Content uploaded by Liz Franz
Author content
All content in this area was uploaded by Liz Franz on Mar 29, 2014
Content may be subject to copyright.
Exp Brain Res (2004) 157: 174–180
DOI 10.1007/s00221-004-1831-3
RESEARCH ARTICLES
Elizabeth A. Franz
.
Tamara Packman
Fooling the brain into thinking it sees both hands moving
enhances bimanual spatial coupling
Received: 13 May 2003 / Accepted: 22 December 2003 / Published online: 7 February 2004
# Springer-Verlag 2004
Abstract This study examined the hypothesis that the
mirror reflection of one hand’s movement directly
influences motor output of the other (hidden) hand, during
performance of bimanual drawing. A mirror was placed
between the two hands during bimanual circle drawing,
with one hand and its reflection visible and the other hand
hidden. Bimanual spatial coupling was enhanced by the
mirror reflection, as shown by measures of circle size.
Effects of the mirror reflection differed significantly from
effects of vision to one hand alone, but did not differ from
a control task performed in full vision. There was no
evidence of a consistent phase lead of the visible hand,
which indicates that the observed effects on spatial
coupling were immediate and not based on time-consum-
ing feedback processes. We argue that visual mirror
symmetry fools the brain into believing it sees both hands
moving rather than one. Consequently, the spatial proper-
ties of movement of the two hands become more similar
through a process that is virtually automatic.
Keywords Motor processes
.
Efference copy
.
Mirror
reflections
.
Spatial coupling
.
Bimanual symmetry
Introduction
When viewing a mirror placed directly in front of the body
along the midline axis that separates the left and right
sides, the reflection of one moving hand gives the
appearance of symmetrical bimanual movements. The
vividness of this illusion was demonstrated elegantly when
Ramachandran and colleagues reflected amputees’ intact
hands producing different grips and postural motions and
observed participants’ astonishment and surprise at the
observable movements of their phantom limbs (Rama-
chandran and Rogers-Ramachandran 1996; Ramachan-
dran et al. 1996; Ramachandran and Hirstein 1998). An
initial interpretation of this observation was that visual
information associated with the intact limb’s motion, when
reflected to give an appearance of the other (amputated)
limb, also engages the appropriate sensory areas of the
amputated limb. Another possibility (though not mutually
exclusive) is that the mirror reflection fools the brain into
thinking both hands are in view, and this visual mirror
symmetry directly activates central processes associated
with movement (even though the limb itself was no longer
in existence). The precise parameters of limb movement
that are influenced by the mirror manipulation have not
been measured.
In a related study, Franz and Ramachandran (1998)
investigated whether a form of bimanual spatial coupling
would persist even in the absence of feedback processes.
We asked normally-intact participants and people with
single limb amputation to perform drawing movements on
a digitizer tablet under three different experimental
conditions. In a single limb control condition, participants
were instructed to draw lines with one limb (the intact limb
in the case of amputees), and no task was assigned to the
other (phantom) limb. In the parallel bimanual condition,
participants were instructed to draw lines with one limb
(intact limb), and to simulate finger tapping movements of
the other (phantom) limb in the same orientation as the
line drawing movements. In the orthogonal bimanual
condition, participants were instructed to draw lines with
one limb (intact) and to simulate twirling movements of
the index finger of the other (phantom) limb. For the latter
two conditions, instructions were to produce simulated
finger movements (rather than whole arm movements) of
the phantom because the insertion of finger muscles was
below the level of amputation; this would eliminate the
possible influence of afferent feedback that might emanate
from movements of the residual stump. In all conditions,
movement of the limbs was obstructed from view of the
participants, and no visual templates were used. Partici-
pants were therefore required to guide their movements
E. A. Franz (*)
Action, Brain and Cognition Laboratory, Department of
Psychology, University of Otago,
Box 56 Dunedin, New Zealand
e-mail: Lfranz@psy.Otago.ac.nz
Tel.: +64-3-4795269
Fax: +64-3-4798335
from internal representations (memory). Spatial disrup-
tions in the line drawing tasks were measured and then
compared across conditions.
On the basis of findings from normally-intact partici-
pants, larger spatial disruption in the orthogonal condition
compared to the parallel and single hand control
conditions would be interpreted as evidence of spatial
coupling between the limbs. Specifically, when attempting
to produce circular movements with one limb and linear
movements with the other, the lines become circle-like and
the circles become line-like. These disruptions in the
shapes of trajectories are not observed when both hands
are assigned the same shape (Franz et al. 1991; Franz
1997; Franz et al. 1996; see Franz 2003a for review).
Similar to the effects in normally-intact individuals,
Franz and Ramachandran (1998) observed spatial cou-
pling when an amputee with the experience of a vivid
phantom limb produced the bimanual task in the
orthogonal condition. There was significantly more spatial
disruption on the line drawing task in the orthogonal
condition than in the parallel and control conditions. These
results suggest that at least one form of spatial coupling is
not dependent on feedback processes from the limb(s).
These results also indicate that the neural processes that
produce spatial coupling may remain intact even though
the physical properties of a limb may be absent. Indeed,
related studies have demonstrated that the corpus callosum
is one neural structure on which some forms of spatial
coupling depend, given bimanual spatial coupling in
competitive situations (those involving distinct tasks for
the two hands) is substantially reduced in patients with
callosotomy (Franz et al. 1996). Moreover, even without a
corpus callosum, interference between the hands is
minimal in many bimanual cooperative tasks that are
unified by a single behavioral goal that was well-learned
prior to surgery, although this is not the case with novel
bimanual tasks (Franz et al. 2000; see also Franz et al.
2001). As these results imply, clearly there are different
levels of spatial coupling influencing bimanual actions
(Franz 2003a).
Considering again the above-described studies on
amputees, it appears that vivid movement sensation, and
consequent bimanual spatial coupling, may occur in
amputees who experience a phantom limb, echoing the
earlier conclusion that some forms of central (spatial)
coupling remain present even after loss of a limb. The
primary purpose of the present study was to examine the
nature of this form of spatial coupling by closely
investigating the effect of the mirror manipulation applied
by Ramachandran. Specifically, the aim was to investigate
whether the mirror reflection manipulation is powerful
enough to influence motor output properties in normally-
intact participants. We examined this issue by measuring
the movement properties in a bimanual circle drawing task
while applying the mirror reflection manipulation of
Ramachandran.
The task of bimanual circle drawing has been
thoroughly investigated in recent years and provides a
feasible means of measuring both the spatial and the
temporal properties of coupling between the limbs (Carson
et al. 1997; Franz 2000a, 2000b, 2003b; Franz et al. 2002;
Semjen et al. 1995; Stucchi and Viviani 1993; Swinnen et
al. 1996; Wuyts et al. 1996). We hypothesized that specific
spatial properties would likely be altered by the use of the
mirror reflection, due to the known relation between
spatial and visual processes of movement. In bimanual
circle drawing, for example, circle size has been shown to
alter depending on visual attention to one hand or the
other. Specifically, when one limb is attended and the
other limb is hidden from view during bimanual circle
drawing, circles drawn by the attended limb are produced
larger than circles drawn by the unattended limb. These
effects are most pronounced when attention is visual, but
they also occur when attention is nonvisual (Franz 2003b).
Because viewing one’s motions in a mirror depends both
on vision and on attention processes, the task of bimanual
circle drawing seemed suitable to begin this investigation.
Moreover, this task involves a common spatial goal for
both hands, thereby eliminating the spatial interference
that may occur when distinct spatial goals are implemen-
ted (Franz et al. 2001; Franz 2003a, 2003b).
The critical question with respect to spatial coupling
concerns whether between-hand differences in circle size
are altered when the mirror illusion is employed. It was
predicted, based on inferences gained from the application
of mirror bimanual reflections in amputees, that circle size
would be more similar for the tasks of the two hands when
the mirror was employed compared to when vision of only
one hand was employed.
The task of bimanual circle drawing also provides a
means to examine the temporal variable of phase lag
between the hands. If the hidden hand consistently lags
behind the visible hand, then one cannot rule out the
possibility that the performer follows the visual feedback
of one hand to guide movement of the other hand.
However, if the hidden hand does not consistently lag
behind the visible hand, then the effects of the mirror
would seem to be somewhat automatic rather than based
on time-consuming feedback processes.
Materials and methods
Participants
The participants were 15 undergraduate volunteers, 5 males and 10
females, from the University of Otago. They ranged in age from 18
to 23 years (mean=19.5). They were right handed (mean
score=0.78) according to an adapted Edinburgh Handedness Inven-
tory (Oldfield 1971). All procedures herein were approved by the
Human Ethics Committee of the University of Otago.
Apparatus
The experiment was conducted in an enclosed soundproof
experimental booth. The inner wall of the booth was lined with
black curtains to prevent visual distraction or reflection. Participants
were seated with the center of the body directly in line with the
division between two digitizer tablets (30×30 cm) placed side by
175
side. Magnetic pens that did not leave a visible trace were used to
draw. The digitizer tablets were connected to a PC computer and
were run by in-house software routines. The x and y coordinates of
each pen’s position were recorded at 100 Hz with .0025 cm spatial
accuracy.
A wooden box with a mirror attached on one side was placed
directly over one tablet so that the mirror would reflect one of the
participant’s hands while the box hid the other hand from the
subject’s view (see Fig. 1). The mirror box was 46 cm long, 44 cm
wide, and 27 cm high, with a 46 × 38 cm mirror attached on one
side. The mirror box had open ends to allow participants to insert
their hands. The open ends were covered with a black silk curtain to
prevent the participant from seeing one hand or its movement. A
cape protruding from the box was placed over the participant’s
shoulders so that no proximal motion of the arm could be seen (not
shown in Fig. 1 so that the apparatus would be clearly visible). The
mirror could be exposed to reflect the participant’s hand (and its
motion), or covered so that no reflection was visible. The mirror box
could be positioned to reflect either the left or the right tablet (and
hand).
Task, design, and procedure. The experiment involved a within
subjects design, with each subject completing four conditions in
randomized order. The factors mirror (exposed, hidden) and hand
(left, right) were crossed to produce four experimental conditions:
vision of left hand with mirror absent (vision left), vision of left
hand with mirror present (mirror left), vision of right hand with
mirror absent (vision right), and vision of right hand with mirror
present (mirror right). In all conditions, one hand and arm was
completely hidden from view. Each experimental condition
consisted of eight trials, each lasting 8 s. With rest between trials
as needed, each testing session lasted approximately 40 min.
On each testing session, the tablets were first calibrated. An
instruction sheet was then read aloud to participants and they were
given a chance to ask any questions before signing an informed
consent form. The experiment then commenced with a demonstra-
tion of the task. Participants were given an opportunity to familiarize
themselves with the procedures by practicing. They were then
instructed to draw at a comfortable pace in a mirror symmetrical
manner to produce bimanual circles of constant size through
continuous cycles of movement. Participants were instructed to
begin each trial with both pens at approximately the 12:00 position
on the circles. Note that there was no circle template present, thus
subjects drew circles from their own internal representations.
Participants were instructed to focus visual attention on the
boundary between the tablet and the mirror adjacent to the visible
hand’s drawing space. The direction of gaze was at about a 45-
degree angle with the drawing surface. The same instructions about
gaze were given whether the mirror was covered or exposed. This
allowed consistency in angle of gaze across conditions, and also
ensured that the bimanual illusion was clearly visible from the
mirror reflection.
When it was clear that the participant understood the instructions,
the first trial began with the experimenter’s verbal ‘go’ command.
The tablets began collecting approximately 400 ms after movement
commenced, to eliminate any initial variability associated with
beginning to draw. After the computer completed data collection for
each trial, the experimenter said ‘stop’. A short rest was given
between trials. After all trials within a condition were collected, the
experimenter rearranged the mirror box in preparation for the next
condition. After all testing, participants were asked to complete the
handedness inventory, and they were then debriefed completely and
invited to provide comments.
Data reduction
The first two trials in each condition were regarded as practice,
leaving six continuous trials per condition for analysis. The primary
variables of interest were circle duration, circle size (radius), and
phase difference. Algorithms for the computation of variables have
been published in detail in earlier reports (Franz et al. 2002; Franz et
al.
2003b), and will be only briefly described here.
Continuous phase was calculated across the trajectories as an
initial step, given the trajectories may not form perfect circles, and
they may not revolve around a stable center. Tangential angle (TA),
Fig. 1 Mirror box apparatus
and bimanual drawing setup,
showing examples of the right
mirror (left panel) and right
vision (right panel) experimen-
tal conditions. Not shown are
the left mirror and left vision
conditions, which were the mir-
ror-image opposite of the con-
ditions shown. The cape worn
by subjects is not shown so that
the apparatus is clearly visible.
In the mirror conditions, parti-
cipants viewed one side of the
mirror, and the reflection of the
visible hand created the illusion
of bimanual movements (left
panel)
176
or ‘bearing’ was calculated for each point on the trajectory. A virtual
circle of 360 degrees was calculated for each point on the trajectory
by searching backward 180 degrees and forward 180 degrees along
the TA profile. Instantaneous values of each dependent variable
were then computed for each associated point on the circle.
To calculate the radius, the circle center was defined as the
midpoint of the x and y values bounded by the virtual circle. The
radius was calculated as the distance between the reference point of
each virtual circle and the circle center. For the computation of phase
difference, an ‘angle of displacement’ was calculated for each
hand’s movement, where angle of displacement refers to the
orientation of a line drawn from the circle center to a point on the
trajectory. The angle of displacement is defined in degrees, using
(12:00) as a reference. To compute a measure of phase, the
difference between the angles of displacement of the left and right
hands was computed. For example, if at some point in time the left
hand was at a 30-degree orientation and the right hand was at 15
degrees, the phase difference would be a left hand lead of 15
degrees. For each variable, a measure of variance was also computed
to assess consistency in performance.
For mean radius, the signed difference between the hidden hand
and the visible hand was computed. The average difference score
across trials within each block was entered into a repeated measures
ANOVA on the factors visible hand (right, left), × mirror (absent,
present) × block (6). Mean phase was analyzed in a similar manner
(using phase differences between left and right hand movements).
Mean duration was not a difference score, and was therefore
analyzed using the within-subjects factors of condition (4) × hand
(2) × block (6). Within the text, SE will be used to report standard
error of the means.
Results
Given participants were encouraged to draw at a self-
selected pace, mean duration was analyzed to give a
measure of average cycle period. Mean duration across all
participants was 965 ms per circle cycle (SE=66). This
drawing speed is within the range observed in past reports
of self-paced circle drawing. Importantly, participants
drew at similar average speeds in the different experi-
mental conditions, F
(3,42)
=1.70, p>0.05, across the two
hands, F
(1,14)
=2.96, p>0.05, and across blocks, F<1.00.
Table 1 (first four rows) shows the means and standard
errors for the primary dependent variables in each of these
four experimental conditions.
The mean circle radius collapsed across all conditions
and both hands was 5.7 cm (SE=0.49), also within the
range reported in previous studies. There was no hint of a
difference in the radius scores depending on which hand
was visible, F
(1,14)
<1.00. Of primary importance, differ-
ence scores were much larger for mirror absent (difference
score=.65 cm) than for mirror present trials (difference
score=0.24 cm) on average, F
(1,14)
=10.781, p<0.002. In
order to illustrate these effects in the context of the actual
circle sizes, the mean circle radii and standard errors for
each condition are illustrated in Fig. 2 (rather than
illustrating difference scores which, of course, do not
reveal actual circle size).
As can be seen from Fig. 2, it is clear that with vision to
one hand (left vision and right vision conditions) the
visible hand drew larger circles than the hidden hand,
consistent with previous findings (Franz 2003b). The
novel effect is that when the mirror reflection was applied
(left mirror and right mirror conditions), this difference in
circle size between hands was significantly reduced. The
difference between the mirror absent and mirror present
conditions was highly significant when the right hand was
visible (right vision–right mirror comparison), and margin-
ally significant when the left hand was visible (left vision–
left mirror comparison), respectively F
(1,14)
=12.39,
p<0.002, and F=4.46, p=0.053. These results support the
hypothesis that the mirror reflection enhances bimanual
spatial coupling. There were no other interactions on mean
radius (all p>0.05).
As can be clearly seen from Fig. 2, addition of the
mirror reflection appeared to primarily alter radii of the
circles drawn by the left hand. This adjustment of the left
hand circle size occurred regardless of whether the left
Table 1 Means and standard errors of primary dependent variables
in the single hand vision, mirror reflection, and full vision control
conditions. Variables of mean radius and cycle duration are shown
averaged across both hands to illustrate that they are comparable
across conditions. Mean phase difference is also shown. Negative
values of phase difference indicate a right hand lead
Mean radius Cycle duration Phase difference
Right vision 5.57 cm (.51) 1,000 ms (84) −3.07 ms (3.2)
Right mirror 5.86 cm (.41) 985 ms (60) −3.20 ms (4.1)
Left vision 5.85 cm (.53) 932 ms (68) −1.29 ms (3.2)
Left mirror 5.72 cm (.49) 946 ms (64) −3.09 ms (3.3)
Full vision control 6.69 cm (.30) 1,088 ms (55) −2.40 ms (3.0)
Fig. 2 Mean radius and standard error are shown for the four
conditions of the main experiment. In the two conditions with the
mirror, the mean radius is more similar for the two hands than in
conditions with vision only
177
hand was in view or the right hand was in view. This effect
will be revisited below.
To examine whether one hand consistently led the other,
phase difference was analyzed. We hypothesized that if the
visible hand consistently leads, then we could not rule out
the hypothesis that a time-consuming process of overt
monitoring may have guided the hidden hand’s movement.
An alternative possibility is that a more direct and
automatic process is responsible for the engagement of
spatial properties of the hidden hand when the mirror is
applied. This would be rather immediate, and therefore
would not be expected to show up as a consistent lead of
the visible hand.
Of primary importance, there was a right hand lead on
average, in all conditions. The average lead ranged from –
1.3 ms to –3.1 ms (the negative sign indicates a right hand
lead), and the miniscule differences across conditions were
not close to being reliable, all F<1.00. Because it was not
the visible hand but rather the right hand that consistently
led, the present findings support the hypothesis that a
direct and rapid coupling occurs between parameters of the
two hands as a result of the mirror reflection.
Discussion
This experiment examined the effects of a mirror reflection
of one hand’s circle drawing movements on the motor
output of a bimanual task. The primary purpose was to
examine whether the mirror reflection alters the motor
output of a hidden hand’s movement. The primary effects
were clear and unambiguous. The effect of a larger circle
size produced by the visible hand compared to the hidden
hand was significantly reduced when the mirror reflection
was applied. This effect was found whether the left hand
or the right hand was the visible hand. No apparent effects
occurred in the temporal parameters (duration or phase) as
a result of the mirror manipulation.
There are a few possible ways in which the observed
enhancement of spatial coupling could occur as a result of
the mirror reflection. One account is that the feedback of
one hand’s movement is used to actively guide movement
of the other hand. This active corrective process based on
visual feedback is believed to take some time to operate
(Keele and Posner 1968; Carlton 1981). By this hypoth-
esis, we would expect the hidden hand to lag behind the
visible hand. Our data clearly do not support this account.
An alternative account is that a more rapid and direct form
of coupling occurs between the visual information and the
processes governing motor output of the hidden hand. This
may require the operation of feedforward or efference
copy that allows for a rapid speed of information
transmission between the perception of movement and
the motor output (Franz and Ramachandran 1998); or this
process could operate on the global properties of the
trajectory (i.e., on the internal representation of the circle:
i.e., Franz et al. 1991; Franz 2003a, 2003b). Either of these
processes would have to allow for rapid, almost automatic,
alterations in the movement plan. The results on phase lag
clearly support the hypothesis that coupling occurs in a
rapid manner rather than as a result of active monitoring of
feedback, given that the hidden hand did not consistently
lag behind the visible hand.
One might probe further into the nature of the effects of
the mirror. Are the effects on spatial coupling the result of
application of the mirror per se? Or, does the visual
illusion of actually seeing two hands moving increase
bimanual spatial coupling between the hands?
1
A possible
way to test this issue would be to allow participants to see
both hands while performing the bimanual drawing task
and then compare the present mirror conditions with the
new control condition using full vision. If the full vision
condition produces results that are identical to those found
in the mirror conditions, then findings would support the
hypothesis that enhanced bimanual spatial coupling is due
to the illusion that the participant is seeing both hands
moving in a symmetrical manner. In contrast, if the full
vision condition produces different results from the mirror
condition, then findings based on application of the mirror
would remain unexplained.
In a control experiment on 15 naïve subjects of
approximately the same description as in the main
experiment, bimanual circle drawing data were collected
under full vision conditions (allowing subjects to see both
hands). The methods were identical to those employed
earlier, except that full vision of both hands was allowed,
no mirror box was used, and only one condition was
tested. We then conducted between-subjects comparisons
of all dependent variables across mirror and full vision
conditions.
It is perhaps important to point out that the comparison
between the left mirror condition and the full vision
control would not be clearly interpretable, due to issues
related to hand dominance. Recall that in the main
experiment, the left hand adjusted its circle size to
accommodate that of the right hand when the mirror was
applied (compared to vision only conditions), regardless of
which hand was mirrored (see Fig. 2). In addition, the
right hand was the lead hand, on average, in all conditions
(see Table 1). These findings illustrate the well-known
property that the right hand is dominant in bimanual
symmetrical movements, especially for right-handed
subjects (Carson et al. 1997; Franz 2000a, 2000b, 2003a,
2003b; Franz et al. 2002; Semjen et al. 1995; Stucchi and
Viviani 1993; Swinnen et. al. 1996; Wuyts et al. 1996).
There is also strong support for the claim that the right
hand tends to be the primary focus of attention during
bimanual performance (Franz 2003b; Peters 1981, 1985),
perhaps accounting for the hand dominance effects
observed in the present study. With the right hand in
view, these two properties (hand dominance and attention
dominance) both favor the right hand, thereby allowing for
direct interpretations of the comparison between the full
vision control condition and the right mirror condition.
However, this is not the case for the corresponding
1
We thank an anonymous reviewer for suggesting we conduct the
additional control experiment.
178
comparisons involving the left mirror condition because
the dominant right hand is hidden in one condition (left
mirror) and visible in the other (full vision). Our
conclusions, therefore, are based only on data from the
comparison of right mirror versus full vision conditions.
The results for the full vision versus right mirror
conditions were clear and unambiguous. There was
absolutely no hint of a difference in the radius scores
produced in the two conditions, nor was there any effect
on mean phase difference, both F
(1,28)
<1.00. In addition,
mean duration was not reliably different across the two
conditions, F
(1,28)
=1.748, p>0.05. The means and standard
errors for each variable in the full vision condition appear
with the other conditions in Table 1. These results strongly
support the argument that the mirror reflection results in
the illusion that both hands are in view, and this visual
symmetry of apparent bimanual movement enhances
spatial coupling of the two hands in a manner similar to
actual vision of both hands.
Our findings are consistent with the hypothesis that
visual capture (Rock 1966) of the observed mirror
symmetry promotes fusion of the visual symmetry on
the proprioception of the moving hands (see Rosetti et al.
1995 for experiments using unimanual reaching). This
proposed process of binding the visual information with
the proprioception of movement would have to be rather
immediate, given there is no evidence that the visible hand
consistently led the hidden hand. One possibility is that the
corpus callosum, which is critical to some forms of
bimanual spatial coupling (Franz et al. 1996), enables the
coactivation of both hemispheres when visual symmetry is
perceived. Accordingly, this coactivation would ensure
that fusion of visual and proprioceptive information occurs
rapidly.
Another possibility is that bimodal neurons (also
referred to as ‘mirror neurons’: Gallese et al. 1996)
provide a direct interface between seen actions and
produced actions; although this hypothesis would extend
what is currently known about these neurons because the
requirement would be that they respond not only to an
external agent but also to one’s own reflection in a mirror.
Another hypothesis that is tempting to invoke is derived
from studies on tracking of predictable targets.
2
In an
elegant set of studies on grip force adjustments using
hand-held loads, Flanagan and Wing demonstrated that
modulations in grip force occur in parallel with changes in
load. This 0 lag in the modulation of grip force is
suggestive of a forward model in which the load force on
the hand is predicted and incorporated into the internal
model (Flanagan and Wing 1993, 1994, 1997). Applying
this type of idea to the present study, it is reasonable that
the visible hand that is reflected in the mirror acts as a
highly predictable target, especially given its movement is
governed under the subject’s voluntary control. The
hidden hand might actually track this highly-predictable
target with virtually a 0 lag, consistent with the findings of
Flanagan and Wing. Indeed, this framework would appear
to generalize to bilateral effectors, based on a related
experiment performed sometime earlier that required
subjects to produce grip force adjustments based on
frictional properties of an object using one hand followed
by grip force adjustments using the other hand on
sequential trials. Stored information related to frictional
properties of the object that resulted in adjustments in grip
force appeared to influence subsequent trials performed by
the other hand (Johansson and Westling 1984). This type
of bilateral access of the internal model would be
necessary to account for the present findings given the
circle size of the left hand movements adjusted to that of
the right hand (in comparison with vision only conditions),
even when the left hand’s predictable target was the visible
one. In other words, if the predictable target information
directly updated an internal representation that is acces-
sible to both hands, then this hypothesis seems plausible.
In sum, although the precise theoretical explanation and
neural mechanisms for the effect remain to be determined,
the present findings demonstrate that bimanual spatial
coupling is enhanced with the application of a mirror
reflection. This enhancement occurs relatively automati-
cally, rather than as a result of more time-consuming
feedback-dependent processes.
Acknowledgements Thanks to Barry Dingwall, Jeremy Anderson,
and Michael Bowden for their technical expertise and helpful
suggestions. This work was funded in part by Otago Research Grant
MFU B26 to Liz Franz
References
Carlton LG (1981) Processing visual feedback information for
movement control. J Exp Psychol Hum Percept Perform
7:1019–1030
Carson RG, Thomas J, Summers JJ, Walters MR, Semjen A (1997)
The dynamics of bimanual circle drawing. Q J Exp Psychol
50A:664–683
Flanagan JR, Wing AM (1993) Modulation of grip force with load
force during point-to-point arm movements. Exp Brain Res
95:131–143
Flanagan JR, Wing AM (1995) The stability of precision grip forces
during cyclic arm movements with a hand-held load. Exp Brain
Res 105:455–464
Flanagan JR, Wing AM (1997) The role of internal models in
motion planning and control: evidence from grip force
adjustments during movements of hand-held loads. J Neurosci
17:1519–1528
Franz EA (1997) Spatial coupling in the coordination of complex
actions. Q J Exp Psychol 50A:684–704
Franz EA (2000a) A double dissociation in the cerebral dominance
of bimanual control: callosotomy versus callosal agenesis.
International meeting of Attention and Performance XIX,
Kloster Irsee, Germany, July 2000
Franz EA (2000b) A double dissociation between callosal agenesis
and callosotomy participants on a bimanual task. Forum for
European Neurosciences, United Kingdom, July 2000
Franz EA (2003a) Bimanual action representation: a window to
human evolution. In: Johnson-Frey SH (ed) Taking action:
cognitive neuroscience perspectives on the problem of inten-
tional acts. MIT Press, Cambridge, MA, pp 259–288
Franz EA (2003b) The attentional distribution of task parameters to
the two hands during bimanual performance of right handers
and left handers. J Motor Behav (in press)
2
We again thank a reviewer for this helpful insight.
179
Franz EA, Ramachandran VS (1998) Bimanual coupling in
amputees with phantom limbs. Nature Neurosci 1:443–444
Franz EA, Zelaznik HN, McCabe G (1991) Spatial topological
constraints in a bimanual task. Acta Psychol 77:137–151
Franz EA, Eliassen J, Ivry R, Gazzaniga MS (1996) Dissociation of
spatial and temporal coupling in the bimanual movements of
callosotomy patients. Psychol Sci 7:306–310
Franz EA, Waldie KE, Smith MJ (2000) The effect of callosotomy
on novel versus familiar bimanual actions: a neural dissociation
between controlled and automatic processes? Psychol Sci
11:82–85
Franz EA, Zelaznik HN, Swinnen SP, Walter C (2001) Spatial
conceptual influences on the coordination of bimanual actions:
when a dual task becomes a single task. J Motor Behav 33:103–
112
Franz EA, Rowse A, Ballantine B (2002) Does handedness
determine which hand leads in a bimanual task? J Motor
Behav 34:402–412
Gallese V, Fadiga L, Fogassi L, Rizzolatti G (1996) Localization of
grasp representations in humans by positron emission tomo-
graphy. 2. Observation compared with imagination. Exp Brain
Res 112:103–111
Johansson RS, Westling G (1984) Roles of glabrous skin receptors
and sensorimotor memory in automatic control of precision
group when lifting rougher or more slippery objects. Exp Brain
Res 56:550–564
Keele SW, Posner MI (1968) Processing visual feedback in rapid
movement. J Exp Psychol 77:155–158
Oldfield RC (1971) The assessment and analyses of handedness:
The Edinburgh Inventory. Neuropsychologia 9:97–113
Peters M (1981) Attentional asymmetries during concurrent biman-
ual performance. Q J Exp Psychol 33A:171 –196
Peters M (1985) Constraints in the performance of bimanual tasks
and their expression in unskilled and skilled subjects. Q J Exp
Psychol 37A:171–196
Ramachandran VS, Rogers-Ramachandran D (1996) Synaesthesia
in phantom limbs induced with mirrors. Proc R Soc Lond B
263:377–386
Ramachandran VS, Hirstein W (1998) The perception of phantom
limbs. Brain 121:1603–1630
Ramachandran VS, Levi L, Stone L, Rogers-Ramachandran D,
McKinney R, et al. (1996) Illusions of body image: what they
reveal about nature. In: Llinas R, Churchland PS (eds) The
mind-brain continuum. MIT Press, Cambridge, MA, pp 29–60
Rock I (1966) The nature of perceptual adaptation. Basic, New York
Rosetti Y, Desmurget M, Prablanc C (1995) Vectorial coding of
movement: vision, proprioception, or both? J Neurophysiol
74:457–463
Semjen A, Summers JJ, Cattaert D (1995) Hand coordination in
bimanual circle drawing. J Exp Psychol Hum Percept Perform
21:1139–1157
Stucchi N, Viviani P (1993) Cerebral dominance and asynchrony
between bimanual two-dimensional movements. J Exp Psychol
Hum Percept Perform 19:1200–1220
Swinnen SP, Jardin K, Meulenbroek R (1996) Between-limb
asynchronies during bimanual coordination: effects of manual
dominance and attentional cueing. Neuropsychologia 34:1203–
1213
Wuyts IJ, Summers JJ, Carson RG, Byblow WD, Semjen A (1996)
Attention as a mediating variable in the dynamics of bimanual
coordination. Hum Mov Sci 15:877–897
180