperative to the onset of the EMG burst in the responding FDI. EMG onset was determined to
be the first data point following the imperative onset that was 3 standard deviations from the mean
of the rectified signal for the entire trial epoch and that exceeded 0.1 mV. The EMG records were
also analyzed to identify trials in which the participant failed to withhold a response on catch trials
using the same criteria to detect the presence of EMG activity. Trials in which EMG activity was
detected 100 ms prior to the MEP or preceded the onset of the imperative (go or catch) were
excluded from analysis. Individual participants mean RT was used for statistical analyses.
Mean MEP amplitudes during the delay period were converted to percentage scores,
relative to mean baseline MEP amplitudes, for statistical analysis. Paired-samples one-tailed t-
tests were used to compare MEP amplitudes in all conditions against baseline. We also compared
the easy selected and easy non-selected conditions with a t-test since this contrast provides a
replication of the main conditions included in previous studies (Duque et al., 2010; 2012; Labruna
et al., 2013). For each experiment, the MEP data were analyzed by a repeated measures ANOVA
with the factors Complexity (easy vs. complex), Condition (choice vs. simple), and Relevance
(selected vs. non-selected/irrelevant). Note that we grouped the irrelevant hand in the simple
condition with the non-selected hand in the choice condition. The RT data were analyzed by a
repeated measures ANOVA with the factors TMS (delay vs. absent), Complexity (easy vs.
complex), Condition (choice vs. simple), and Hand (left vs. right). Note that levels of the Relevance
factor in the MEP analysis and levels of the Hand factor in the RT analysis refer to the same task
conditions. We chose these different terminologies for the sake of clarity.
We also performed an exploratory analysis on the MEP data between the two experiments.
The Easy conditions in the two experiments were identical. Therefore, a mixed ANOVA restricted
to these conditions was used to assess between-group differences and/or effects of task context
given that the Complex conditions were different in the two experiments. This ANOVA included the
between-subject factor Experiment (E1 vs. E2), and within-subject factors Condition (choice vs.
simple) and Relevance (selected vs. non-selected/irrelevant). To compare between the two
different types of complex responses, we calculated a difference score (Complex Easy) for each
pair of response conditions (i.e., complex choice easy choice and complex simple easy simple
for selected and non-selected/irrelevant responses). These difference scores were submitted to the
same mixed ANOVA as employed for the Easy conditions to compare between the two
experiments.

Results
Experiment 1
The MEP data for Experiment 1 are depicted in Figure 2a. Relative to baseline, MEPs were
attenuated during the preparatory delay period. This inhibition was significant in all conditions (all
ps < 0.05) except on trials in the easy choice block in which the response was made with the right
index finger (non-selected condition), and even here, the effect was marginally significant (t(11)=-
1.6, p = 0.07, one-tailed). The pattern for the easy choice conditions was similar to that observed in
prior studies (Duque et al., 2010; 2012; Labruna et al., 2013): inhibition was greater when the left
FDI was selected for the forthcoming response compared to when it was not selected (t(11) = 2.8,
p < 0.01, one-tailed).
MEP amplitudes during the preparation of complex responses were significantly more
suppressed than during the preparation of easy responses (F(1,11) = 7.4, p < 0.05; Figure 2a).
This effect was more pronounced in MEPs measured from the non-selected/irrelevant hand than
MEPs measured from the selected hand, resulting in a significant interaction between complexity
and relevancy (F(1,11) = 6.5, p < 0.05), and also tended to be stronger in the choice than in the
simple condition, although this interaction was marginally significant (F(1,11) = 4.6, p = 0.06). As
noted above, MEP amplitudes were significantly smaller in the selected compared with the non-
selected hand in the choice condition, but this difference between the two hands (relevant vs.
irrelevant) was not present in the simple condition. This pattern contributed to a significant
condition by relevancy interaction (F(1,11) = 8.4, p < 0.05). There were no other significant main
effects or interactions.
For EMG onset time (Table 1), the repeated measures ANOVA showed a main effect of
TMS (F(1,11) = 30.6, p < 0.001), with earlier EMG onset time for TMS than no-TMS trials. This is
consistent with previous studies in which TMS facilitated response time on this task (Duque et al.,
2012; Labruna et al., 2013). There was a trend-level effect of hand (F(1,11) = 3.8, p = 0.08), with
earlier EMG onset time for right hand than left hand responses. There were no other significant
main effects or interactions. An effect of response difficulty was not expected because of the
relatively long preparatory interval and training on the complex responses. Participants responded
on 24 14% of catch trials.
In summary, Experiment 1 showed that motor excitability was suppressed more during the
preparation of complex compared to easy responses. This effect was more pronounced in the non-
selected/irrelevant hand than in the selected hand, and tended to be stronger in the choice
condition than in the simple condition. In addition, replicating previous studies, motor excitability
was suppressed more in the selected than in the non-selected hand in the choice condition.
However, a parallel comparison between the selected and non-selected hand in the simple
condition did not exhibit this difference. Collectively, these results support our hypothesis that the
motor system is more suppressed when choosing between complex responses, possibly to
counteract increased recruitment of ipsilateral M1. To follow up on these results, we wished to test
whether the observed effects were restricted to complex responses involving coordinated gestures
or whether the same pattern is observed for complex responses comprised of sequential gestures.

Experiment 2
In Experiment 2, we repeated the design of Experiment 1 using sequential gestures for the
complex responses. Sequential responses enabled us to match the initial gesture to the easy
response, i.e. a lateral index finger movement, and also addressed the question of whether the
greater suppression of MEPs observed in Experiment 1 was unique to coordinated actions. The
MEP results of Experiment 2 are presented in Figure 2b and the EMG onset times are in Table 1.
In agreement with Experiment 1 and other previous studies, MEP amplitudes during the
preparatory delay were significantly smaller than those measured at baseline (all ps < 0.05) in
every condition. As in Experiment 1, MEP amplitudes were significantly smaller in the easy choice
selected than in the easy choice non-selected condition (t(11) = 4.6, p < 0.001, one-tailed). The
MEP amplitudes measured during the task baseline did not differ significantly from resting MEP
amplitudes measured before (t(11) = -0.70, p > .05, two-tailed) or after (t(11) = -0.59, p > .05, two-
tailed) the experiment.
In contrast to Experiment 1, MEP amplitudes tended to be larger during the preparation of
complex sequential responses relative to easy responses, although this effect did not reach
statistical significance (F(1,11) = 4.5, p = 0.06; Figure 2b). MEP amplitudes across all conditions
were significantly smaller during preparation of responses involving the selected/relevant hand
compared to the non-selected/irrelevant hand (F(1,11) = 8.6, p < 0.05), and this effect was
somewhat more pronounced in the choice than in the simple condition resulting in a trend-level
interaction (F(1,11) = 3.2, p = 0.10). There were no other significant main effects or interactions.
Similar to Experiment 1, EMG onset times were earlier for TMS than no-TMS trials (F(1,11)
= 13.2, p < 0.001; Table 1). EMG onset times for complex responses were later than for easy
responses in the simple condition, but this pattern was slightly reversed in the choice condition
resulting in a significant interaction (F(1,11) = 5.1, p < 0.05). Right FDI EMG onset tended to be
earlier than left (F(1,11) = 3.7, p = 0.08), and this trend-level effect was stronger for the simple than
the choice condition, also at trend-level (F(1,11) = 4.5, p = 0.06). There were no other significant
main effects or interactions. Participants responded on 18 8% of catch trials.

Between Experiment Exploratory Analyses
As can be seen by comparing Figure 2a with Figure 2b, MEP amplitudes differed between our two
experiments, particularly for the choice condition. Therefore, we conducted two post-hoc
exploratory analyses to compare MEP amplitudes between the two experiments. The first mixed
ANOVA was restricted to only easy response trials, which were the same across the two
experiments. This test yielded a large main effect of experiment (F(1, 22) = 203.3, p < 0.001), with
participants in the second experiment exhibiting reliably smaller MEP amplitudes than participants
in the first experiment. There was also a main effect of relevancy (F(1, 22) = 9.4, p < 0.01), with
generally smaller MEP amplitudes in the selected than in the non-selected/irrelevant hand. This
effect tended to be greater in the choice than in the simple condition resulting in a trend-level
interaction (F(1, 22) = 3.8, p < 0.06). Notably, there was not a difference in baseline MEP
amplitudes between the two experiments, (t(11) = 0.3, p = 0.75).
Because the complex condition differed between the two experiments, but the easy
condition did not, we chose to compare the effect of complexity on MEP amplitudes across the two
experiments by calculating a Complex-Easy difference score, presented in Figure 3. This
difference score was significantly more negative in the non-selected/irrelevant hand than in the
selected hand (F(1, 22) = 5.3, p < 0.05). Interestingly, there was also a significant interaction
between condition and experiment (F(1, 22) = 6.8, p < 0.05). This interaction reflects the observed
pattern that complex simultaneous responses were more suppressed than easy responses in the
choice condition of Experiment 1 while complex sequential responses were less (or equivalently)
suppressed relative to easy responses in the choice condition of Experiment 2 (Figure 3 white
bars). In contrast, the pattern for the simple responses was highly similar across the two
experiments (Figure 3 gray bars). Notably, the difference score analysis did not yield a significant
main effect of experiment or any other significant effects.

Discussion
In two experiments we showed that response complexity influences motor excitability during
response preparation, especially in the context of a choice. Importantly, the easy responses were
the same across both experiments and were also a subcomponent of the complex responses, such
that an FDI muscle was involved in every response. In both experiments, MEP amplitudes
measured during response preparation were significantly reduced relative to an inter-trial baseline
in all but one condition, which was at trend-level, suggesting that motor system excitability may be
suppressed whenever a response is being prepared, even outside the context of a choice. For
Experiment 1, MEP amplitudes were smaller during the preparation of complex responses relative
to easy responses, an effect that was most pronounced in the choice condition. For Experiment 2,
this pattern was reversed; with larger MEP amplitudes for the complex relative to the easy
condition, although this effect did not reach significance (p = 0.06). A Complex-Easy difference
score for MEP amplitudes was calculated for both experiments and revealed that the effect of
difficulty interacted with whether or not the task involved a choice. Overall, and in line with our
predictions, the results suggest that complexity influences motor system excitability during the
preparation of responses, especially when a choice is involved. This result lends support to the
existence of a distinct inhibitory process involved in competition between candidate response
options. Moreover, where other studies have explored the influence of different effectors on
signatures of motor suppression during response preparation, we present the first evidence that
such suppression may be sensitive to a different behavioral dimension.
As noted above, neuroimaging studies indicate that complex or difficult movements recruit
representations in ipsilateral motor cortex (Shibasaki et al., 1993; Hackley and Miller, 1995;
Verstynen et al., 2005; Verstynen and Ivry, 2011a). This ipsilateral recruitment may contribute to
the expression of mirror movements in contralateral homologous effectors. When a choice pits
homologous effectors against each other, this ipsilateral activity may need to be suppressed to
facilitate the choice. Evidence from motor system stimulation experiments implicates a distinct
brain mechanism for carrying out this type of competition resolution, evident in the suppression of
non-selected response representations during the preparation of choice responses (Duque et al.,
2010; 2012). Based upon these separate lines of evidence, we predicted that response complexity
would selectively modulate motor excitability during the preparation of responses when there is a
competition, i.e. a choice, between homologous effectors of the two hands. This is more or less
what we observed. Unexpectedly, we also observed an interaction between this effect and the type
of complex response, i.e. simultaneous vs. sequential. This suggests that the preparation of
simultaneous and sequential responses may have opposite effects on motor excitability during the
preparatory period.
Interestingly, we also observed that MEP amplitudes were reduced significantly below
baseline during all of the simple response conditions in both experiments. This pattern suggests
that at least one inhibitory process may always be engaged during the preparation of any
response. However, it seems unlikely that this signature of inhibition results from a mechanism
involved in competition since participants were executing the same response throughout simple
task blocks. Moreover, we did not observe an effect of response complexity for the simple task
blocks in either experiment. It is possible that this pattern is representative of the impulse control
mechanism identified by Duque et al. (2010 & 2012), which was shown to influence excitability at
the spinal level as measured by the H-reflex. Future studies will be able to measure spinal
excitability to investigate whether signatures associated with the impulse control process are
present in different types of simple response conditions. Considered in conjunction with the results
from a choice condition, manipulating the complexity of the prepared response may be a novel way
of teasing apart these two putative inhibitory mechanisms.
Nevertheless, our interpretation is complicated by some unexpected results. First, the
participants in Experiment 2 exhibited reliably smaller MEP amplitudes relative to baseline than the
participants in Experiment 1. This effect was highly significant, and may represent a spurious group
difference. Alternatively, this effect may represent the influence of the context of the different
complex response conditions used in the two experiments. Importantly, despite the overall
difference in MEP amplitudes across the two experiments, the Complex-Easy difference scores for
the simple condition did not differ between the two experiments. This pattern further reinforces the
interpretation that the effect of response complexity selectively influences motor excitability in the
choice task. We also note that EMG onset time was significantly later in Experiment 2 than
Experiment 1. However, this difference in EMG onset time is unlikely to account for the interaction
across experiments observed for MEP amplitudes because the difference in EMG onset time was
not specific to only choice or simple responses. Surprisingly, in Experiment 2 the EMG onset times
were faster for complex responses than easy responses in the choice condition, but not in the
simple condition. This effect may offer a clue about the mechanisms that differentiate sequential
responses from simultaneous responses. However, this pattern is also unlikely to explain the MEP
effects because TMS was always delivered prior to imperative stimulus onset, i.e. 100 ms before
the imperative stimulus, and participants responded on relatively fewer catch trials in Experiment 2
than in Experiment 1, indicating that responses were not more likely to be prematurely executed
despite the earlier EMG onset times. Despite these complications, our results suggest that motor
excitability during response preparation is especially sensitive to response difficulty in the context
of a choice.
Several brain mechanisms are likely involved in making a choice between the two hands
and could be influenced by response complexity at multiple levels. A distributed network of brain
areas is hypothesized to support the large repertoire of human behaviors involving interlimb
coordination (Swinnen, 2002). Our results are a first step toward establishing that motor system
excitability dynamics during response preparation, possibly mediated by this network, are sensitive
to task demands such as response complexity. Some previous evidence implicates a role for
lateral prefrontal cortex (LPFC) in response selection, and this brain region may be especially
sensitive to the influence of response complexity. Disruption of the LPFC with repetitive TMS was
shown to release MEP suppression of both the selected and non-selected response
representations during the same delayed response task used here (Duque et al., 2012).
Interactions between LPFC and medial frontal cortex are believed to play an important role in
action monitoring, putatively for the purpose of correcting errors that may arise from competition
between responses (Gehring and Knight, 2000). According to current models, the medial frontal
cortex signals the LPFC to regulate action representations (Ridderinkhof et al., 2004). We propose
that the LPFC is recruited during the preparation of complex responses in anticipation of greater
action selection demands imposed by an increased likelihood of response execution errors. Future
studies may be able to explore a relationship between preparatory inhibition and the relationship to
subsequent action errors.
Furthermore, we showed that manipulating the complexity of a response alters the level of
motor excitability despite no explicit differences in the primary component action, since a lateral
flexion of the index finger was a component of every response in both experiments. This aspect of
the results suggests that the degree of motor suppression is contingent upon task goals and is not
merely an all-or-none process that occurs whenever the same effector is prepared to respond. This
finding motivates future experiments that may be able to exploit this dynamic property of motor
suppression to better understand its relationship to differences in non-motor task demands. For
example, it may be possible to determine whether or not signatures of motor suppression predict
individual differences or trial-to-trial fluctuations in behavioral performance by manipulating
dimensions of task difficulty, reward, or emotional contexts.

Conclusion
One possible function of the transient suppression of the motor system during response
preparation may be to facilitate the execution of more complex responses. Here we showed that
the level of corticomotor suppression is sensitive to response complexity, but only when choosing
between the two hands. Complexity did not influence the level of motor suppression when the
same response was executed on every trial. This pattern may reflect the selective engagement of
a previously proposed competition resolution mechanism, implicated in the elimination of a non-
selected candidate response. A role for this mechanism in intermanual coordination agrees with
other evidence indicating that the execution of complex responses engages representations in
ipsilateral motor cortex.




!"#$% '( Mean (sLd) LMC onseL Llme ln mllllseconds relaLlve Lo Lhe lmperaLlve sLlmulus.

)*+%,-.%/0 ' )*+%,-.%/0 1

Llul LMC CnseL 8lul LMC CnseL Llul LMC CnseL 8lul LMC CnseL
easy cholce no-1MS 338 (43) 319 (46) 279 (37) 282 (30)
easy cholce 1MS 293 (48) 289 (46) 233 (38) 236 (30)
easy slmple no-1MS 328 (33) 322 (38) 264 (33) 239 (43)
easy slmple 1MS 284 (63) 273 (31) 219 (40) 213 (37)
complex cholce no-1MS 336 (38) 338 (39) 271 (32) 267 (36)
complex cholce 1MS 317 (77) 307 (38) 232 (48) 239 (32)
complex slmple no-1MS 340 (32) 326 (38) 297 (64) 276 (30)
complex slmple 1MS 308 (63) 290 (43) 260 (32) 224 (38)
Figure 1. During the delayed response task a fully informative cue indicated which hand to
prepare for a response following an imperative stimulus. Transcranial magnetic stimulation
(TMS) was delivered during the fixation baseline or during the preparatory delay. Motor evoked
potentials (MEPs) were recorded from the left first dorsal interosseous muscle when that hand
was selected/relevant to the task or non-selected/irrelevant to the task (A). In both experiments
easy responses involved lateral flexion of the index finger. In Experiment 1, hard responses
involved lateral flexion of the index finger with simultaneous abduction of the pinky finger. In
Experiment 2, hard responses involved the lateral flexion of the index finger followed by the
abduction of the pinky on the same hand.

Figure 2. Motor evoked potential (MEP) amplitudes for Experiment 1 (A) and Experiment 2 (B)
are represented as a percentage of the baseline for the choice and simple conditions.

Figure 3. Hard-Easy MEP amplitude difference scores for Experiments 1 and 2 reveal an
opposing influence of difficulty for the choice (white bars), but not the simple condition (gray
bars).


References
Cramer SC, Finklestein SP, Schaechter JD, Bush G, Rosen BR. Activation of distinct motor
cortex regions during ipsilateral and contralateral finger movements. Journal of
Neurophysiology 81: 383387, 1999.
Duque J, Ivry RB. Role of corticospinal suppression during motor preparation. Cereb Cortex
19: 20132024, 2009.
Duque J, Labruna L, Verset S, Olivier E, Ivry RB. Dissociating the Role of Prefrontal and
Premotor Cortices in Controlling Inhibitory Mechanisms during Motor Preparation. J Neurosci
32: 806816, 2012.
Duque J, Lew D, Mazzocchio R, Olivier E, Ivry RB. Evidence for two concurrent inhibitory
mechanisms during response preparation. J Neurosci 30: 37933802, 2010.
Gehring WJ, Knight RT. Prefrontal-cingulate interactions in action monitoring. Nat Neurosci 3:
516520, 2000.
Hackley SA, Miller J. Response complexity and precue interval effects on the lateralized
readiness potential. Psychophysiology.
Hanakawa T, Parikh S, Bruno MK, Hallett M. Finger and Face Representations in the
Ipsilateral Precentral Motor Areas in Humans. Journal of Neurophysiology 93: 29502958,
2005.
Hasbroucq T, Kaneko H, Akamatsu M, Possama CA. Preparatory inhibition of cortico-spinal
excitability: a transcranial magnetic stimulation study in man. Brain research Cognitive brain
research 5: 185192, 1997.
Hasbroucq T, Kaneko H, Akamatsu M, Possama CA. The time-course of preparatory spinal
and cortico-spinal inhibition: an H-reflex and transcranial magnetic stimulation study in man.
Experimental brain research Experimentelle Hirnforschung Exprimentation crbrale 124:
3341, 1999.
Kanouchi T, Yokota T, Isa F, Ishii K, Senda M. Role of the ipsilateral motor cortex in mirror
movements. J Neurol Neurosurg Psychiatr 62: 629632, 1997.
Kobayashi M, Hutchinson S, Schlaug G, Pascual-Leone A. Ipsilateral motor cortex
activation on functional magnetic resonance imaging during unilateral hand movements is
related to interhemispheric interactions. NeuroImage (2003). doi: 10.1016/s1053-
8119(03)00220-9.
Labruna L, Lebon F, Duque J, Klein P-A, Cazares C, Ivry RB. Generic Inhibition of the
Selected Movement and Constrained Inhibition of Nonselected Movements during Response
Preparation. J Cogn Neurosci (September 18, 2013). doi: 10.1016/S0926-6410(96)00069-9.
Meyer-Lindenberg A, Ziemann U, Hajak G, Cohen L, Berman KF. Transitions between
dynamical states of differing stability in the human brain. Proc Natl Acad Sci USA 99: 10948
10953, 2002.
Rao SM, Binder JR, Bandettini PA, Hammeke TA, Yetkin FZ, Jesmanowicz A, Lisk LM,
Morris GL, Mueller WM, Estkowski LD. Functional magnetic resonance imaging of complex
human movements. Neurology 43: 23112318, 1993.
Ridderinkhof KR, Ullsperger M, Crone EA, Nieuwenhuis S. The role of the medial frontal
cortex in cognitive control. Science 306: 443447, 2004.
Shibasaki H, Sadato N, Lyshkow H, Yonekura Y, Honda M, Nagamine T, Suwazono S,
Magata Y, Ikeda A, Miyazaki M, Fukuyama H, Asato R, Konishi J. Both primary motor
cortex and supplementary motor area play an important role in complex finger movement.
Brain 116: 13871398, 1993.
Swinnen SP. Intermanual coordination: from behavioural principles to neural-network
interactions. Nat Rev Neurosci 3: 348359, 2002.
van den Berg FE, Swinnen SP, Wenderoth N. Excitability of the motor cortex ipsilateral to
the moving body side depends on spatio-temporal task complexity and hemispheric
specialization. PLoS ONE 6: e17742, 2011.
Verstynen T, Diedrichsen J, Albert N, Aparicio P, Ivry RB. Ipsilateral Motor Cortex Activity
During Unimanual Hand Movements Relates to Task Complexity. 93: 12091222, 2005.
Verstynen T, Ivry RB. Network dynamics mediating ipsilateral motor cortex activity during
unimanual actions. J Cogn Neurosci 23: 24682480, 2011a.
Verstynen T, Ivry RB. Network dynamics mediating ipsilateral motor cortex activity during
unimanual actions. J Cogn Neurosci 23: 24682480, 2011b.
Wexler BE, Fulbright RK, Lacadie CM, Skudlarski P, Kelz MB, Constable RT, Gore JC. An
fMRI study of the human cortical motor system response to increasing functional demands.
Magnetic Resonance Imaging 15: 385396, 1997.