Exp Brain Res (2007) 179:191–198
DOI 10.1007/s00221-006-0780-4
R E SEARCH ART I CLE
Returning home: location memory versus posture memory
in object manipulation
Matthias Weigelt · Rajal Cohen · David A. Rosenbaum
Received: 21 June 2006 / Accepted: 25 October 2006 / Published online: 22 November 2006
Springer-Verlag 2006
Abstract Previous studies of object manipulation
have suggested that when participants return an object
to the place from which they just carried it, they tend to
grasp the object for the target-back-to-home trips close
to where they just grasped it for the home-to-target
trips [Exp Brain Res 157(4):486–495, 2004; Psychon
Bull Rev, 2006]. What was unclear from these previous
studies was whether participants recalled postures or
locations. According to the posture hypothesis, they
remembered what body positions they adopted when
they last held the object. According to the location
hypothesis, they remembered where they held the
object and then took hold of it there or nearby again.
To distinguish between these possibilities, we had
participants mount or dismount a platform after hometo-target moves and before target-back-to-home
moves. In the control condition, they did not change
their vertical position relative to the shelf containing
the home and target platforms (they merely stepped
sideways). We found that participants grasped the
M. Weigelt (&)
University of Bielefeld, Bielefeld, Germany
e-mail: matthias.weigelt@uni-bielefeld.de
M. Weigelt
Max-Planck Institute for Human Cognitive and Brain
Sciences, Munich, Germany
R. Cohen · D. A. Rosenbaum
Department of Psychology,
Pennsylvania State University,
University Park, PA 16802, USA
e-mail: rajal@psu.edu
D. A. Rosenbaum
e-mail: dar12@psu.edu
object at nearly the same place along its length as they
had before, even if this meant adopting very diVerent
postures than before. This outcome is consistent with
the location-recall account and is inconsistent with the
posture-recall account. The implications for motor
planning are discussed.
Introduction
Previous research has shown that human participants
take hold of objects diVerently depending on what they
plan to do with the objects. Such anticipatory eVects
were Wrst discovered by Marteniuk et al. (1987), who
found diVerences in the speed of the hand as it
approached a tennis ball to be tossed or a light bulb to
be screwed into a socket. Later, Rosenbaum et al.
(1990) showed that university students grasped the
same rod with diVerent hand orientations depending
on what they were going to do with the rod. When the
rod was going to be brought to an orientation that
would cause the hand to occupy an awkward Wnal posture (thumb pointing down rather than up), participants generally took hold of the rod in a way that
prevented this awkward Wnal posture. Rosenbaum
et al. (1990) called the tendency to avoid uncomfortable Wnal postures the end-state comfort eVect. The
behaviors that reXected this eVect were observed in
subsequent studies (Rosenbaum et al. 1990; Rosenbaum and Jorgensen 1992; Rosenbaum et al. 1995,
1996, 2006; Short and Cauraugh 1997, 1999; Weigelt
et al. 2006). The end-state comfort eVect was taken to
suggest that actors anticipate future body states. This
view is consistent with current conceptions of the
importance of goal-state representations in action
123
192
planning (Blakemore et al. 2002; Hommel et al. 2001;
Kunde and Weigelt, 2005; Rosenbaum et al. 2001;
Wolpert and Flanagan 2001).
More recent experiments designed to test the generality of the end-state comfort eVect have shown that
university students took hold of a vertically oriented
rod at diVerent heights depending on the height to
which the rod would be carried (Cohen and Rosenbaum 2004; Rosenbaum et al. 2006). The rod that was
used was the wooden shaft of a standard bathroom
plunger (i.e., the rod extending up from a rubber base).
When the plunger stood at a Wxed initial height and
then had to be moved to a high target, participants
grasped the plunger low (close to the base), but when
the plunger stood at the same initial height and then
had to be moved to a low target, participants grasped
the plunger high (farther from the base). In general,
participants grasped the plunger at a height that was
inversely related to the height to which the plunger
would be carried. Analyses of the grasp heights at the
initial and terminal positions showed that participants
behaved much as did participants in the earlier demonstrations of the end-state comfort eVect, modifying how
they initially took hold of the plunger to avoid extreme
uncomfortable postures (extreme joint angles) at the
ends of movements.
An unexpected Wnding that emerged in the studies
of grasp-height control pertained to returns of the
plunger to its home position. After participants
brought the plunger to the target position, they lowered their hand and then reached out once again, this
time to grasp the plunger and return it to its (Wxed)
home position. If participants had obeyed the end-state
comfort eVect for the return moves, they would have
grasped the plunger at a point along its length that
would have ensured a comfortable Wnal position back
at the home site. Thus, they would have grasped the
plunger at the same point along its length no matter
what the target position was. This is not what happened. Instead, participants grasped the plunger close
to where they had grasped it before, when they carried
the plunger from the home site to the target site.
Cohen and Rosenbaum (2004) called this the grasp
height recall eVect.
What does the grasp height recall eVect signify?
Cohen and Rosenbaum (2004) and Rosenbaum et al.
(2006), who replicated the eVect, suggested that it signiWed that participants were making use of memory in
their formation of action plans. According to this
explanation, the Wrst time a participant encounters a
task condition, an action plan is computed based on an
assessment of the task demands. When a similar condition is encountered within a very short time frame,
123
Exp Brain Res (2007) 179:191–198
participants recall key features of what they just did
and Wne tune the plan. This explanation is reminiscent
of two-stage models of cognitive problem solving,
where it has been hypothesized that the solution to a
(math) problem is initially computed (e.g., the product
of 13 times 13 is arrived at through multiplication) but
is later recalled as a previously stored instance (e.g.,
the participant recalls the proposition “13 times 13
equals 169”); see Logan (1988).
How can one determine what speciWc information is
recalled when participants exhibit the grasp height
recall eVect? The approach taken here is to distinguish
between two hypotheses about what that information
could be. One possibility is that participants returned
to the posture they adopted when they released the
plunger. Another possibility is that participants
returned to the location along the length of the plunger
where the hand was for the just completed home-totarget trip. The Wrst hypothesis says that what was
recalled was represented in intrinsic (body-based)
coordinates. The second hypothesis says that what was
recalled was represented in extrinsic (allocentric) coordinates. Distinguishing between these alternatives was
not only important for shedding light on the sources of
the grasp height recall eVect; it was also important for
shedding light, more generally, on what is recalled in
physical action tasks.
Method
We tested the alternative hypotheses by creating situations in which recalling locations and recalling postures
would result in diVerent behaviors. As shown in Fig. 1,
participants reached out to move the plunger from a
home position to a target position, then lowered the
hand, and then reached out to move the plunger back
to the home position again. This simple procedure was
the same as the one used in the previous studies of
grasp height. What was new here was that between the
home-to-target move and the target-back-to-home
move, participants stepped to the right, either down
from a platform, up onto a platform, or, in the control
condition, merely sideways. By having participants step
up onto or down from a podium or merely step sideways, we could ask whether, for target-back-to-home
moves, participants tried to grasp the plunger close to
where they had done so before in extrinsic coordinates
(i.e., at a point close to where they had grasped it relative to the base of the plunger) or in intrinsic coordinates (i.e., at a point close to where they had grasped it
before relative to their feet). The main prediction, as
shown in Fig. 2, was that if locations were recalled in
Exp Brain Res (2007) 179:191–198
193
to a malfunction of the video recording system. This
study was approved by the local ethics committee and
carried out according to the 1964 Declaration of Helsinki.
Apparatus and materials
Fig. 1 Three podium arrangements used in the experiment. Left
panel Podium (P) on the left; middle panel no podium; right panel
podium on the right. The home shelf (HS) is on the left and the
target shelves (TS) are on the right in each arrangement. The arrow shows the step participants were asked to take after moving
the plunger from the home shelf to any one of the three target
shelves. Numbers indicate cylinder height, podium height, and
target heights (all in cm)
Fig. 2 Possible strategies for grasping a plunger after holding it
while standing on the Xoor (left panel). One strategy is to grasp it
on the same shaft location as before (middle panel). Another
strategy is to grasp it using the same posture as before to replay
the just-performed movement in reverse (right panel). Schematic
diagram only, not drawn to scale
the grasp height recall eVect, grasp height for return
moves should remain invariant with respect to extrinsic
coordinates. By contrast, if postures were recalled in
the grasp height recall eVect, then grasp height for
return moves should remain invariant with respect to
intrinsic coordinates.
Participants
Fifteen students (9 female and 6 male; mean
age = 24.7 years; age range 20–32 years) from the University of Munich participated. All the participants
characterized themselves as both right-handed and
neurologically healthy, and all were naive to the purpose of the study. Each participant was tested individually. Before testing, each participant provided his or
her informed consent. After the experiment was completed, each participant received 5 Euros (the equivalent of approximately 6 US Dollars) for participation.
One participant had to be excluded from the study due
The apparatus was similar to the one used by Cohen
and Rosenbaum (2004) except for a wooden podium
(30 cm wide, 50 cm deep and 10 cm high), which was
added to the set-up in particular conditions. When subjects were not standing on the podium, they were asked
to stand on one of two rectangular pieces of paper
(21.0 cm by 29.6 cm) taped to the Xoor, 30 cm in front
of the bookshelf, on the left or on the right.
The bookshelf consisted of three shelves at heights
of 50, 85, and 120 cm. A wooden target platform rested
on each of these shelves on the right, and a home platform extended 15 cm from the middle shelf on the left.
On the home platform stood a standard toilet plunger
(only used in this and similar experiments) whose
wooden shaft was cylinder, 2.5 cm in diameter and
50 cm high, and was supported by a circular rubber
base, 10 cm in diameter and 5 cm high. The weight of
the cylinder was counterbalanced by a brick positioned
on the inner end of the wooden platform (i.e., behind
the base from the participant’s perspective). Each of
the three target shelves could be pushed into the bookshelf or pulled out of the bookshelf, causing it to extend
15 cm from the edge of the bookshelf, as was the case
for the home shelf. The target shelves also had bricks in
the back to counterbalance the cylinder and base.
Design and procedure
Each participant was tested with all three target shelf
heights in all three conditions (podium-absent,
podium-left and podium-right). The order of podium
conditions was randomly assigned to each participant
prior to testing. When a particular podium condition
was selected, the participant proceeded until all target
shelf heights were tested in that podium condition. The
target shelf heights were randomly presented within
each podium condition. After all target shelf heights
were tested for one podium condition, the experimenter changed the set-up to continue with the next
block, until all podium conditions were tested.
Before the participant started the experiment, she/
he was told that we were recording short video
sequences to be shown in a memory experiment with
another group of participants. This cover story was necessary because a video camera stood on a tripod to the
right of the bookshelf, in full view of the participants.
123
194
We asked the participants to perform the task in a
relaxed, natural fashion, but to pay close attention to
the instructions of the experimenter, who announced
each action sequence to be performed. The camera
lens was focused to fully capture the whole length of
the cylinder on the home platform. A colored dot was
taped to the back of the right hand of each participant.
The participants were instructed to keep their left
hands by their sides at all times and only to move their
right hands when told to do so by the experimenter.
They were further instructed to hold the cylinder
securely, and to move it at a comfortable, unhurried
speed while performing the task.
At the beginning of each trial in the podium-absent
conditions, the participant was asked to step to the
paper on the left, in front of the home shelf. The experimenter then pulled out one of the three target platforms on the right, indicating the next placement
position of the cylinder. The participant was instructed
to take hold of the cylinder with the right hand and to
move it to the target platform, then to set the cylinder
with its base down on that shelf, and then to return the
hand back to the side of the body. This completed the
home-to-target move. Next, the experimenter asked the
participant to step over to the right, onto the paper in
front of the target shelves. For the following targetback-to-home move, the participant was instructed to
grasp the plunger, step back over to the left, onto the
paper in front of the home shelf, and return the plunger
to the home shelf. This sequence of events was repeated
for each target shelf height, constituting one block.
When the moves to and from one target were completed, the experimenter pushed back the platform
into the bookshelf and consulted a previously prepared
design sheet before pulling out the next target shelf,
whereupon the next home-to-target move and targetback-to-home move sequence was tested.
In the podium-present conditions, the experimenter
placed the podium either in front of the home shelf or
in front of the target shelves, where it remained until
the plunger had been taken by the participant to all target shelf heights twice. Then, the experimenter either
moved the podium to the other side of the bookshelf or
removed it, depending on which block was next.
If the podium was placed in front of the home shelf
(podium-left), the participant was asked to step on to
the podium and then to move the cylinder to the target
platform. After the cylinder was placed on the target
platform, the participant took a step down from the
podium and stood with his or her feet on the paper in
front of the target shelves. She/he then reached for the
plunger again and next took hold of it before taking a
step back up on to the podium, whereupon she/he
123
Exp Brain Res (2007) 179:191–198
placed it back on the home shelf. As in the no-podium
condition, this procedure was repeated twice for each
trial before the next target shelf height was tested.
If the podium was placed in front of the target
shelves (podium-right), the participant was asked to
stand on the piece of paper in front of the home shelf
and take hold of the plunger on the home shelf, next to
place the plunger on the target shelf, and then to step
up the podium. To return the plunger from the target
platform back to the home platform, the participant
reached for the plunger again, took a step down from
the podium (with the plunger in his or her hands), and
stood on the paper in front of the home shelf, before
Wnally placing the plunger on the home shelf. As in the
other two conditions, this procedure was repeated
twice for each target shelf height.
Each participant was tall enough to comfortably
reach the top of the cylinder when it was placed on the
top platform. The entire session lasted for about
20 min. The participants were debriefed after the session.
OV-line video analysis
Because the performance of each participant was captured on videotape, it was possible to estimate each
participant’s grasp heights on the plunger shaft after
the testing session. We estimated the position of the
right hand on the cylinder at two critical moments in
each transport cycle: (1) when the participant took
hold of the plunger to carry it to the target platform
and (2) when the participant returned the plunger from
the target platform and set it down on the home shelf.
Because the participants were instructed to Wrmly
grasp the plunger and hold it securely during the transport action (i.e., not to let the cylinder slip through
their Wngers), we assumed that the grasp heights on the
plunger for the target-back-to-home moves were faithfully reXected back at the home shelf at the ends of the
target-to-home moves. Measuring the grasp heights at
this one position for all movement cycles ensured standardization of measurement.
The way we used the videos to measure participants’
grasp heights was as follows. For each measurement,
the experimenter froze the picture frame of interest
and measured the on-screen distance between the colored dot (which was placed at the back of the right
hand of each participant before the experiment) and
the bottom of the cylinder base. The experimenter also
measured the on-screen length of the cylinder for each
measurement sample. This was done as an extra precaution in case the participants slightly varied the position of the plunger on the home shelf, thus changing
Exp Brain Res (2007) 179:191–198
195
Results
30
20
10
0
50
85
120
Shelf Height (cm)
Left Platform
50
40
30
20
10
0
50
85
120
Shelf Height (cm)
Extrinsic coordinates
The main eVects for shelf height, F(2, 26) = 19.648,
P < 0.001, direction, F(1,13) = 7.262, P < 0.05, and platform, F(2, 26) = 4.957, P < 0.05, were signiWcant, whereas
the main eVect of repetition was not, F(1,13) = 0.203,
P < 0.092. In addition, the direction £ platform interaction, F(2, 26) = 8.357, P < 0.01, was signiWcant, as was
the shelf height £ repetition interaction, F(2, 26)
= 3.848, P < 0.05, but none of the other interactions
reached or approached signiWcance.
To further examine the direction £ platform interaction, we collapsed the data over the factors shelf height
and repetition and conducted an additional 2
(direction) £ 3 (platform) ANOVA. A graphic depiction
of this interaction can be seen in the lower panel of Fig. 4,
where the diVerent grasp heights for home-to-target
No Platform
50
40
30
20
10
0
50
85
120
Shelf Height (cm)
No Platform
50
40
30
20
10
0
50
85
120
Shelf Height (cm)
Fig. 3 Mean grasp height relative to feet (top panels) and relative
to plunger base (bottom panels) when the platform was on the left
(left panels), when the platform was on the right (right panels),
and when there was no platform (middle panels). The two curves
are for the Wrst home to target moves (right-pointing triangles)
and for the Wrst target-back-to-home moves (left-pointing trian-
Grasp Height Relative To Plunger Base (cm)
40
Grasp Height Relative To Feet (cm)
Left Platform
50
Grasp Height Relative To Plunger Base (cm)
Grasp Height RelativeTo Plunger Base (cm)
Grasp Height Relative To Feet (cm)
As seen in Fig. 3, the grasp heights for the target-backto-home moves were much more similar to the grasp
heights for the home-to-target moves when the data
were plotted relative to the plunger base (i.e., in extrinsic coordinates) than when the data were plotted relative to the feet (i.e., in intrinsic coordinates).
To evaluate these impressions, we analyzed participants’ grasp height in extrinsic and intrinsic coordinates and we submitted the data to two separate
repeated-measures analyses of variance whose designs
were 3 (shelf height: 50 vs. 85 vs. 120 cm) £ 2 (direc-
tion: HT vs. TH) £ 3 (platform: left vs. none vs.
right) £ 2 (repetition: 1st vs. 2nd). One ANOVA used
the data relative to the plunger base (extrinsic coordinates). The other used the data relative to the feet
(intrinsic coordinates).
Grasp Height Relative To Feet (cm)
the apparent length of the cylinder. To analyze the
data, we divided the distance from the bottom of the
cylinder base to the colored dot by the distance from
the bottom of the cylinder to the top of the cylinder.
Then, to express the grasp height in centimeters, we
multiplied the actual length of the cylinder by the ratio
obtained in the preceding step. That value (in cm) was
reported as the grasp height.
Right Platform
50
40
30
20
10
0
50
85
120
Shelf Height (cm)
Right Platform
50
40
30
20
10
0
50
85
120
Shelf Height (cm)
gles). Because the home position was on the left, participants carried out home-to-target moves while standing on the platform in
the left platform condition and carried out target-back-to-home
moves while standing on the platform in the right platform condition. Error bars show §1 SE
123
Exp Brain Res (2007) 179:191–198
grasp height relative to feet (cm)
196
35
HT
30
TH
25
20
15
10
5
grasp height relative to plunger base (cm)
left platform
no platform
right platform
35
HT
30
TH
25
20
15
10
5
left platform
no platform
right platform
Fig. 4 Direction £ platform interaction using, in the upper panel,
grasp height relative to the feet (intrinsic coordinated) and, in the
lower graph, grasp height relative to the plunger base (extrinsic
coordinates) when the platform was on the left, when the platform was on the right, and when there was no platform. The two
curves are for home to target moves (solid circles) and for targetback-to-home moves (empty circles). Because the home position
was on the left, participants carried out home-to-target moves
while standing on the platform in the left platform condition and
carried out target-back-to-home moves while standing on the
platform in the right platform condition. Error bars show §1 SE
moves (HT) and target-back-to-home moves (TH) are
displayed for the various platform conditions. Followup t-tests of the signiWcant direction £ platform interaction, F(2, 26) = 8.440, P = 0.001, revealed that mean
grasp height was about 2 cm lower for TH moves than
for HT moves when the platform was placed to the left
(22.31 vs. 24.29 cm, respectively) and when there was
no platform (20.16 vs. 22.15 cm, respectively), with the
signiWcance of both of these two-tailed tests being high,
P < 0.01. This small but signiWcant diVerence was
absent when the platform was placed to the right (22.17
vs. 21.54 cm).
Intrinsic coordinates
The main eVects for shelf height, F(2, 26) = 19.626,
P < 0.001, direction, F(1,13) = 7.213, P < 0.05, and platform, F(2,26) = 138.141, P < 0.001, were signiWcant,
123
whereas the main eVect for repetition, F(1,13) = .202,
P = 0.092, was not. Also, the direction £ platform
interaction, F(2,26) = 280.566, P < 0.001, and the shelf
height £ repetition
interaction,
F(2,26) = 3.813,
P < 0.05, both reached signiWcance. None of the other
interactions reached or approached signiWcance.
To further examine the direction £ platform interaction, we collapsed the data over shelf heights and
repetitions and conducted an additional 2
(direction) £ 3 (platform) ANOVA. A graphic depiction of this interaction can be seen in the upper panel
of Fig. 4, where the diVerent grasp heights for home-totarget moves (HT) and target-back-to-home moves
(TH) are displayed for the various platform conditions.
Follow up t-tests of the signiWcant direction £ platform
interaction, F(2,26) = 281.141, P < 0.001, revealed that
the mean grasp height was about 2 cm lower for TH
moves than for HT moves when there was no platform
(20.16 vs. 22.15 cm, respectively). When the platform
was placed to the left, grasp heights were about 8 cm
higher for TH moves than for HT moves (32.31 vs.
24.29 cm, respectively). Conversely, when the platform
was placed to the right, TH moves were about 9 cm
lower than HT moves (12.17 vs. 21.54 cm, respectively).
Discussion
Previous studies of object manipulation have suggested
that when an object is returned to the place from which
it was just carried, participants tend to grasp the object
for the target-back-to-home trips close to where they
had just grasped it for the home-to-target trips (Cohen
and Rosenbaum 2004; Rosenbaum et al. 2006). What
was unknown from these previous studies was whether
participants recalled postures or locations. According
to the posture hypothesis, participants recalled the
body positions they adopted when they last held the
plunger and adopted that posture again upon taking
hold of the plunger for the return trip. According to
the location hypothesis, participants recalled the place
along the length of the plunger occupied by the hand
for the just-completed home-to-target trip.
To distinguish between these possibilities, we had
participants mount or dismount a platform after some
of their home-to-target moves and before some of their
target-back-to-home moves. In the control condition,
they did not change their vertical position relative to
the shelf containing the home and target platforms
(they merely stepped sideways). We found that participants grasped the plunger at nearly the same place
along the length of the plunger as they had before (the
Exp Brain Res (2007) 179:191–198
same height above the base of the plunger), even if this
meant adopting very diVerent postures than before
(diVerent heights above the feet). Figure 3 reXects this
behavioral strategy. The grasp heights for target-backto-home moves were much more similar to the grasp
heights for the home-to-target moves when the data
were plotted relative to the plunger base (i.e., in extrinsic coordinates) than when the data were plotted relative to the feet (i.e., in intrinsic coordinates). This
outcome shows that participants recalled locations on
the plunger, rather than their previously attained goal
postures, a Wnding that is consistent with the locationrecall account and inconsistent with the posture-recall
account.
The present Wndings also help to address an even
more basic concern about the interpretation of the previous grasp-height results. That concern was whether
participants were in fact relying on memory to select
their grasp heights for return moves. One need not
assert that if participants grasped the plunger for target-to-home moves close to where they had grasped
the plunger for home-to-target moves, they did so
because they remembered either the previous grasp
location or the previous grasp posture. It might just be
a coincidence that the grasp heights were so closely
related. The present results argue against this possibility because, as shown in Fig. 3, grasp heights were
much better explained by assuming location invariance
than posture invariance. If successive grasp heights
were related by coincidence only, one would not
expect one dimension of control (location) to account
so much better for the relation than another dimension
of control (posture).
What are the broader implications of our Wndings?
First, our results agree with the notion that memory for
information expressed in extrinsic coordinates is more
robust than, or is relied on more, memory for information expressed in intrinsic coordinates (see Smyth 1984,
for a review). Interestingly, studies of serial reaction
time have led to the same conclusion. In those studies,
it has been found that participants pressing buttons
associated with lights as quickly as they can are far
more disrupted when the mapping of lights to buttons
changes than when the mapping of Wngers to buttons
changes (Willingham et al. 2000).
Second, our results have implications for the role
that posture memory may play in motor planning.
According to the posture-based motion planning theory (Rosenbaum et al. 1995, 2001), goal postures are
speciWed before movements to those goal postures are
speciWed. The greater ease of repeating movements
(i.e., shorter initiation latencies and decreasing endpoint variability) is explained in the theory by the
197
availability of more and better stored goal postures for
those movements. Having these goal postures available in memory reduces the time needed for motor
planning because, according to the hypothesis, recalling postures provide a quicker basis for specifying goal
postures than generating goal postures de novo on the
basis of new computations; see Rosenbaum et al.
(1995, 2001) for details. The latter proposal is reminiscent of Logan’s distinction between generation and
recall of problem solutions, mentioned earlier in this
article.
Previous work showed that when participants had to
reach to a speciWed target location repeatedly with
alternating hands, left–right errors were mirrored
across hands, as would be expected if postures were
copied from one arm to the other (Rosenbaum et al.
1999). The present study demonstrates that in addition
to remembering postures, participants can remember
locations in extrinsic space when they change position
vis-à-vis an object during two consecutive motor
actions. The question of why participants rely on location memory in some situations and on posture memory in other situations remains to be answered.
References
Blakemore SJ, Wolpert DM, Frith CD (2002) Abnormalities in
the awareness of action. Trends Cognit Sci 6:237–242
Cohen RG, Rosenbaum DA (2004) Where grasps are made reveals how grasps are planned: generation and recall of motor
plans. Exp Brain Res 157(4):486–495
Hommel B, Muesseler J, Aschersleben G, Prinz W (2001) The
theory of event coding (TEC): a framework for perception
and action planning. Behav Brain Sci 24(5):849–878
Kunde W, Weigelt M (2005) Goal-congruency in bimanual object
manipulation. J Exp Psychol Human Percept Perform
31(1):145–156
Logan GD (1988) Toward an instance theory of automatization.
Psychol Rev 95:492–527
Marteniuk RG, MacKenzie CL, Jeannerod M, Athenes S, Dugas
C (1987) Constraints on human arm movement trajectories.
Can J Psychol 4:365–378
Rosenbaum DA, Jorgensen MJ (1992) Planning macroscopic aspects of manual control. Hum Move Sci 11:61–69
Rosenbaum DA, Marchak F, Barnes HJ, Vaughan J, Slotta JD,
Jorgensen MJ (1990) Constraints for action selection: overhand versus underhand grip. In: Jeanerod M (ed) Attention
and performance XIII. Lawrence Erlbaum Associates, Hillsdale, pp 321–342
Rosenbaum DA, Loukopoulos LD, Meulenbroek RG, Vaughan
J, Engelbrecht SE (1995) Planning reaches by evaluating
stored postures. Psychol Rev 102:26–67
Rosenbaum DA, van Heugten CM, Caldwell GE (1996) From
cognition to biomechanics and back: the end-state comfort
eVect and the middle-is-faster eVect. Acta Psychol 94:59–85
Rosenbaum DA, Meulenbroek RJ, Vaughan J (1999) Remembered positions: stored locations or stored postures? ExpBrain Res 124:503–512
123
198
Rosenbaum DA, Meulenbroek RG, Vaughan J, Jansen C (2001)
Posture-based motion planning: applications to grasping.
Psychol Rev 108:709–734
Rosenbaum DA, Halloran E, Cohen RG (2006) Precision
requirements aVect grasp choices. Psychon Bull Rev
Short MW, Cauraugh JH (1997) Planning macroscopic aspects of
manual control: end-state comfort and point-of-change
eVects. Acta Psychol 96:133–147
Short MW, Cauragh JH (1999) Precision hypothesis and the endsate comfort eVect. Acta Psychol 100:243–252
123
Exp Brain Res (2007) 179:191–198
Smyth MM (1984) Memory for movements. In: Smyth MM, Wing
AM (eds) The psychology of human movement. Academic,
London, pp 83–117
Weigelt M, Kunde W, Prinz W (2006) End-state comfort in
bimanual object manipulation. Exp Psychol 53(2):143–148
Willingham DB, Wells LA, Farrell JM, Stemwedel ME (2000)
Implicit motor sequence learning is represented in response
locations. Memory Cognit 28:366–375
Wolpert DM, Flanagan JR (2001) Motor prediction. Curr Biol
11:729–723