ArticlePDF Available

Abstract and Figures

The influence of a secondary task on speeded responses, and its effect on the outcome of more complex tasks has been studied in detail. However, the consequence of task interference on specific movement parameters other than speed and accuracy has been largely ignored. The current study examines how performing a secondary task impacts the drawing of an unseen shape. Without vision of the hand, 15 subjects traced a shape on a graphics tablet. The shape and cursor were projected onto a screen. The shape disappeared and the subject attempted to draw three consecutive identical shapes. In the visual single-task condition, hand positions were represented by a cursor, but the resultant drawings could not be seen; in another, there was no visual feedback. In four remaining conditions, the 15 subjects drew the previously seen shapes without visual feedback while performing a secondary task of reporting the orientation of an arrow which appeared on the screen either in random or periodic timing. Subjects indicated the direction of the arrow either verbally or manually. Shapes were analyzed for scale, error of the corner angles as compared with 90 degrees, and drift, compared to the reference shape and across conditions. In dual-task conditions, performance of the primary, shape-drawing task deteriorated with respect to location and orientation, but not with respect to the pattern and proportion aspects of the shape. Vision was important for controlling position of the drawing, and also for controlling the shape and proportion of the drawing suggesting separate mechanisms for the location of a drawing and its shape and proportion. Furthermore, we propose that internal representations are more important than proprioception in the shape aspect of drawing well-known figures.
Content may be subject to copyright.
The effects of secondary task interference on shape reproduction
Blake Cameron Wesley Martin Denise Y. P. Henriques
Received: 10 July 2009 / Accepted: 20 November 2009 / Published online: 5 December 2009
Springer-Verlag 2009
Abstract The influence of a secondary task on speeded
responses, and its effect on the outcome of more complex
tasks has been studied in detail. However, the consequence
of task interference on specific movement parameters other
than speed and accuracy has been largely ignored. The
current study examines how performing a secondary task
impacts the drawing of an unseen shape. Without vision of
the hand, 15 subjects traced a shape on a graphics tablet.
The shape and cursor were projected onto a screen. The
shape disappeared and the subject attempted to draw three
consecutive identical shapes. In the visual single-task
condition, hand positions were represented by a cursor, but
the resultant drawings could not be seen; in another, there
was no visual feedback. In four remaining conditions, the
15 subjects drew the previously seen shapes without visual
feedback while performing a secondary task of reporting
the orientation of an arrow which appeared on the screen
either in random or periodic timing. Subjects indicated the
direction of the arrow either verbally or manually. Shapes
were analyzed for scale, error of the corner angles as
compared with 90, and drift, compared to the reference
shape and across conditions. In dual-task conditions, per-
formance of the primary, shape-drawing task deteriorated
with respect to location and orientation, but not with
respect to the pattern and proportion aspects of the shape.
Vision was important for controlling position of the
drawing, and also for controlling the shape and proportion
of the drawing suggesting separate mechanisms for the
location of a drawing and its shape and proportion. Fur-
thermore, we propose that internal representations are more
important than proprioception in the shape aspect of
drawing well-known figures.
Keywords Attention Proprioception
Motor performance Dual-task Drawing
Daily activities often require the simultaneous performance
of more than one task: we walk and chew gum at the same
time, stir the vegetables while shaking salt into the potatoes
or, even where prohibited by law and common sense, talk
on a cell phone while driving a car. Most people believe
that dual-task performance decrements are negligible;
however, laboratory results show the effects of dividing
attention can be substantial, even for well-practiced,
seemingly compatible tasks (Pashler 1990). Much of the
dual-task literature is concerned with performance of a
primary motor task (such as a sequence of button presses)
concurrent with a perceptual or cognitive secondary task
(such as vigilance for the presentation of stimuli), or some
kind of mental load (such as word pairing or mental
arithmetic). In these experiments, subjects’ reaction times
and sequence errors are measured in the primary task
alone, and while performing the secondary task.
Dividing attention has direct costs for motor learning.
Taylor and Thoroughman (2007) divided attention by pro-
viding subjects with a tone discrimination task during per-
turbed reaching, and Gold and Park (2009) required subjects
to repeat a list of digits at a delay, either during learning or
recall of the building of an object such as a pinhole camera.
B. C. W. Martin (&)D. Y. P. Henriques
Centre for Vision Research, York University, 4700 Keele Street,
Toronto, ON M3J 1P3, Canada
B. C. W. Martin D. Y. P. Henriques
School of Kinesiology and Health Science, York University,
4700 Keele Street, Toronto, ON M3J 1P3, Canada
Exp Brain Res (2010) 202:65–77
DOI 10.1007/s00221-009-2112-y
For both experiments, interference was greatest during
learning, and the researchers hypothesized that some
chunking occurs at the time of encoding (Gold and Park
2009). Such automaticity allows for better resource sharing
when performing more than one task simultaneously (Brown
and Carr 1989). As a motor sequence is learned, it becomes
encapsulated into a ‘‘chunk’’ that can be later retrieved more
easily. Secondary loads on attentional, planning and motor
systems cause specific interference depending on the degree
to which automaticity has been achieved (Eversheim and
Bock 2001). There is some evidence that there is a timing
cost even for highly stereotyped movements such as con-
current saccades and reaching (Bekkering et al. 1994),
although later research by the same group identified the
intention for the action as being the source of the time cost
(Bekkering and Neggers 2002), rather than divided attention
per se. In a different set of experiments, tool grasping was
impacted by a secondary semantic task, while a tracking task
deteriorated with a secondary spatial task, showing that
certain particular sub-systems of a given secondary task may
interfere differently depending on the nature of the primary
task, which therefore demonstrates the role of cognition in
dual-task performance (Creem and Proffitt 2001). In all of
the preceding experiments, however, subjects have learned
new skills, and it is not possible to determine if the perfor-
mance cost was due to effects of late stages of learning, or
performance. Equally, the secondary task is rarely motor in
nature, but rather cognitive or perceptual.
To a lesser extent, other experiments have examined
‘real-world’’ dual-task situations which investigate the cost
to completing a goal-oriented behavior while also carrying
out a secondary task, such as golf putting while performing
an auditory search task (Beilock et al. 2002), or performing
a soccer slalom while identifying geometric shapes (Smith
and Chamberlin 1992). These studies observe the net
impact of task interference on the accomplishment of the
goal. However, neither the speeded response/error designs
nor the ‘‘real-world’’ designs reveal how the secondary task
might impact individual movement parameters such as the
scale, orientation or drift of component movements in a
primary task. In those instances of research involving
continuous performance of a dual-task, the study of the
parameters of the movements has not been the explicit
goal. Summers et al. (2008), for example, measured the
synchronization of circles drawn simultaneously with the
left and right hands while verbally responding to a random
tone. They investigated the variability in the location of a
drawn shape, but without reporting overall location biases.
They also measured proportion, but found no significant
main effect, possibly because the bimanual task was
designed to be in phase. More recently, Otte and van Mier
(2006) had children simultaneously perform tapping with
one hand and tracing a shape in a grooved track with the
other. In that instance, tapping was the primary task, and
the measures examined related to movement velocity rather
than the qualities of the movement performance. Move-
ment scale and variability have been studied for dual-task
interference in postural control and gait, particularly in
aging subjects. Not surprisingly, many researchers have
found that the introduction of an additional task increases
mean variability of center of pressure (COP), and alters
other specific movement parameters of gait such as
obstacle contact, as well as measures of the secondary task
such as reaction time and accuracy [see Woollacott and
Shumway-Cook (2002) for a comprehensive review and
discussion]. However, many aspects of posture and gait are
not under direct cortical control in contrast to hand
movements such as reaching, grasping and manipulating.
These actions require both attention and dexterity, and
consequently, may reveal greater performance decrement
under the load of a secondary task than do posture or gait.
Previous research from Henriques et al. (2004) sug-
gested that errors in the reproduction of felt shapes from
individual line segments are due to higher cognitive pro-
cesses. In one experiment, subjects used a robot manipu-
landum to ‘‘feel’’ the virtual borders of an irregular, closed
shape then reproduced the shape using the same manipu-
landum: the subjects could not see the hand or manipu-
landum. Subjects tended to draw the shape about 15%
larger than the reference shape. In terms of location, right-
handed subjects tended to draw the shapes further from
their body and to the left, and left-handed subjects drew the
shapes closer to themselves. Last, subjects tended to make
the inner angles of the shapes more regular than they
actually were, leading the researchers to hypothesize that
regular shapes may be more compactly stored in memory.
For our current study, we simplified the shapes, and added
a secondary task. Thus, we hypothesized that any addi-
tional performance decrement would be due to a compe-
tition for attentional resources during either motor
planning, or execution.
To better understand processes and mechanisms
involved in the continuous performance of a primary task,
we asked subjects to draw three reproductions of a shape
while simultaneously reporting the orientation of a series of
visual stimuli. The task of drawing quadrilaterals allowed
us to examine well-known movement sequences having
clear and specific measures for accurate completion
(straight lines, 90corners, equal-length sides and consis-
tent location): moreover, subjects did not need to learn how
to draw the shape. We hypothesized that with the sec-
ondary load, shapes would be less square both in terms of
proportion and size, and the location of the shape would
drift more than in conditions without load. Last, we varied
the method of response between verbal and manual modes.
We hypothesized that if any effect on the drawing is shared
66 Exp Brain Res (2010) 202:65–77
by both response modalities, it is likely that the effect is
attributable to the impact of perceptual processes; alter-
natively an effect in only one response modality indicates
the effect may be caused by the planning and execution of
the response modality.
Fifteen healthy right-handed subjects (10 male, 5 female,
ages 17–40 years, mean age 23.3) with normal or corrected
to normal vision, participated in six conditions discussed
below. Twelve new subjects and three from the previous
experiment (10 male, 5 female, age 17–24 years, mean age
20.5) with normal or corrected to normal vision, partici-
pated in the control conditions of responding manually or
verbally to the secondary task only. The subjects were
volunteers who gave informed consent and the experiments
were conducted in accordance with the provisions of the
York University Human Participants Review Sub-
In all conditions, subjects were seated comfortably at a
desk equipped with a Wacom graphics tablet,
(23.5 943.3 95.8 cm.) sampling at 50 Hz, connected to
a PC (Fig. 1). The center of the drawing surface was sit-
uated 30 cm from the body midline at table height. The
computer display, with a resolution of 1,270 9764 pixels,
was projected onto a screen 67 cm in front of the subject,
while a barrier on the desk occluded the subject’s view of
the room. An 8-inch-wide board just below eye level also
restricted field of vision so that subjects were not able to
see their hand on the tablet, and a circular mask on the
screen removed immediate vertical and horizontal orthog-
onal references (Williamson and McKenzie 1979). Custom
software was designed to provide the experimental condi-
tions and collect the data. For the manual and verbal
responding dual-task paradigms, subjects responded,
respectively, on mouse buttons with the left hand or their
spoken answer was recorded by an experimenter.
All subjects participated in six separate conditions. Two
conditions examined the ability to draw a shape with or
without vision of the cursor. These two single-task condi-
tions were performed without any additional load and
allowed us to isolate the effect of vision in the single-task
condition. Four additional dual-task conditions were
performed both with the primary task of drawing a shape
without vision of the cursor, and a secondary task of
identifying the orientation of a visually presented arrow.
Single-task conditions
In the single-task, no-cursor (SNC) condition, the subject
traced a template shape, then completed a second ‘‘drawing
phase’’ without any vision of the reference shape or the
cursor or hand: the screen displayed only white pixels. The
second, single-task, cursor, condition (SC) was identical to
the SNC condition, except that the cursor was visible
during the drawing phase. For each condition, 32 trials
consisting of one trace and three drawings were drawn, for
a total of 128 repetitions. For all subjects the SNC
Fig. 1 Experimental apparatus and display of one template shape. On
a graphics tablet, subjects traced a square or diamond displayed on a
vertical screen. As the shape was drawn, an arrow oriented either to
the left or right was displayed either at a fixed time interval or at a
random time during the drawing. A horizontal bar occluded vision of
the hand, and the template shape disappeared. Shapes were either
494or898 cm, and appeared either centered on the screen or
shifted 4 cm to the right
Exp Brain Res (2010) 202:65–77 67
condition was provided first, since deprived of any feed-
back of performance, it is quite unlikely that any learning
took place. For all subjects SNC and SC conditions were
completed on separate days. In the drawing phase, the
cursor condition allowed us to examine the subjects’ per-
formance replication of a remembered shape with vision of
the end effector, while the single task no-cursor condition
allowed us to view replication of a shape relying only on
efference copy, proprioception, and haptic feedback.
Dual-task conditions
For the dual-task conditions, all subjects performed the
tracing phase with vision of the cursor, as before; however,
during the drawing phase without vision of the drawing
cursor, subjects were presented with arrows displayed
randomly on the screen, within an area of 19.8 96.4 cm,
centered on the middle of the screen. Each arrow was 3 cm
in length with a head measuring 0.5 cm across and com-
prising 0.5 cm of the total arrow length. The arrows were
tilted either to the right or left of vertical by 3,5or 10.
Subjects were told that the most important task was to
accurately and precisely reproduce the reference shape.
Arrows were presented either at fixed (timed) or random
intervals. Unlike PRP experiments, where the timing of the
stimuli is carefully manipulated to influence the timing of a
speeded response (Pashler 1994), we used fixed and ran-
dom intervals to vary the difficulty of the perceptual task
on correctly identifying arrow orientation. Thus, we were
able to determine whether performance deficits in the pri-
mary task were attributable to attending and perceiving the
arrows, or attributable to verbal versus manual responding.
In the timed conditions, we presented arrows after a
1,000 ms inter-stimulus interval. In the random timed
conditions, the inter-stimulus interval was random,
between 500 and 1,500 ms. In either instance after the
interval, the arrow was displayed until the subject
responded whether the arrow was tilted to the right or left,
or a for maximum time of 2,000 ms: following a response
or maximum presentation time the arrow disappeared.
After the next inter-stimulus interval, a new arrow
appeared in a new location with a different degree of tilt.
Correct and incorrect answers were recorded, as well as a
failure to respond within the 2,000-ms display time. Sub-
jects were continuously provided with new arrows after
every response until they indicated the end of the primary
drawing task by moving the stylus to the bottom of the
tablet. As a consequence, some subjects responded to more
arrows and others to fewer.
Subjects responded to the secondary task while contin-
uing to draw reproductions of the shape. In the manual-
response conditions the responses were manual, with the
subject using the left hand to click a mouse positioned
beside the graphics tablet, clicking left or right buttons to
indicate the corresponding arrow-tilt. In the verbal
response conditions, subjects spoke the words ‘‘right’’ or
‘left’’ to an experimenter who then manually responded
using the mouse. In this condition, the subject’s left hand
was placed in the lap just below the desk.
Conditions were counterbalanced across subjects with
respect to timed and random arrow presentation and verbal
and manual responses, giving four dual-task combinations:
timed arrows, manual response (DTM); timed arrows,
verbal response (DTV); random arrows, manual response
(DRM); and random arrows, verbal response (DRV).
The appearance of a blue dot (0.5 cm in diameter) on the
screen indicated the start position for each trial in all
conditions. Subjects positioned the stylus so that the point
of the cursor was within the dot. After 1,000 ms, a quad-
rilateral with sides of either 4 or 8 cm appeared oriented as
a square or diamond. As drawn on the tablet, the center of
the drawn shape was approximately 30 cm from the sub-
ject’s sternum. In order to maintain the subjects’ interest,
the shapes were presented pseudo-randomly with the fol-
lowing combinations of variables: either centered on the
display and midline of the body or offset 4 cm to the right;
either 4 or 8 cm, in either square or diamond orientation.
All eight combinations were randomly presented in each
block of eight trials. In the centered condition, all shapes
were centered on the same point, regardless of size or
orientation, meaning they each had a different starting
point. The same was true for all right-shifted shapes. We
varied the size, orientation and start location of the shapes
to maintain subject interest, and presented these varied
stimuli in a pseudo-randomized, counterbalanced fashion.
The upper left or topmost corner of the square or diamond
(respectively) was given as a starting point, and a clock-
wise drawing direction was arbitrarily chosen to facilitate
programming and analysis of the experimental data.
Subjects used a computer cursor to trace the outline of a
shape which was projected on a screen. This gave imme-
diate visual feedback regarding performance on the tracing
since subjects could compare the position of the cursor
against the reference shape, although no visible path was
displayed on the screen. The tracing phase provided the
template shape for each trial in all conditions: these data
were not analyzed, but are included in some figures to
provide an index of baseline performance. At the end of the
trace phase, subjects paused, and after 1,000 ms, the ref-
erence shape disappeared. The pause allowed for a clear
separation of trace and drawing phases, facilitating both
stimulus presentation and data collection. Subjects then
made three additional continuous squares or diamonds, as
68 Exp Brain Res (2010) 202:65–77
much like the trace shape as possible during a second
‘drawing phase’’. Subjects were instructed in advance to
make an additional shape if they were uncertain whether
they had made two or three cycles.
Secondary task control
In order to determine whether performance in the main
drawing task affected the response rate to arrow-tilt in the
dual-task (mean number of responses per subject, per
condition =219.18 ±78.50, SD), we ran a separate group
of subjects who performed 300 trials of the arrow task only,
(no primary drawing task) using the same experimental
Data analysis
Data points were smoothed using a first order, low-pass
Butterworth filter, and line paths for each shape were
selected using a Matlab custom-written GUI, and verified
and corrected by hand when computer-selected corners did
not correspond with the corners of the drawn shape. Sta-
tistical analysis was performed in SPSS. We assessed
drawing performance using measures of size as determined
by area, drift of the shape as determined by displacement of
the center of area over three repetitions (left/right, up/
down), mean rotation of the shape (tilt), inner angle error
(squareness), and movement time. These measures were
compared to those for the reference shape and with respect
to those for the preceding shape in each drawing phase.
All data were compared using a 2-way within-subjects
repeated measures analysis of variance (RM-ANOVA)
comparing 6 conditions 93 cycles. By comparing the SC
(single task, cursor) condition against the SNC (single task,
no cursor) condition, we were able to determine to what
extent vision of the cursor aided in single-task perfor-
mance. The comparison of the SNC condition against the
dual-task conditions allowed us to examine the effect of a
secondary task on the primary task, both in the absence of
vision. The ‘‘trace’’ phase of each trial was only used as a
template, and is not included in any analysis. Greenhouse–
Geisser correction was used where appropriate on the
repeated measures analyses, and a modified Bonferroni
correction was applied to all post hoc paired comparisons.
The area of each shape was calculated using the formula
A=(Rx19y2xn 9y(n?1)) -(Ry19x2yn 9
x(n-1)))/2, to calculate the area of any polygon using
Cartesian coordinates: this formula was used to allow for
the inclusion and influence of every data point on the area
of the drawing, regardless of its concavity or convexity
relative to a straight line. We then calculated the ratio of
area of the drawn shape compared to the ideal shape. A
perfect sized shape would have a ratio of 1. Inner angle
error is the angle between two adjacent drawn lines sub-
tracted from 90 (the ideal shape), with positive values
indicating an obtuse inner angle, and 0 representing a 90
degree inner angle. To measure the tilt of the drawn shape,
we calculated the rotation of each line segment relative to
the line’s origin, and calculated the mean rotation for each
shape. To measure drift in the shape reproduction, we
calculated the center of area for each shape, and computed
this two-dimensional location relative to the two start
position centers for the shapes.
For certain trials in load conditions, subjects completely
missed the corner of a shape and proceeded to the next
corner (Fig. 9a, shows a typical missed-corner trial from
one subject). These trials were omitted from the main
analysis. The number of missed corners were summed and
averaged for each subject and condition. One sample ttests
(two-tailed) revealed the occurrence of missed corners to
be reliably different from zero in all load conditions except
DTM (p[0.1), (DTV: t (14) =2.188, p\0.05; DRM:
t(14) =2.571, p\0.02; DRV: t(14) =2.311, p\0.04).
We analyzed these ‘‘missed corner’’ trials for the effect of
angle of the stimulus arrow, as discussed below.
Responses to the arrow orientation on the secondary task
were summed as correct, incorrect or missed for each
subject and condition. ttests were used to compare them
with the control condition, and correlations were performed
to determine if the number of arrow responses impacted
drift, squareness, tilt or movement time.
Figure 2illustrates changes in the size (as measured by
area) of the drawing, pictured as a gray square over a
wireframe representing the ideal size, normalized across
small and large sizes (F
=8.2, p=0.002, e=0.385).
When subjects had vision of the cursor and no load (Fig. 2,
single task, cursor condition, SC, they drew shapes about
33–60% smaller than all other conditions (p\0.013) (as
indicated by the relatively larger shaded shapes in Fig. 2)
and also 14% smaller than the ideal shape (p\0.001, one
sample t-test). The DRM (dual-task random manual) con-
dition (top, right) was drawn larger than either the SNC
(single task, no cursor) or DTV (dual-task, timed verbal)
conditions, and the DRV (dual-task, random verbal) con-
dition (bottom, right) was also larger than the DTV con-
dition (all p\0.016). Vision of the cursor (in the cursor
condition) helped subjects to draw shapes closer in size to
the reference shape, and the random presentation of the
arrow stimulus caused subjects to draw shapes approxi-
mately 6–26% larger than the timed load and no-cursor
conditions, perhaps indicating that performance was
affected when vigilance needed to be greater.
Exp Brain Res (2010) 202:65–77 69
Subjects tended to draw the shapes below the location of
the ideal shape. To quantify this, we calculated the location
of the shape as the center of the drawing as defined by the
centroid of the endpoints of each line segment, and found
that drift was not significant between subsequent cycles of
the drawing within a single trial (i.e., between first and
second repetition), but only became significant over
cumulative repetitions within a single cycle (between first
and third repetitions). These centers, averaged across rep-
etitions, trials and subjects, are shown for all conditions by
the squares in Fig. 3, with the origin of the axes repre-
senting the center of an ideally drawn shape, and down
representing locations closer to the subject. The black
square represents a shape drawn with view of the cursor:
even with sight of the cursor, subjects drew the center of
the shape with a downward drift (p\0.0001). All other
conditions without vision of the cursor were drawn sig-
nificantly lower than the cursor condition (F
p\0.001, e=0.517, post hoc comparison p\0.001), but
there was no effect of load. In other words, hiding the
cursor caused increased downward drift (closer to the
subject) compared to the cursor condition, although no
additional downward drift was caused by the secondary
task. The DTV condition, represented by the grey circle in
Fig. 3, drifted more to the right than all other conditions (5,
70) F=2.8, p\0.041, e=0.708, post hoc p\0.015)
except the SNC (white square) and DRV condition (black
Shapes not only drifted more without vision of the
cursor and with the load of a secondary task, they also
became more rotated around their central axis. The black
bar in Fig. 4a shows that when subjects had vision of the
cursor, the shapes were less rotated than for all other
conditions except DRV. (F
=3.2, p\0.02, e=0.842,
paired comparisons p\0.014). Figure 4b shows that
shapes became more rotated after the first drawing cycle
=17.29, p\0.0001, e=0.576, paired compari-
sons p\0.001); however, the third cycle is not more
rotated than the second. It is not possible with such few
cycles to speculate if this lack of change between the
second and third cycles represents an asymptote. The shape
and proportion aspects of the drawing were also impacted
by lack of visual feedback. Subjects drew shapes which
were less distorted, with corners closer to 90when they
could see the cursor (F
=12.7, p\0.001, e=0.648,
paired comparisons p\0.001). Removing vision of the
cursor led to distortion of the shape, however, adding an
additional load did not as shown by the black bar in
Fig. 5a. Also, Fig. 5b shows there was a main effect of
Fig. 2 The scale of the drawing as represented by area across
conditions. The ideal, normalized area of shapes is represented by the
black wireframe, while the actual performance is shown by the grey
quadrilaterals. Error bars represent the SEM averaged across
participants. Significant differences are shown with an asterisk.
sc single task, cursor; snc single task, no cursor; dtm dual task, timed
manual responding; dtv dual task, timed verbal responding; drm dual
task, random manual responding; drv dual task, random verbal
Fig. 3 Average drift of the drawing across three repetitions for all
subjects (in cm). The squares represent the center of the drawn
shapes, averaged across subjects and trials, with the intersection of the
axes representing the ideal center of the drawn shape. Results of all
subjects comparing the single-task cursor condition (black) and single
task c no-cursor condition (white) against manual (triangles) and
verbal (circles) responding; dark grey represent fixed arrow timings,
and the light grey shapes represent random arrow timings. sc single
task, cursor; snc single task, no cursor; dtm dual task, timed manual
responding; dtv dual task, timed verbal responding; drm dual task,
random manual responding; drv dual task, random verbal responding
70 Exp Brain Res (2010) 202:65–77
cycle, with shapes becoming more distorted after the
completion of the first cycle but no more so after two
cycles. (F
=14.4, p\0.001, paired comparisons,
On average, subjects took about 1.2 s to draw each side
of the shape (i.e. each line), but this movement time varies
across conditions (F
=6.91, p\0.0001, e=0.527),
for example Fig. 6a illustrates that movement time of the
SNC and DTM conditions (white and light grey bars,
respectively) were drawn around 300 ms more slowly than
either the timed verbal or random verbal conditions
(p\0.002). In Fig. 6b, each successive cycle can be seen
to be drawn more quickly than the preceding
=34.85, p\0.0001, comparisons p\0.014).
These movement times may seem long in comparison with
transport times of *1 s for both 15- and 30-cm reaches in
studies such as that by Rand et al. (2006), but it must be
remembered that in the present experiment subjects are
drawing rather than reaching. Thus, they have the con-
straint of making straight paths that correspond to the
template shape and also respond to a secondary task.
Secondary task
We compared the performance of the secondary task of
correctly identifying arrow orientation alone against its
performance concurrent with the primary task, considering
correct, incorrect and missed responses (Fig. 7). For the
DTV condition, there were more correct responses and
fewer missed responses for the secondary task alone (sec-
ond row left, grey and white wedges, respectively). For the
DRV condition, there were more correct responses and
fewer incorrect responses for the secondary task alone
(bottom row left, grey and black wedges, respectively; all
p\0.001, two-tailed). Still, performance of the detection
task was very good in both dual-task verbal response
conditions at 92.5% for DRV and 85.7% for the DTV
Fig. 4 The rotation of drawings around a central axis. Bars show the
mean rotation, averaged across trials and subjects, with taller bars
being more rotated. aMean rotations across each condition. bMean
rotations across cycles. Error bars represent the SEM averaged across
participants. Significant differences are shown with an asterisk.
sc single task, cursor; snc single task, no cursor; dtm dual task, timed
manual responding; dtv dual task, timed verbal responding; drm dual
task, random manual responding; drv dual task, random verbal
Fig. 5 The inner angle error, which is the difference between the
ideal angle (90) and that drawn by the subject, averaged across trials
and subjects. The grey horizontal line shows the inner angle error for
the trace condition, when subjects could see the drawing template.
Mean inner angle error for the drawing are shown across conditions
(a) and cycles (b). Error bars represent the SEM averaged across
participants. Significant differences are shown with an asterisk.
sc single task, cursor; snc single task, no cursor; dtm dual task, timed
manual responding; dtv dual task, timed verbal responding; drm dual
task, random manual responding; drv dual task, random verbal
Fig. 6 Time required to draw a single side of a shape, averaged
across subjects and trials. Movement times in milliseconds across
conditions (a), and cycles (b). Error bars represent the SEM averaged
across participants. For asignificant differences are shown with an
asterisk, for b, each cycle is significantly different. sc single task,
cursor; snc single task, no cursor; dtm dual task, timed manual
responding; dtv dual task, timed verbal responding; drm dual task,
random manual responding; drv dual task, random verbal responding
Exp Brain Res (2010) 202:65–77 71
compared to a chance level of 50%. In all other conditions
detection of orientation in the dual-task condition was
equal to performance in the secondary task alone.
Because subjects were free to respond to as many arrow
stimuli as they wished, it was possible that the number of
responses may have influenced the various metrics. Cor-
relations were not significant except between the number of
responses and the measure of movement time, where we
found r[0.93 for all load conditions (data not shown),
indicating that an increased number of responses was
coupled with an increased movement time.
As one might expect, we found that subjects typically
coordinated their response to the arrow-tilt so that they
responded more often as they were approaching a corner of
the drawing; in other words when the velocity of the
drawing action was likely to be slowest, since subjects
briefly came close to stopping at each corner. This is shown
in Fig. 8where we plot the number of responses in the
secondary task relative to the distance to the nearest corner.
Figure 8a shows the responses counted into 20-ms bins.
The vertical line represents the nearest corner. The same
data is also shown in Fig. 8b, plotted against an idealized
corner on a shape with sides that would take the mean time
of 1,215 ms to complete. Most responses fell into the two
bins immediately before the corner, with the third most
populous group occurring immediately after the corner.
In a small but significant (p\0.006, ttest) number of
trials (140 trials, which is 7.2% of all trials) subjects missed
drawing one of the corners: although these trials were
excluded from the main analysis, we examined the features
of these trials. Figure 9a shows a typical trial in which a
subject completely missed a corner and proceeded to the
next corner in the shape. In these missed-corner trials, we
analyzed the remaining corners of the square to see if there
were any differences between the end locations for corners
that were before and after a missed corner and corners that
did not flank a missed corner. Separate one-way ANOVAs
were performed on the left/right and up/down coordinates
and no differences were found between the means or
variabilities of corners flanking a missed corner (pre- and
post-corners in Fig. 9b) and corners of shapes where there
were no missing lines in any of the conditions (normal
corners in Fig. 9b). In other words, subjects located corners
before and after a missed corner with both the same
accuracy and precision as with normal corners. It is as
though they had never missed the corner at all.
We considered that arrow-angles closer to vertical may
have caused subjects to miss corners due to additional
resources needed to identify the less oblique arrow-angles
(overall, subjects were slightly less accurate for 3arrows,
=8.23, p\0.0001); however, all angles occurred
before missed corners with the same frequency (Chi-
square =1.41, df =3, p\0.7). Also, subjects were less
Fig. 7 Mean of correct, incorrect and missed responses to presented
secondary task arrow plotted as a percentage of the total number of
arrows presented. The left group represents the secondary task control
condition in which participants responded to arrow orientations
without drawing, and the right represents arrow responses within the
dual task conditions. dtm dual task, timed manual responding;
dtv dual task, timed verbal responding; drm dual task, random
manual responding; drv dual task, random verbal responding
Fig. 8 a Responses to the arrow were placed into 20-ms bins,
according to their appearance relative to the closest corner drawn by
the subject in the shape-drawing task. The vertical line (0 corner)
shows the timing of the corner relative to the arrow responses. bThe
same data shown on an idealized shape with the length of each side
representing the average time taken to draw a side (1,215 ms)
72 Exp Brain Res (2010) 202:65–77
likely to miss drawing the first line (0.8% compared to
change rate of 8.3%) or last (5.4 versus change of 8.3%)
than ones in between, (93.8 versus change rate of 83.33%,
Chi-square =9.2, df =2, p\0.01). In other words, sub-
jects were less likely to miss drawing the first or last line of
a drawing than one in between.
In this study, we examined the effect of a secondary load
on a continuous drawing task. Specifically, subjects had to
reproduce a square shape while indicating the orientation
(left or right) of briefly displayed arrows. Indicating the
orientation of arrows manually or verbally allowed us to
identify whether perceptual or motor demands of the sec-
ondary task caused deterioration in performance of shape
drawing. We found that the addition of this secondary load
affected the topocinetic aspects (scalar and positional
aspects of a movement), but not the morphocinetic (shape
or pattern) aspects of the primary task. Overall, subjects
were reasonably good at drawing squares and diamonds,
although the shape and placement of the drawings did
deteriorate in some aspects when they drew these shapes
without visual feedback (no cursor) and when performing
the secondary task. Without vision of the cursor, shapes
were drawn less square and more rotated than with vision
of the cursor. The addition of a secondary task did not
worsen this decrement. During concurrent performance of
the secondary task, there were changes in the size and
location of the drawn shapes compared to the single-task
no-cursor condition. Regarding the placement of the drawn
shapes, subjects drew shapes below the reference shape,
especially when made without the cursor, and slightly more
so (though not significantly) in the load conditions. Shapes
were drawn more to the right for the verbal responding
conditions. In the load conditions, subjects sometimes
missed a corner, but this did not affect the location of
preceding or following corners.
Vision, attention and drift
The impact of the secondary task on the location but not on
the shape of the drawing suggests that in a shape-drawing
task, those parameters may be processed separately. This is
consistent with findings by Brown et al. (2003) on a sim-
pler point-to-point reaching task and by Zelaznik and
Lantero (1996) in a circle drawing task. There is mounting
evidence that position sense is processed separately from
dynamic movement, suggesting separate mechanisms for
the regulation of limb position and shape parameters in
primary and/or supplementary motor cortices (Proske
In terms of the position aspects of the movement, even
with vision of the cursor there was significant drift of the
stylus toward the subject across repetitions within each
trial. Drift became worse without vision of the cursor;
however, in the up/down direction we found no effect of
the secondary task. Most previous studies reporting drift
use more than 50 repetitions without visual feedback, while
in our experiment there were only 3 repetitions for each
trial, and it is possible that with more repetitions we would
have shown an effect of the secondary task on drift. Our
starting location represents a normal workspace position
for writing and typing tasks and is a common starting
position in other experiments: (15–45 cm, Henriques and
Soechting 2005; 30 cm, Henriques et al. 2004; 30 cm,
Klatzky 1999). These studies did not report any drift, in
contrast with Brown et al. (2003) who found a lateral drift
from the starting location when the hand was started at
approximately 30 cm from the subject, and movement
speed was higher. Alternatively, the complexity of drawing
a square may be so great that even with vision of the
cursor, but no external reference point, the near/far position
component could not be maintained.
Drawings also drifted to the right in verbal responding
load conditions. A possible explanation is that motor
planning and execution of speech, rather than perceptual
mechanisms, competed with resources for accurate place-
ment of the shape. In other words in the verbal conditions,
only the manner of response changed so we can conclude
that conflict over synergies at the level of motor production
caused the disruption. A second possibility is that the
placement of the left hand in the manual responding con-
dition ‘‘anchored’’ the right hand. This explanation is not
supported since there was no rightward drift in either the
Fig. 9 a A typical drawing in which a subject completely missed
drawing a corner due to load from the secondary task. bThe
intersection of axes shows the ideal location of a corner. The three
overlapping ellipses show the variance for normal corners, corners
before subjects missed a corner (dotted line), and corners after
subjects missed a corner (dashed line). The overlapping dots show the
mean values (in cm) for those same corners
Exp Brain Res (2010) 202:65–77 73
cursor or no-cursor conditions, and in each of these con-
ditions the left hand was not in the workspace but on the
knee, immediately below the workspace. It might also be
posited that a challenge to working memory may be
responsible for the proprioceptive drift. A study by Des-
murget et al. (2000), however, suggests that variable delay
periods, even up to 20 s between target presentation and
reaching, do not on average affect proprioceptive drift
when vision of the hand is occluded. In the present
experiment, there is only 1-s delay between the termination
of the trace phase and the beginning of the drawing phase,
and the mean time for the drawing phase is less than 4 s.
Following the presentation of an arrow, subjects fre-
quently drew two or more additional sides of the quadri-
lateral before responding to the arrow task. At that point,
they responded coincidentally at the point of minimum
drawing velocity, most frequently within 20 ms of starting
a new line. Subjects apparently ‘‘buffered’’ the response to
the secondary task, holding it in memory then integrating
its execution with a change of direction in the primary
drawing task. The prefrontal cortex (PFC) is a good can-
didate for a possible site of conflict causing a performance
decrement in our verbal task as it has been implicated in
movement planning and execution (Bullock 2004; Aver-
beck et al. 2003a,b). Sohn et al. (2000) demonstrated
functional connections between the superior posterior
parietal cortex, (important for movement guidance) and the
PFC, particularly when subjects had foreknowledge of the
action to be performed. Miller and Cohen (2001) argue that
the PFC modifies and integrates behavior based on ‘‘rules’
in concert with extensive interconnections with sensory,
association and motor cortices, as well as connections with
sub-cortical structures, such as the thalamus and basal
ganglia, and provides control for top–down behavior.
Furthermore, they suggest that the PFC maintains a
response in memory until it is executed. If the PFC is
responsible for the performance decrement, why is only the
verbal response modality in our task affected? The answer
may lie in an interconnected structure.
The PFC is known to have extensive interconnections
with the pre-supplementary movement area (pre-SMA)
(Bates and Goldman-Rakic 1993), which is implicated in
planning of internally guided movements (Deiber et al.
1996), especially movements prepared in advance as
compared to speeded response-type movements (Krams
et al. 1998). Gowen and Miall (2007) also found increased
activations of the pre-SMA in an internally generated
shape-drawing task. Additionally, the pre-SMA has been
found to be involved in internally cued speech production
(Tremblay and Gracco 2009). The verbal responding con-
dition of our load task required movement preparation,
internally generated drawing, geometric shape production,
and internally cued speech production. Given the
coincidence of these activities, a possible candidate for
aiding in drift control of the primary movement task is the
pre-SMA. In the Tremblay and Gracco study (2009), they
found no effect of attention on the pre-SMA; however,
their high attention task required greater semantic pro-
cessing, but did not actually require selective attention
among distractors nor did it divide attention among pos-
sible targets or tasks. This internal cueing for speech
preparation and response selection attributed to the pre-
SMA would also explain the higher incidence of missed
and incorrect responses for the DTV and DRV conditions
(respectively, Fig. 7). Equally, while some literature finds
specific slowing of a primary tapping task due to story
telling or recitation of a word list (Hiscock and Chipuer
1986) or reading (Hiscock et al. 1989), the movement times
in our dual task verbal conditions were quicker under the
verbal responding conditions than either the single task, no
cursor or dual task timed manual responding conditions.
This enigmatic result in the present study may be the
consequence of choosing and preparing a verbal response:
in the aforementioned studies the response was previously
selected. We propose that when subjects concurrently
performed the secondary task, performance suffered addi-
tionally only for the verbal responding conditions because
of the competition for resources in the distributed pro-
cessing of the pre-SMA.
Vision, attention and shape
The shape and proportion components of the quadrilateral
became worse without vision but no additional deteriora-
tion was observed when load was applied: the shapes did
not become less square with the secondary task. Even when
a corner was missed (which occurred only rarely and only
in the load conditions), subjects were able to find the fol-
lowing corners with remarkable fidelity. It seems that, even
when subjects were so distracted by the secondary task that
they entirely missed a corner, the brain still maintained an
accurate map of the relative locations of the remaining
points: the subjects had no vision of the hand or end
effector, and were therefore relying either on propriocep-
tive information, and/or an internal representation of the
shape. It is possible that if we had used a secondary
movement requiring the movement of the elbow or
shoulder as opposed to just the index and middle finger,
there may have been an effect on the shape components.
Separate mechanisms for location and shape parameters
In the present study, we observed drift of location but
preservation of pattern and proportion aspects of drawing
movements made under an attentional load. A similar
effect has been observed previously under various
74 Exp Brain Res (2010) 202:65–77
conditions aimed at challenging performance through
removing vision, ischemic compression, vibrating the ten-
dons, or observing patients with complete afferent neur-
opathies. Vision plays an important role in limb
localization, but less so in maintaining the pattern and
proportion of the shape. We found that subjects drifted
downward when vision was removed, but there was no
alteration of the pattern and proportion aspects of the
shape. Similarly, in another study using repetitive circle
drawing, researchers found that when vision was removed,
the spatial components of the task were altered in terms of
the size and location, but the shape and roundness of the
figure were maintained (Zelaznik and Lantero 1996).
In order to degrade proprioceptive information during
a letter-drawing task, Laszlo and Bairstow (1971) applied
a blood pressure cuff to subjects’ arms. They found that
the shape and proportion aspects of the letters were not
maintained, concluding that proprioception is necessary
for the structural components of a motor task. However,
Kelso et al. (1975) found that during ischemic com-
pression, impairment of motor function occurs before
loss of proprioceptive information, making it likely that
both motor function and proprioception were degraded.
Likewise, when another group of researchers vibrated the
tendons to disrupt proprioception, they found that the 16-
cm circles were drawn smaller and the location also
drifted relative to un-vibrated control conditions (Ver-
schueren et al. 1999). The circles were drawn consis-
tently wider than tall in all conditions, indicating that
although distorted overall, pattern and proportion infor-
mation was preserved.
This preservation of form despite drift is also observed
in patients with total sensory neuropathy. In a deafferented
patient, researchers found that the proportion and pattern
aspects of handwriting were preserved even when vision
was removed; however, the position, scale and orientation
components were compromised. Moreover, the patient was
observed to drift more than controls in a repetitive ellipse
drawing task, while the overall shape of the figure was
preserved (Teasdale et al. 1993). Ingram et al. (2000) found
that a different deafferented patient, IW, was able to adapt
to a visuomotor perturbation for a single joint movement,
although a cognitive load of counting backward degraded
his performance. In both experiments, the patients main-
tained the shape aspects of the drawings, but drifted con-
siderably. It is probable that the attention necessary for
maintaining a seated posture had some impact on the out-
come in terms of drift (Ingram et al. 2000), but proprio-
ception was not necessary to make the shape accurately. In
summary, drift occurs without vision, and even more so
when proprioception is degraded, absent, or challenged for
attentional resources as seen in load conditions in the
present experiment.
Missed corners and serial action
We found that in some cases, subjects entirely missed a
corner of these well-known shapes, but were unaware that
they had done so. They were able to continue drawing the
shape and subsequent repetitions of the shape quite faith-
fully. This disruption and ability for continuation suggests
that each component of the shape was prepared in advance,
rather than being dependent on the preceding correct action;
otherwise in Fig. 9a, our subject would have continued the
drawing with a diagonal up and to the left, rather than toward
the apex of the diamond. Lashley (1951) proposed that
before execution, sequential acts such as typing, or indeed
the drawing of quadrilaterals, were prepared in the brain as
complete actions. Hence, transpositions of keystrokes in
typing or spoonerisms in speech show the parallel prepara-
tion of all elements in the sequence. Bullock (2004) refers to
this collective preparation of the sequence before its exe-
cution as a ‘‘fluent succession of acts’’ and implicates
Brodmann’s area 46, the dorsolateral prefrontal cortex.
Lashley’s hypothesis has been confirmed experimentally.
Averbeck et al. (2003a,b) made recordings from ensembles
of neurons from the right prefrontal cortex of monkeys as the
monkeys drew different shapes copied from visually pre-
sented templates. During this serial motor behavior, the
neuronal ensembles displayed distinctive patterns of acti-
vation corresponding to the beginning of each line segment.
Before the drawing, all line segments in the shape were
represented at the level of neuron populations, with the first
segment of the sequence having a relatively stronger rep-
resentation after the template stimulus had been presented
but before the monkeys had begun to draw. As the monkeys
began drawing each segment, the neural activation corre-
sponding to that segment increased until the hand path was
partially completed, at which point the next segment began
to increase in neural activity. The peak of each activation
corresponded to the commencement of each successive
Averbeck et al. (2003a) also found that the strength of
the neural representation for line segments was stronger for
early and late segments than for middle segments, corre-
lating to the commission of more errors in the middle
segments. This corresponds to the ideas of primacy and
recency in the recall of serially presented stimuli (Robinson
and Brown 1926), where the researchers proposed that in a
memory task, both the first and most recent elements in a
series were most well remembered. When our subjects
mistakenly missed corners of their drawings, more than
93.8% of those missed corners were in the middle of the
drawing, contrasting with 0.8% for beginning segments and
5.4% for end segments. These animal studies and our
behavioral results indicate that the prefrontal cortex may be
a candidate location for some of the conflict experienced by
Exp Brain Res (2010) 202:65–77 75
our subjects, since it is involved in the parallel preparation
of serial acts. Overall, our finding of the strong preserva-
tion of the shape of the drawing over the location of the
drawing suggests that prefrontal cortex–pre-SMA networks
give priority to the shape of a drawing movement.
Was drift a result of identifying arrows or the motor act
of responding?
It is reasonable to consider whether drift across repetitions in
conditions with load and without vision were the conse-
quences of perceptual processes or of processes related to the
preparation and execution of movement plans. For drift we
found different effects for verbal and manual responding,
suggesting that the effects were not due to processing, but
rather a consequence of competition at the level of motor
preparation and production, with verbal and manual responses
competing for different resources. Furthermore, in the cases
where subjects completely missed a corner, there was no
correlation between missing a corner and the angle of the
arrow preceding the missed corner. Although the 3arrow was
more challenging to judge correctly, it did not occur more
frequently before a missed corner than other arrow-angles.
In addition to the verbal and manual differences, we also
manipulated the interval of the arrow presentation between
fixed and random. In the secondary-task-only control
group, subjects performed equally well at identifying the
arrows whether the timing of the presentation stimulus was
fixed or random, with correct responses at more than 92%.
Clearly, the perceptual process was not a problem, nor was
coordinating the index or middle fingers to indicate the
correct response in either fixed or random intervals. When
subjects were faced with randomly presented arrows, while
concurrently drawing a shape in the dual-task conditions,
they made their shapes larger than the template shape
(Fig. 2). One possibility is that the increased uncertainty of
the random presentation made the coordination of the pri-
mary task of drawing and the secondary task of responding
more difficult, leading subjects to delay braking of the
drawing movements in order to allow for synchronization
of the responding task. Another possibility is that subjects
were using the presentation of the arrow as an external
timing cue for judging appropriate movement amplitude;
however, this seems unlikely since arrows were presented
both faster and slower than 1,000 ms, and any effect of
cuing from the arrows would be neutralized, and there
would simply be an increase in variability.
When subjects performed two tasks, performance of the
more complex, primary drawing task was degraded in
terms of location, but not shape and proportion compo-
nents. This supports previous work showing a separate
control and regulation for location and pattern aspects of
limb movement. Vision played an important role in the
preservation of the position, scale and orientation elements
of the drawing. When vision was removed, drift across
repetitions within a trial increased, and when a load was
added, this drift increased to the right showing that while
proprioception alone is inadequate for accurately main-
taining position of a limb, proprioceptive performance
deteriorates when attention is divided. For the shape and
pattern components, overall performance was not degraded
under the secondary load, except when subjects naively
missed a corner. This finding, along with literature
involving proprioceptive disruption suggests that proprio-
ception is not necessary for producing drawing movements
of known shapes, rather that internal representations of the
movement may be employed.
Acknowledgments The authors gratefully acknowledge Dr. Erin
Cressman and Aidan Thompson for their helpful suggestions on the
manuscript and Tarek Kazem and Gissell Suarez for their assistance
with data selection.
Averbeck BB, Chafee MV, Crowe DA, Georgopoulos AP (2003a)
Neural activity in prefrontal cortex during copying geometrical
shapes. I. Single cells encode shape, sequence, and metric
parameters. Exp Brain Res 150(2):127–141. doi:
Averbeck BB, Crowe DA, Chafee MV, Georgopoulos AP (2003b)
Neural activity in prefrontal cortex during copying geometrical
shapes II decoding shape segments from neural ensembles. Exp
Brain Res 150(2):142–153. doi:10.1007/s00221-003-1417-5
Bates JF, Goldman-Rakic PS (1993) Prefrontal connections of medial
motor areas in the rhesus monkey. J Comp Neurol 336(2):211–228
Beilock SL, Wierenga SA, Carr TH (2002) Expertise, attention, and
memory in sensorimotor skill execution: impact of novel task
constraints on dual-task performance and episodic memory. Q J
Exp Psychol A 55(4):1211–1240
Bekkering H, Neggers SFW (2002) Visual search is modulated by
action intentions. Psychol Sci 13(4):370–374
Bekkering H, Adam JJ, Kingma H, Huson A, Whiting HT (1994)
Reaction time latencies of eye and hand movements in single-
and dual-task conditions. Exp Brain Res 97(3):471–476
Brown TL, Carr TH (1989) Automaticity in skill acquisition:
mechanisms for reducing interference in concurrent perfor-
mance. J Exp Psychol Hum Percept Perform 15(4):686–700
Brown LE, Rosenbaum DA, Sainburg RL (2003) Limb position drift:
implications for control of posture and movement. J Neurophys-
iol 90(5):3105–3118
Bullock D (2004) Adaptive neural models of queuing and timing in
fluent action. Trends Cogn Sci 8(9):426
Creem SH, Proffitt DR (2001) Grasping objects by their handles: a
necessary interaction between cognition and action. J Exp
Psychol Hum Percept Perform 27(1):218–228
Deiber MP, Ibanez V, Sadato N, Hallett M (1996) Cerebral structures
participating in motor preparation in humans: a positron
emission tomography study. J Neurophysiol 75(1):233–247
76 Exp Brain Res (2010) 202:65–77
Desmurget M, Vindras P, Grea H, Viviani P, Grafton ST (2000)
Proprioception does not quickly drift during visual occlusion.
Exp Brain Res 134(3):363–377
Eversheim U, Bock O (2001) Evidence for processing stages in skill
acquisition: a dual-task study. Learn Mem 8(4):183–189
Gold D, Park N (2009) The effects of dividing attention on the
encoding and performance of novel naturalistic actions. Psychol
Res 73(3):336–349
Gowen E, Miall RC (2007) Differentiation between external and
internal cuing: an fMRI study comparing tracing with drawing.
Neuroimage 36(2):396–410
Henriques DY, Soechting JF (2005) Approaches to the study of haptic
sensing. J Neurophysiol 93(6):3036–3043
Henriques DY, Flanders M, Soechting JF (2004) Haptic synthesis of
shapes and sequences. J Neurophysiol 91(4):1808–1821
Hiscock M, Chipuer H (1986) Concurrent performance of rhythmi-
cally compatible or incompatible vocal and manual tasks:
evidence for two sources of interference in verbal-manual
timesharing. Neuropsychologia 24(5):691–698
Hiscock M, Cheesman J, Inch R, Chipuer HM, Graff LA (1989) Rate
and variability of finger tapping as measures of lateralized
concurrent task effects. Brain Cogn 10(1):87–104
Ingram HA, van Donkelaar P, Cole J, Vercher JL, Gauthier GM,
Miall RC (2000) The role of proprioception and attention in a
visuomotor adaptation task. Exp Brain Res 132(1):114–126
Kelso JA, Wallace SA, Stelmach GE, Weitz GA (1975) Sensory and
motor impairment in the nerve compression block. Q J Exp
Psychol 27(1):123–129
Klatzky RL (1999) Path completion after haptic exploration without
vision: implications for haptic spatial representations. Percept
Psychophys 61(2):220–235
Krams M, Rushworth MF, Deiber MP, Frackowiak RS, Passingham
RE (1998) The preparation, execution and suppression of copied
movements in the human brain. Exp Brain Res 120(3):386–398
Lashley KS (1951) The problem of serial order in behavior. In:
Jeffress LA (ed) Cerebral mechanisms in behavior: the Hixon
symposium, California Institute of Technology, Pasadena. Haf-
ner, New York, 1967, pp 112–146
Laszlo JI, Bairstow PJ (1971) Accuracy of movement, peripheral
feedback and efference copy. J Mot Behav 3(3):241–252
Miller EK, Cohen JD (2001) An integrative theory of prefrontal
cortex function. Annu Rev Neurosci 24:167–202
Otte E, van Mier HI (2006) Bimanual interference in children
performing a dual motor task. Hum Mov Sci 25(4–5):678–693
Pashler H (1990) Do response modality effects support multiproces-
sor models of divided attention? J Exp Psychol Hum Percept
Perform 16(4):826–842
Pashler H (1994) Dual-task interference in simple tasks: data and
theory. Psychol Bull 116(2):220–244
Proske U (2006) Kinesthesia: the role of muscle receptors. Muscle
Nerve 34(5):545–558
Rand MK, Squire LM, Stelmach GE (2006) Effect of speed
manipulation on the control of aperture closure during reach-
to-grasp movements. Exp Brain Res 174(1):74–85
Robinson ES, Brown MA (1926) Effect of serial position upon
memorization. Am J Psychol 37:538–552
Smith MD, Chamberlin CJ (1992) Effect of adding cognitively
demanding tasks on soccer skill performance. Percept Mot Skills
75(3 Pt 1):955–961
Sohn MH, Ursu S, Anderson JR, Stenger VA, Carter CS (2000)
Inaugural article: the role of prefrontal cortex and posterior
parietal cortex in task switching. Proc Natl Acad Sci USA
Summers JJ, Maeder S, Hiraga CY, Alexander JR (2008) Coordina-
tion dynamics and attentional costs of continuous and discon-
tinuous bimanual circle drawing movements. Hum Mov Sci
27:823–837. doi:10.1016/j.humov.2007.11.003
Taylor JA, Thoroughman KA (2007) Divided attention impairs
human motor adaptation but not feedback control. J Neurophys-
iol 98(1):317–326
Teasdale N, Forget R, Bard C, Paillard J, Fleury M, Lamarre Y (1993)
The role of proprioceptive information for the production of
isometric forces and for handwriting tasks. Acta Psychol 82(1–
Tremblay P, Gracco VL (2009) Contribution of the pre-SMA to the
production of words and non-speech oral motor gestures, as
revealed by repetitive transcranial magnetic stimulation (rTMS).
Brain Res 1268:112–124
Verschueren SM, Swinnen SP, Cordo PJ, Dounskaia NV (1999)
Proprioceptive control of multijoint movement: bimanual circle
drawing. Exp Brain Res 127(2):182–192
Williamson AM, McKenzie BE (1979) Children’s discrimination of
oblique lines. J Exp Child Psychol 27(3):533–543
Woollacott M, Shumway-Cook A (2002) Attention and the control of
posture and gait: a review of an emerging area of research. Gait
Posture 16(1):1–14
Zelaznik HN, Lantero D (1996) The role of vision in repetitive circle
drawing. Acta Psychol 92(1):105–118
Exp Brain Res (2010) 202:65–77 77
Guiding a reach when the spatial location of the viewed target and the hand movement are not congruent (i.e. decoupled) can rely on the use of explicit cognitive rules (strategic control) or on the implicit recalibration of gaze and the limb position (sensorimotor recalibration). We previously demonstrated, in a patient with optic ataxia (OA) having bilateral superior parietal lobule damage, an increased reliance on strategic control when performing a decoupled reach (Granek et al. 2013). To understand fundamental mechanisms of decoupled visuomotor control more generally and to test whether we could distinguish these two modes of movement control more specifically, we tested healthy participants in a cognitively-demanding dual task. Participants continuously counted backwards while simultaneously reaching towards horizontal (left or right) or diagonal (~top-left or ~top-right) targets with either veridical or rotated (90°) cursor feedback. By increasing the overall neural load and selectively compromising potentially overlapping neural circuits responsible for strategic control, the complex dual task served as a non-invasive means to disrupt the integration of a cognitive rule into a motor action. Complementary to our previous results observed in patients with OA, here our dual task led to greater performance deficits during movements that required an explicit rule, implying a selective disruption of strategic control in decoupled reaching. Our results suggest that distinct neural processing is required to control these different types of reaching, since, in considering the current results and previous patient results together, the two classes of movement could be differentiated depending on the type of interference. Copyright © 2014, Journal of Neurophysiology.
Full-text available
Previously, we observed changes in the scale, rotation, and location of drawn shapes when subjects simultaneously performed a secondary task, but not in the shape or proportion of the drawing. We suggested the secondary task impacted motor planning and execution or proprioception of the primary task. To isolate for proprioceptive effects, here we used the same secondary task during passive shape perception. A robotic manipulandum moved the subject's hand around the perimeter of a template shape and then a test shape differing in size, proportion, or location. Subjects also performed the same primary task while simultaneously performing a secondary task of reporting the orientation of right or left tilted arrows. We compared the performance between single and dual task, and different workspaces. In single-task conditions, subjects perceived scale, location, and proportion very close to the actual (all biases under 1 cm). A secondary task only increased the uncertainty range for judgment of scale, with no other effect. Subjects judged shapes in the centered workspace to be smaller and closer relative to the template compared with those in the peripheral workspace, although in that workspace, it was more difficult to discern changes in the proportion of the shape. The result for scale in the current passive paradigm is not different from our active study in which efference copy was available. This suggests that the scale parameters of the shape, whether actively or passively encountered, are disrupted by task interference at the level of proprioception or sensory integration rather than motor planning and execution.
Full-text available
This study investigated the mechanisms that mediate changes in dual-task interference. Sixteen Ss in each of 4 transfer-of-training conditions completed 4 hrs of practice on a speeded keypressing task, with a block of concurrent digit-span trials administered early in practice to 3 groups and late in practice to all 4 groups. Changes in the amount of interference were examined in relation to the transfer conditions, the length of the keypressing sequences performed, the frequency of practice for specific sequences, and the amount of experience with dual-task trials. Results indicated that reductions of dual-task interference are consistent with formulations of intratask automaticity in which chunking and progressive information encapsulation underlie the formation of an integrated motor program. They are not consistent with formulations of strategic deployment of resources between tasks based on attention switching or task integration. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Full-text available
The goal of this study was to investigate whether ocular and hand motor systems operate independently or whether they share processes. Using dualtask methodology, reaction time (RT) latencies of saccadic eye and hand motor responses were measured. In experiment 1, the hand and eye motor systems produced rapid, aimed pointing movements to a visual target, which could occur either to the left or right of a central fixation point. Results showed that RT latencies of the eye response were slower in the dual-task condition than in the single-task condition, whereas the RT latencies of the hand response were virtually the same in both conditions. This interference effect indicated that the ocular and manual motor systems are not operating independently when initiating saccadic eye and goal-directed hand movements. Experiment 2 employed the same experimental paradigm as experiment 1, except for one important modification. Instead of a goal-directed hand movement to the target stimulus, subjects had to make a button-press response with either the index or middle finger of the right hand dependent upon whether the stimulus occurred to the right or left of the control fixation point. The aim of experiment 2 was to investigate the issue whether the observed interference effect in experiment 1 was specific or non-specific (e.g. overhead costs due to coordinating any two responses). The finding that saccadic eye movements and button-press responses in the dual-task condition could be initiated without delay relative to the single-task conditions, supports the specific interference interpretation.
The influence of action intentions on visual selection processes was investigated in a visual search paradigm. A predefined target object with a certain orientation and color was presented among distractors, and subjects had to either look and point at the target or look at and grasp the target. Target selection processes prior to the first saccadic eye movement were modulated by the different action intentions. Specifically, fewer saccades to objects with the wrong orientation were made in the grasping condition than in the pointing condition, whereas the number of saccades to an object with the wrong color was the same in the two conditions. Saccadic latencies were similar under the different task conditions, so the results cannot be explained by a speed-accuracy trade-off. The results suggest that a specific action intention, such as grasping, can enhance visual processing of action-relevant features, such as orientation. Together, the findings support the view that visual attention can be best understood as a selection-for-action mechanism.
Investigated the locus of interference effects and the nature of the retrieval process in secondary memory and compared RT performance on the same materials as W. James's (1980) primary memory (PM) and secondary memory (SM) conditions in the S. Sternberg (1969) paradigm. 90 undergraduates participated in 2 experiments. Identical materials were used in a PM and SM Sternberg situation, the SM process being produced by introducing a distractor task between the termination of the memory set and the probe. As in a release from the proactive interference (PI) paradigm, 3 consecutive trials from 1 taxonomic category were given, and a shift was made to another category for 24 consecutive categories. The 1st trial on a category was defined as low in interference, and the 3rd was defined as high. Overall, results strongly support a response-set, list-differentiation, and an interference at retrieval interpretation of PI, in contrast to an encoding (perceptual) one, and stress the view that the initial retrieval act is retrieval of the address of the set and not of individual items. (47 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
An emerging theoretical perspective, largely based on neuroimaging studies, suggests that the pre-SMA is involved in planning cognitive aspects of motor behavior and language, such as linguistic and non-linguistic response selection. Neuroimaging studies, however, cannot indicate whether a brain region is equally important to all tasks in which it is activated. In the present study, we tested the hypothesis that the pre-SMA is an important component of response selection, using an interference technique. High frequency repetitive TMS (10 Hz) was used to interfere with the functioning of the pre-SMA during tasks requiring selection of words and oral gestures under different selection modes (forced, volitional) and attention levels (high attention, low attention). Results show that TMS applied to the pre-SMA interferes selectively with the volitional selection condition, resulting in longer RTs. The low- and high-attention forced selection conditions were unaffected by TMS, demonstrating that the pre-SMA is sensitive to selection mode but not attentional demands. TMS similarly affected the volitional selection of words and oral gestures, reflecting the response-independent nature of the pre-SMA contribution to response selection. The implications of these results are discussed.
In three experiments using delayed-matching tasks with 5-year-old children (one of which also included 7-year-olds) no support was found for the hypothesis that children code for line orientation using a match-mismatch strategy. The hypothesis that errors in discrimination of obliques can be attributed mainly to confusion over the left-right direction of tilt was rejected in a fourth experiment. It was concluded that young children do not spontaneously encode orientation in relation to a matching environmental cue and that there is no evidence that memory for the direction of obliquity is inferior to memory for the degree of obliquity.