Exp Brain Res (2009) 197:311–319
DOI 10.1007/s00221-009-1916-0
RESEARCH ARTICLE
Preparation and inhibition of interceptive actions
Welber Marinovic Æ Annaliese M. Plooy Æ
James R. Tresilian
Received: 26 April 2009 / Accepted: 17 June 2009 / Published online: 30 June 2009
Ó Springer-Verlag 2009
Abstract Two experiments aimed to provide an estimate
of the last moment at which visual information needs to be
obtained in order for it to be used to initiate execution of an
interceptive movement or to withhold execution of such a
movement. In experiment 1, we sought to estimate the
minimum time required to suppress the movement when
the participants were first asked to intercept a moving
target. In experiment 2, we sought to determine the minimum time required to initiate an interceptive movement
when the participants were initially asked to keep stationary. Participants were trained to hit moving targets using
movements of a pre-specified duration. This permitted an
estimate of movement onset (MO) time. In both experiments the requirement to switch from one prepared course
of action to the other was indicated by changing the colour
of the moving target at times prior to the estimated MO.
The results of the experiments showed that the decision to
execute or suppress the interception must be made no less
than about 200 ms before MO.
Keywords Human Inhibition Interception
Motor control Movement Preparation
W. Marinovic (&) A. M. Plooy
Perception and Motor Systems Laboratory,
School of Human Movement Studies,
The University of Queensland, Blair Drive,
St Lucia 4072, Brisbane, QLD, Australia
e-mail: w.marinovic@uq.edu.au
J. R. Tresilian
Department of Psychology,
University of Warwick, Warwick, UK
Introduction
Interceptive actions like hitting and catching moving
objects involve initiating movement at an appropriate
moment in time so that the hand, bat or racquet reaches a
location on the object’s path (the interception location)
coincident with the object’s arrival. Although much
research has been conducted to determine the basis for
initiating and executing simple interceptions (for reviews
see Lacquaniti et al. 1993; Merchant and Georgopoulos
2006; Tresilian 2005), very little is known about how
people abort or withhold the production of an interception.
There have been a few studies that examined the ability to
withhold responses in coincidence anticipation tasks
(Carlsen et al. 2008; McGarry et al. 2003; Slater-Hammel
1960), which resemble interceptive actions in some
respects (for review see Tresilian 1995), but no experimental studies of people’s ability to withhold interceptions
have been conducted. An ability to quickly abort execution
can be advantageous in racquet sports like tennis or badminton where a player does not want to make contact with
the ball or shuttle if it is going ‘out’. In these situations the
player has a very limited period of time within which to
determine that the ball is going out and to abort the process
of interception.
It has long been accepted that interceptive actions are
initiated when a perceived variable reaches a threshold or
criterion value (e.g. Lacquaniti and Maioli 1989; Tresilian
2005). Both behavioural (Lacquaniti et al. 1993; Smith
et al. 2001) and neurophysiological (Merchant et al. 2004;
Merchant and Georgopoulos 2006) evidence supports this
proposal. There is some controversy concerning exactly
what perceived variable is used for initiation (Caljouw
et al. 2004), but it appears to be one that carries information about the moving target’s time-to-contact (TTC) with
123
312
the interception location (Lacquaniti et al. 1993; Merchant
and Georgopoulos 2006; Smith et al. 2001; Tresilian 2005).
All existing models agree, therefore, that the production of
interceptive actions involves a motor command generation
process that is prompted into generating commands by a
trigger signal produced when the perceived variable
reaches criterion (Zago et al. 2009) and so share the basic
structure shown in Fig. 1a. There is some debate concerning the nature of the command generating process.
Some have proposed that it is a kind of motor program that
is capable of generating commands without sensory input
(Tresilian 2005; Tyldesley and Whiting 1975); others
propose that it is a kind of sensorimotor transformation that
continuously converts sensory input into motor commands
(Dessing et al. 2002; Peper et al. 1994). Although it is not
clear which is correct (Zago et al. 2009), the two proposals
both suppose that a triggering process produces an initiating signal when a perceived quantity reaches criterion.
Figure 1b shows the temporal sequence of events that
occurs when making an interception according to the basic
model structure of Fig. 1a. Command generation is initiated by a trigger signal produced when perceived TTC
Fig. 1 a Functional block
diagram of the program model
of interception. b Sequence of
events in the program model.
When the TTC of the moving
target reaches criterion (a
stimulus event) some time is
needed to process the
information and start command
generation (the visual
processing and transmission
time). c Sequence of events in a
simple RT experiment: a
warning stimulus is presented
first, followed by the IS. The
command generation that
triggers the movement is issued
once the IS has been detected
(EMD electromechanical delay)
123
Exp Brain Res (2009) 197:311–319
(assumed to be the relevant variable) reaches a criterion
value (TTCcrit). This temporal sequence may be fit into a
very small period of time: in baseball, for example, the
time between the pitcher’s release of the ball (moving
target first visible) and the hit may be less than 500 ms and
can be similar for a batsman in cricket or a tennis player
returning serve (Regan 1992; Watts and Bahill 1990).
Thus, if the interception that the player is prepared to make
is to be withheld, the suppression process must work
quickly.
Although people’s ability to withhold or abort execution
of prepared interceptions has not been explicitly studied,
the ability to withhold manual responses in laboratory
reaction time (RT) tasks has been studied extensively (De
Jong et al. 1990; Logan and Cowan 1984; Logan et al.
1984; Logan and Irwin 2000; McGarry and Franks 1997;
Mirabella et al. 2009; Mirabella et al. 2006). In the standard simple RT protocol the participant’s task is to make a
response as soon as possible following presentation of an
imperative stimulus (IS). This protocol can be modified to
study the ability to withhold responding by embedding
within a sequence of standard RT trials, a few trials in
Exp Brain Res (2009) 197:311–319
313
which a stimulus is presented to indicate that the response
should not be executed (the stop signal, SS): participants
are instructed not to respond to the IS if an SS is presented.
Typically, the SS is presented after the IS during the RT
interval (Fig. 1c shows the temporal sequence of events in
a simple RT task).
A major empirical issue investigated in RT tasks is to
determine the last moment that the SS can be presented and
for there still to be sufficient time to prevent the response
being executed. The primary purpose of the experiments
reported here is to address this same issue using an interceptive task. As in the RT protocol, a SS was used on some
trials to indicate to participants that they were not to
intercept the target, whereas on others no SS was presented
and interception was to proceed normally. Comparison of
Fig. 1b and c shows that there are certain basic similarities
between the events that occur in an RT task and those that
occur in an interceptive task. However, an important feature of RT tasks is that the stimulus events that an action is
made in response to (the IS) is under experimental control.
In an interceptive task, the action is made in response to
TTC (or similar stimulus information) reaching a criterion
and this is not something that is under experimental control. This presents a problem: how to control the timing of
the SSs so that the last moment at which there is still
sufficient time for the response to be withheld can be
determined. This was done by adopting a protocol that we
have used in previous studies of preparatory states in
interception (Marinovic et al. 2008b; Tresilian and Plooy
2006): participants are trained prior to the experimental
session to make interceptive movements that last a specific
time (the movement time, MT). This allows an estimate to
be made of when the movement will start (the time of
movement onset, MO) since the MT is known as is the time
when the target is at the interception location (see
‘‘Materials and methods’’). The SS can then be presented at
different times prior to the estimated MO time, permitting
an investigation of the latest time at which the SS can be
presented and the interceptive response withheld. The
results of the experiments presented in this report provide
estimates for the time delays involved in the processes of
inhibition and preparation of rapid interceptive actions.
Fig. 2 Plan view of the experimental setup of the interceptive task.
The bat starts a distance of 28 cm from the target’s path. Point P is the
point at which the centre of the bat meets the path of the target face.
Point Q is at the centre of the target face. The grey shaded area is the
region swept out by the bat as it is moved along its track. The time to
arrival of Q at P (the TTC) is Z(t)/V, where Z(t) is the distance of Q
from P at time t
Materials and methods
Subjects
Ten volunteers participated in the experiments and all gave
their informed consent prior to commencement of the
study, which was approved by the local Ethics Committee
of the University of Queensland. All participants reported
normal or corrected to normal vision and stated they were
right handed. Their ages ranged from 25 to 38 years.
Apparatus and task
The experimental task was to hit a moving target performing a movement of 180 ms of duration. The target was
mounted on a carriage, which was attached to a belt system
driven by computer controlled torque motor. The participants were constrained to move a bat along a straight path
above and perpendicular to the target track as shown in
Fig. 2.
The target was made of slightly abraded clear plastic
material with embedded LEDs which illuminated the target
with either a green or red colour. The target was flat and
rectangular (9 cm tall and 6.1 cm in length) and it was seen
for 2.2 s before its arrival at the strike zone in all trials
(grey shaded area in Fig. 2), and it moved with a constant
velocity of 1.5 m/s. Ambient light was dim, so that the
target was distinctive, but it did not impair vision of the
123
314
surrounding objects. Details of the experimental setup can
be found elsewhere (Marinovic et al. 2008).
Infrared emitting diodes (IREDs) were fixed to the bat
and to the carriage in which the target was housed. The
positions of these IREDs were sampled at 200 Hz during
experimental trials using an Optotrak (Northern Digital
Inc.) optoelectronic movement recording system and stored
on computer disc.
Surface EMG data were collected using bipolar preamplified electrodes with an analogue bandpass filter (30–
500 Hz) built into the preamplifiers. The two electrodes
were circular (diameter 4 mm) mounted 9 mm apart in an
insulating block (23 9 15 9 3 mm) which housed the
preamplifier. The preamplifier gain was adjustable from 24
to 2,200. Surface EMG activity was recorded from the
bellies of the right anterior deltoid (AD) and right posterior
deltoid (PD). A grounding electrode was placed on the
right acromion. During experimental trials EMG data were
sampled at 1,000 Hz by the OptotrakTM analogue-to-digital
data acquisition unit (ODAU unit, Northern Digital Inc.)
and recording was time-locked to the recording of the
IRED position data.
Design and procedures
Experiment 1
Prior to the experimental session, the participants were
trained to hit the target with a movement time of 180 ms so
that movement onset time could be estimated in advance.
During training and experimental sessions the participants
were provided with knowledge of results (KR) about their
performance. The KR informed the participants about their
MTs and maximum distance moved.
In experiment 1 participants’ primary task was to
intercept the target with the specified MT. The target was
always initially illuminated by the green (go signal) LED
throughout its trajectory towards the interception location.
However, on 50% of the trials (nogo trials), when the target
colour changed from green to red (stop signal), they were
required to halt their movements. The stop signal was
presented pseudorandomly.
The stop signal (colour change) on nogo trials could
occur at various times prior to the expected moment of
movement onset (i.e. 180 ms prior to the target reaching
the hitting location). Five different stop signal intervals
were used: 50 (SS-50), 100 (SS-100), 150 (SS-150), 200
(SS-200), and 250 ms (SS-250). To balance order effects of
task presentation, half of the participants began the
experimental session with experiment 1, and half of them
began with experiment 2. The participants performed eight
trials for each of the five conditions in which there was a
123
Exp Brain Res (2009) 197:311–319
colour change plus 40 control trials (no colour change) in
each of the two experiments (160 total).
Experiment 2
The participants’ primary task was to keep the bat stationary. The target was initially illuminated by the red
(nogo signal) LED throughout its trajectory towards the
interception location. However, on 50% of the trials (go
trials), when the target colour changed from red to green
(go signal), they were required to strike the target with the
specified MT of 180 ms. The go signal was presented
pseudorandomly.
The go signal on go trials could occur at various times
prior to the expected moment of movement onset. Five
different preparation intervals were used: 100 (PI-100), 150
(PI-150), 200 (PI-200), 250 (PI-250) and 500 ms (PI-500).
Data reduction and analysis
Experiment 1
Data reduction was performed using custom Labview
software (version 7.1, National Instruments). The position
data time series were digitally filtered by dual pass through
a second order Butterworth filter with a cut-off frequency
of 20 Hz. Movement onsets were estimated from the tangential velocity time series (derived by numerical differentiation from the filtered position time series) using the
two-stage algorithm recommended by Teasdale et al.
(1993). When an interception was required, the time at
which the target was hit as well as the temporal error (TE)
were estimated from the position time series of the bat
IRED and the target IRED. Further details of the analysis
procedures can be found elsewhere (Marinovic et al. 2008).
The EMG data obtained from each muscle were fullwave rectified, and digitally enveloped by dual-pass
through a low-pass, second order Butterworth filter with a
51-Hz cutoff (equivalent to fourth order, zero phase lag
filter with a 40-Hz cutoff).
The main measure in experiment 1 was the absence or
presence of movement. A failure to withhold a movement
was considered as such if the participant moved the bat
more than 10 mm and the EMG showed an activation of
the AD which exceeded three standard deviations from
baseline activity for more than 20 samples. To obtain
inhibition functions, the data set of each participant was
fitted with a cumulative Gaussian by using a maximumlikelihood fitting procedure. The time required to suppress
the interceptive action was defined as the point in the
inhibition function at which the probability of inhibiting
the movement was 0.5. In this experiment, we also
Exp Brain Res (2009) 197:311–319
315
analysed: Amax, defined as the highest value in the
acceleration profile. Distance moved, defined as the maximum distance the bat moved away from its initial position.
EMG amplitude defined as the peak of enveloped EMG
activity.
Effects of experimental conditions on Amax, distance
moved, and EMG amplitude were first analysed through a
one-way ANOVA with repeated measures. A Newmann–
Keuls post hoc test, P \ 0.05, was used for comparison of
the means.
Experiment 2
The variables of interest were MT, defined as the time
between movement onset and target strike. Peak acceleration (Amax), defined as in experiment 1. TE, defined as the
difference between target strike time and the time the
centre of the target reaches the interception location. If the
bat arrived before the centre of the target at the interception
location, the temporal error was early and recorded as
negative, otherwise the error was positive. For each participant the mean of the eight trials in each experimental
condition was averaged to give estimates of MT, Amax,
and constant TE (CTE) for each condition (both hits and
misses were used in the calculation of these variables). The
procedures and design for the analyses of these variables
were the same as those used in experiment 1.
Results
Experiment 1
In Fig. 3 is shown the inhibition function for two representative participants. As shown in Fig. 3, the probability
of inhibiting a prepared hitting action was virtually zero up
to 100 ms before movement onset, but rapidly increased
after this point. In fact, for all participants only one
response was completely inhibited when the stop signal
interval was \150 ms.
Since we controlled the timing of the stop signal
appearance based on the estimated movement onset time,
the mean stop signal time required to withhold the hitting
action should be equal to the point at which the inhibition
function reaches 0.5. The mean (±SD) stop signal interval
estimated with this approach across all of the subjects was
192 ms (±26.3).
As shown in Fig. 4a, on average the participants reached
the target’s plane of motion (280 mm from the starting
position) only when the time left to withhold the action was
in the order of 50 ms. A repeated measures ANOVA on
this variable revealed a significant effect of stop signal
interval, F(5, 40) = 138.52, P \ 0.0001, x2 = 0.87. Post
Fig. 3 Inhibition functions for two representative subjects (P8 and
P9). The curves are the cumulative Gaussian functions that best fit the
data. The grey lines in each panel indicate the point in the inhibition
function at which the probability of inhibiting the movement was 0.5.
Filled circles represent the proportion of inhibited trials at that
particular SS interval
hoc analysis showed that the distance moved was greater
for each decrement in the stop signal interval since all
pairwise comparisons were significant.
Figure 4b shows that Amax decreased gradually as the
stop signal interval became longer. A repeated measures
ANOVA on this variable revealed a significant effect of
stop signal interval, F(5, 40) = 97.03, P \ 0.0001,
x2 = 0.86. Post hoc analysis showed that Amax was significantly different for all pairwise comparisons, except for
those between the SS interval of 250 and 200 ms, and
between the SS interval of 50 ms and go-trials in which
Amax did not differ from each other.
Repeated measures ANOVAs on the peak amplitude of
enveloped EMG revealed that significant differences existed in AD, F(5, 40) = 23.86, P \ 0.0001, x2 = 0.62, and
PD, F(5, 40) = 5.78, P \ 0.0001, x2 = 0.22. Pos hoc
analysis on AD showed that peak amplitude was lower than
on go trials only at SS intervals of 150, 200, and 250 ms as
shown in Fig. 4c. By contrast, the post hoc analysis on PD
indicated that peak amplitude was larger than in go trials at
SS intervals of 50, 100 and 150, but not at 200 and 250 ms
as shown in Fig. 4d.
Experiment 2
Figure 5 shows MT, CTE, and Amax data from experiment
2. In Fig. 5a, the dashed line indicates the required MT
with which the moving targets should be intercepted,
whereas the dotted lines indicate the range within which
MT was considered acceptable. As can be seen in Fig. 5a,
MT decreased gradually as the preparation interval became
shorter and when the preparation interval was equal or
shorter than 150 ms the participants could not produce
123
316
Exp Brain Res (2009) 197:311–319
Fig. 4 a Mean distance moved
on nogo and go trials as a
function of SS interval. Dashed
line indicates the distance
between the initial position and
the plane of target movement,
where the target could be hit. b
Mean maximum acceleration on
nogo and go trials as a function
of SS interval. c–d Mean peak
amplitude expressed as a
proportion of mean go trial
amplitude for each muscle as a
function SS interval. c Anterior
deltoid (AD). d Posterior deltoid
(PD). The error bars show 95%
confidence intervals
MTs with the specified duration (180 ms ± 10%). The
repeated measures ANOVA on MT revealed a significant
effect of preparation interval, F(4, 36) = 27.66,
P \ 0.0001, x2 = 0.52. Post hoc analysis confirmed that
MT was greater for each increment in the preparation
interval with all pairwise comparisons being significant,
except for those between PI-100 and PI-150 in which MTs
did no differ from each other.
Figure 5b shows that the participants tended to arrive at
the interception zone late when the preparation interval was
in the order of 150 ms or less. In contrast, with preparation
intervals of 200 ms or greater, there was a tendency to
arrive at the interception zone earlier. The repeated measures ANOVA on CTE indicated a significant effect of
preparation interval, F(4, 36) = 94.34, P \ 0.0001,
x2 = 0.88. Post hoc analysis showed that CTE in conditions PI-100 and PI-150 differed from all other conditions,
whereas pairwise comparisons among conditions PI-200,
PI-250 and PI-500 showed no differences.
Figure 5c shows that Amax decreased gradually as the
preparation interval became longer. The repeated measures
ANOVA on Amax revealed a significant effect of preparation interval, F(4, 36) = 9.48, P \ 0.0001, x2 = 0.40.
Post hoc analysis showed that Amax in conditions PI-100
and PI-150 was greater than those observed in conditions
PI-200, PI-250, and PI-500, but they did not differ from
each other.
participants delaying the time of MO to increase the time
available for inhibition). For this purpose we compared go
trials in experiment 1 with conditions in which the participants had enough time to prepare their responses (PI-250
and PI-500). The variables compared were: MT, CTE and
Amax. A repeated measures ANOVA comparing MT on go
trials of experiment 1 and MT at long preparations intervals
in experiment 2 showed a significant effect of condition,
F(2, 18) = 6.32, P \ 0.01, x2 = 0.18. The post-hoc test
showed that MT at PI-500 ð
x ¼ 190:1 msÞ in experiment
2 was longer than that observed at PI-250 ð
x ¼ 177:6 msÞ
in experiment 2 and also longer that observed in go trials of
experiment 1 ð
x ¼ 177:8 msÞ; which did not differ from
each other. This indicates that the participants produced
accurately the required MT of 180 ms (±10%) on go trials
in experiment 1 since performance on this condition did not
differ from that on PI-250 in experiment 2, when the participants had a better performance in producing the
required MT than on PI-500 (see Fig. 5a). Furthermore, the
repeated measures ANOVAs revealed neither a significant
effect on CTE, F(2, 18) = 3.34, P [ 0.05, x2 = 0.10, or
Amax, F(2, 18) = 0.99, P [ 0.05, x2 = 0.01. These
results suggest that the participants were preparing to strike
the moving target, as trained, without being affected by the
high probability of inhibiting their movements on nogo
trials.
Comparison between experiments 1 and 2
Discussion
We also conducted additional comparisons between the
two experiments. These comparisons are important because
the expectation of a SS in experiment 1 could affect the
time course of preparation for the interception (e.g.
In the experiments reported here we investigated the time
course of inhibitory and preparatory processes of fast
interceptive actions. Whereas some experiments have
studied the time course of movement preparation of
123
Exp Brain Res (2009) 197:311–319
Fig. 5 a Mean movement time on go trial conditions. Dashed line
indicates the required MT. Dotted lines indicate the range within
which MT was considered to be correct (±10%). b Mean constant
temporal error in go-trial conditions. c Mean maximum acceleration
in go trial conditions. The error bars show 95% confidence intervals
interceptions during the 0.5 s prior to movement onset
(Marinovic et al. 2008b), to our knowledge no experiment
to date has investigated peoples’ ability to inhibit this type
of action. A number of experiments investigating movement suppression in RT tasks have used the ‘‘race model’’
put forward by Logan and Cowan (1984) and Logan et al.
(1984) as a tool to investigate the dynamics of inhibitory
processes (De Jong et al. 1990; Logan and Cowan 1984;
Logan et al. 1984; Logan and Irwin 2000; McGarry and
Franks 1997; Mirabella et al. 2009; Mirabella et al. 2006).
In their model, Logan and colleagues proposed that activation and inhibition evolve independently of one another
and compete for control of responding. If the excitatory
process is completed first, a response is produced. If the
inhibitory process is completed first, a response does not
occur. This description of the competition between internal
processes involved in the production and inhibition of RT
317
tasks is somewhat analogous to the triggering and inhibition processes involved in interceptions. The commandgenerating process activated by the perceived variable
(TTCcrit in Fig. 1b) reaching criterion in an interception is
analogous to the activation of excitatory processes activated by an IS in a RT task. Also, the inhibitory processes
in RT tasks and interceptions are both initiated when an
external stop signal is presented (although in RT tasks the
stop signal is presented after the IS, whereas in interceptions it can only be presented in relation to when the
response is expected to occur). Therefore, despite the
obvious differences in the protocols used to study inhibition in RT tasks and interceptions, it seems that the race
model proposed by Logan and colleagues is, at least
qualitatively, equivalent to the competition between the
triggering and inhibition processes we believe are involved
in the suppression of rapid interceptive actions. This seems
to indicate that the MT training protocol used to make our
participants initiate their movements at a particular
moment may be an alternative to study inhibitory processes
in more complex motor acts (e.g. intercepting moving
objects) in a similar manner that the race model allows the
study of inhibition in RT tasks.
In experiment 1, we sought to determine the minimum
time needed to suppress fast interceptive actions. In this
experiment, MT on go trials was produced accurately by
the participants and CTE and Amax were similar to the
values observed in experiment 2 for long preparation
intervals (C250 ms). This suggests that the participants
were not affected by the expectation of suppressing their
movements and prepared their actions as requested. As a
result, movement onset time can be estimated in advance of
the person starting to move and so the amount of time
available to inhibit this type of action can be controlled.
The results of experiment 1 showed an average time in the
order of 192 ms to inhibit a hitting action. This latency is
close to that reported for other manual responses (Logan
et al. 1984; Logan and Irwin 2000; Mirabella et al. 2009;
Mirabella et al. 2006), but greater than the values found for
the suppression of other anticipatory tasks, which seems to
be about 150 ms (Carlsen et al. 2008; McGarry et al. 2003;
Slater-Hammel 1960).
One possible account for the discrepancy between the
results obtained with other anticipatory tasks and the
results here presented is based on the requirements
imposed by the different tasks. The participants in experiment 1 were asked to produce movements of a specified
duration to intercept a moving target. In previous studies
with anticipatory tasks (Carlsen et al. 2008; McGarry et al.
2003; Slater-Hammel 1960) there was no MT requirement
and movement onset had simply to coincide with the
arrival of the target at a specific location. Since our participants had to move for approximately 180 ms after
123
318
movement onset, the critical moment for temporal estimation of the target’s arrival at the contact point was made
well before that of previous studies with anticipatory tasks.
Gray and Regan (1998) found that the variance in TTC
estimation can be expressed as a percentage of TTC.
Consequently, the determination of movement onset time
might have been temporally less precise in our task because
of longer estimations of TTC values in which the movement should be triggered. This may have required our
participants to program and store their responses in
advance (in case they overestimated the real TTC value
and had to initiate their responses sooner than expected). In
addition, whereas in previous experiments (Carlsen et al.
2008; McGarry et al. 2003; Slater-Hammel 1960) the
success in a trial depended mainly on movement onset
timing, in our task movement onset was an initial component of the task since the movement should be programmed to bring the limb to a determined place with a
specific duration. Therefore, motor preparation well in
advance of movement initiation was likely to be more
important in our task than in previous experiments with
anticipatory tasks (Carlsen et al. 2008; McGarry et al.
2003; Slater-Hammel 1960). As a result, it may have been
more difficult to inhibit an action in our experiment
because the preparatory processes were in more advanced
stages than in other anticipatory tasks previously studied.
An alternative account for the additional delay to inhibit
an interceptive action in our experiment may reside on the
utilisation of colour change to indicate that the response
should be halted.1 Pisella et al. (1998) showed that the
latency to stop an ongoing reaching action in response to a
colour change is longer than that observed to stop in
response to a location change (however, see Brenner and
Smeets 2004). Previous research investigating inhibition in
anticipatory tasks used either an acoustic stimulus (Carlsen
et al. 2008) or the sudden stop of the moving target which
indicates the time of movement onset (McGarry et al.
2003; Slater-Hammel 1960). Therefore, an alternative
hypothesis is that the additional delay we obtained was due
to a slower processing rate for colour change in comparison
to acoustic and location stimuli. Note that the two
hypotheses we have raised here, advanced preparation and
colour change, are not mutually exclusive and thus it is
possible that both may have played a role in determining
the latency to inhibit movements in our experiment. Further experiments are warranted to distinguish between
these two alternative hypotheses.
The results of experiment 1 also showed that although
virtually no response could be completely withheld when
the stop signal was presented 100 ms before movement
onset, the response produced was affected. When the stop
1
We thank one anonymous reviewer for raising this possibility.
123
Exp Brain Res (2009) 197:311–319
signal was delivered 100 ms before movement onset, the
value of peak acceleration (Amax) was significantly
reduced. Analysis of peak EMG amplitude indicated that
the prepared responses were affected by the stop signal.
However, rather than a reduction in the activation of agonist muscles (AD) at SS intervals B100 ms, there was a
significant increase in the activation of antagonist muscles
(PD). This result indicates that the stop process could not
affect the initial phase of the prepared response when
delivered 100 ms before movement initiation, but suggests
that the activation of antagonist muscles could be rapidly
increased to stop the ongoing action. When the stop signal
was presented 50 ms before movement onset, however,
there was no time to increase the activation of antagonist
muscles and consequently most responses reached the
interception zone. In fact, many responses on nogo trials at
the stop signal interval of 50 ms were carried out to
completion and 56% (±19.7) of them hit the target. These
findings suggest that about 50 ms before movement onset it
was too late to initiate any amendment to the movement
that could be effectively used during a rapid interceptive
action. The results, therefore, are consistent with a motor
program model for the control of brief interceptive actions
where visual information can only play a minor role after
movement initiation. Note that the increased activation of
antagonist muscles after movement onset refers to a process of movement suppression which is different from the
race between inhibitory and triggering processes described
in the introduction since the race had already been won by
the triggering process. In this case, the SS may have been
responsible for initiating a counteracting response intended
to stop an ongoing action as soon as possible. This counteracting response, however, seemed to be only triggered
when there was an overt action since there was no indication of increased antagonist, nor agonist, muscle activity
on successfully inhibited trials.
In experiment 2, we manipulated the preparation interval of a hitting action by presenting on 50% of the trials a
go signal which was delivered at different moments before
the expected movement onset time. The results showed that
under uncertain knowledge about the requirement to move
or not, participants could produce hitting actions accurately, in terms of movement time and the temporal error to
hit the target’s centre, with a preparation interval as short
as 200 ms. This latency is similar to that we found for the
preparation of an interceptive action in a previous study
where the participants had to prepare for one out of two
distinct movement amplitudes in a trial-by-trials basis
(Marinovic et al. 2008). The comparison between participants’ performance with preparation intervals C200 ms in
experiment 2 and that observed in experiment 1 in go trials
(no SS presented) suggests that the expectation to withhold
the action did not affect the time course of motor
Exp Brain Res (2009) 197:311–319
preparation in experiment 1. Assuming that participants’
performance would deteriorate if the go-signal was delivered after TTC information had reached the criterion value
of the initiating variable, experiment 2 also provides an
estimate of the interval in which TTC reaches criterion.
Since performance was negatively affected when the gosignal was presented after 150 ms to movement onset, we
conclude that the triggering variable was produced between
150 and 200 ms prior to movement onset. This result is
consistent with the triggering signal occurring about
150 ms prior to movement onset as we have recently
shown (Marinovic et al. 2009; Tresilian and Plooy 2006).
In summary, the results of the experiments reported
showed that the decision to perform or inhibit a rapid
interception must be made no less than about 200 ms
before movement onset. Although both preparatory and
inhibitory processes span similar durations, an advantage
of the inhibitory process is that if a visual cue to withhold
the movement is delivered, for example, about 100 ms
before movement onset, the performer might be able to
prevent the completion of an already initiated action. In
contrast, the preparatory process cannot be shorter than
about 200 ms before movement onset without a sacrifice in
performance levels.
Acknowledgments This research was supported in part by a
CAPES (Postgraduate Federal Agency/Brazilian Government) doctoral scholarship to Welber Marinovic and a grant from the Australian
Research Council awarded to J. R. Tresilian and A. Plooy.
References
Brenner E, Smeets JBJ (2004) Colour vision can contribute to fast
corrections of arm movements. Exp Brain Res 158:302–307
Caljouw SR, van der Kamp J, Savelsbergh GJP (2004) The fallacious
assumption of time-to-contact perception in the regulation of
catching and hitting. In: Hecht H, Savelsbergh GJP (eds) Timeto-contact. Elsevier Science, Amsterdam, pp 443–474
Carlsen AN, Chua R, Inglis JT, Sanderson DJ, Franks IM (2008)
Motor preparation in an anticipation-timing task. Exp Brain Res
190:453–461
De Jong R, Coles MGH, Logan GD, Gratton G (1990) In search of the
point of no return—the control of response processes. J Exp
Psychol Hum Percept Perform 16:164–182
Dessing JC, Bullock D, Peper CE, Beek PJ (2002) Prospective control
of manual interceptive actions: comparative simulations of
extant and new model constructs. Neural Netw 15:163–179
Gray R, Regan D (1998) Accuracy of estimating time to collision
using binocular and monocular information. Vision Res 38:499–
512
Lacquaniti F, Maioli C (1989) The role of preparation in tuning
anticipatory and reflex responses during catching. J Neurosci
9:134–138
Lacquaniti F, Corrozzo M, Borghese NA (1993) The role of vision in
tuning the anticipatory motor responses of the limbs. In: Berthoz
A (ed) Multisensory control of movement. Oxford UP, Oxford,
pp 379–393
319
Logan GD, Cowan WB (1984) On the ability to inhibit thought and
action—a theory of an act of control. Psychol Rev 91:295–327
Logan GD, Irwin DE (2000) Don’t look! Don’t touch! Inhibitory
control of eye and hand movements. Psychon Bull Rev 7:107–
112
Logan GD, Cowan WB, Davis KA (1984) On the ability to inhibit
simple and choice reaction-time responses—a model and a
method. J Exp Psychol Hum Percept Perform 10:276–291
Marinovic W, Plooy AM, Tresilian JR (2008a) The time course of
amplitude specification in brief interceptive actions. Exp Brain
Res 188:275–288
Marinovic W, Plooy AM, Tresilian JR (2008b) The time course of
direction specification in brief interceptive actions. Exp Psychol
(in press)
Marinovic W, Plooy AM, Tresilian JR (2009) The utilisation of visual
information in the control of rapid interceptive actions. Exp
Psychol 56:265–273
McGarry T, Franks IM (1997) A horse race between independent
processes evidence for a phantom point of no return in
preparation of a speeded motor response. J Exp Psychol Hum
Percept Perform 23:1533–1542
McGarry T, Chua R, Franks IM (2003) Stopping and restarting an
unfolding action at various times. Q J Exp Psychol A 56:601–
620
Merchant H, Georgopoulos AP (2006) Neurophysiology of perceptual
and motor aspects of interception. J Neurophysiol 95:1–13
Merchant H, Battaglia-Mayer A, Georgopoulos AP (2004) Neural
responses during interception of real and apparent circularly
moving stimuli in motor cortex and area 7a. Cereb Cortex
14:314–331
Mirabella G, Pani P, Paré M, Ferraina S (2006) Inhibitory control of
reaching movements in humans. Exp Brain Res 174:240–255
Mirabella G, Pani P, Ferraina S (2009) The presence of visual gap
affects the duration of stopping process. Exp Brain Res 192:199–
209
Peper L, Bootsma RJ, Mestre DR, Bakker FC (1994) Catching balls:
how to get the hand to the right place at the right time. J Exp
Psychol Hum Percept Perform 20:591–612
Pisella L, Arzi M, Rossetti Y (1998) The timing of color and location
processing in the motor context. Exp Brain Res 121:270–276
Regan D (1992) Visual judgments and misjudgements in cricket, and
the art of flight. Perception 21:91–115
Slater-Hammel AT (1960) Reliability, accuracy, and refractoriness of
a transit reaction. Res Q 31:217–228
Smith MRH, Flach JM, Dittman S, Stanard T (2001) Monocular
optical constraints on collision control. J Exp Psychol Hum
Percept Perform 27:395–410
Teasdale N, Bard C, Fleury M, Young DE, Proteau L (1993)
Determining movement onsets from temporal series. J Mot
Behav 25:97–106
Tresilian JR (1995) Perceptual and cognitive processes in time-tocontact estimation: analysis of prediction-motion and relative
judgment tasks. Percept Psychophys 57:231–245
Tresilian JR (2005) Hitting a moving target: perception and action in
the timing of rapid interceptions. Percept Psychophys 67:129–
149
Tresilian JR, Plooy AM (2006) Effects of acoustic startle stimuli on
interceptive action. Neuroscience 142:579–594
Tyldesley DA, Whiting HTA (1975) Operational timing. J Hum Mov
Stud 1:172–177
Watts RG, Bahill AT (1990) Keep your eye on the ball: the science
and folklore of baseball. Freeman, New York
Zago M, McIntyre J, Senot P, Lacquaniti F (2009) Visuo-motor
coordination and internal models for object interception. Exp
Brain Res 192:571–604
123