Washington University School of Medicine

Digital Commons@Becker
Open Access Publications

2008

Neuroimaging studies of mental rotation: A meta-

analysis and review
Jeffrey M. Zacks
Washington University School of Medicine in St. Louis

Follow this and additional works at: http://digitalcommons.wustl.edu/open_access_pubs

Recommended Citation
Zacks, Jeffrey M., ,"Neuroimaging studies of mental rotation: A meta-analysis and review." Journal of Cognitive Neuroscience.20,1.
1-19. (2008).
http://digitalcommons.wustl.edu/open_access_pubs/3351

This Open Access Publication is brought to you for free and open access by Digital Commons@Becker. It has been accepted for inclusion in Open
Access Publications by an authorized administrator of Digital Commons@Becker. For more information, please contact [email protected].
 Neuroimaging Studies of Mental Rotation:
A Meta-analysis and Review

Jeffrey M. Zacks

Abstract
& Mental rotation is a hypothesized imagery process that has these areas is modulated by parametric manipulations of mental
inspired controversy regarding the substrate of human spatial rotation tasks, supporting the view that mental rotation de-
reasoning. Two central questions about mental rotation remain: pends on analog representations. Mental rotation also is ac-
Does mental rotation depend on analog spatial representations, companied by activity in the medial superior precentral cortex,
and does mental rotation depend on motor simulation? A re- particularly under conditions that favor motor simulation, sup-
view and meta-analysis of neuroimaging studies help answer porting the view that mental rotation depends on motor sim-
these questions. Mental rotation is accompanied by increased ulation in some situations. The relationship between mental
activity in the intraparietal sulcus and adjacent regions. These rotation and motor simulation can be understood in terms of
areas contain spatially mapped representations, and activity in how these two processes update spatial reference frames. &

INTRODUCTION
agery debate, supporting the analog representation view
Mental rotation is a hypothetical psychological opera- (Kosslyn, 1994; but see Pylyshyn, 2003). Neuroimaging
tion in which a mental image is rotated around some studies of mental rotation provide a valuable means to
axis in three-dimensional space. Mental rotation was first test the hypothesis that mental rotation, in particular,
revealed in behavioral experiments (Cooper & Shepard, operates in virtue of analog spatial representations.
1973; Shepard & Metzler, 1971) in virtue of a striking Imaging evidence for such representations should satisfy
finding: The time to make a judgment about a rotated two criteria. First, activation should be observed in areas
object often increases in a near-linear fashion with the known to be spatially mapped. Second, the observed
amount of rotation required to bring the object into activation should be modulated by the amount of men-
alignment with a comparison object or with a previously tal rotation performed in a given trial or block of trials;
learned template. Participants in mental rotation experi- this can be indexed by variables such as the angular
ments generally report strong introspective impressions distance of the rotation or participants’ response time. If
that they perform the experimental tasks by forming a spatially mapped regions of the brain consistently show
mental image of the stimulus and imagine it rotating increases in activation with increasing amounts of men-
until aligned. That is, they intuit that they are perform- tal rotation, this would provide good evidence that men-
ing an operation on analog spatial representations, in tal rotation depends on analog spatial representations.
which ‘‘the intermediate stages of the internal process More recently, a debate has emerged about the role of
have a demonstrable one-to-one relation to intermediate motor processes in mental rotation. Early neuroimaging
stages of the corresponding external process’’ (Shepard studies found activity during mental rotation tasks in areas
& Cooper, 1982, p. 13). in the posterior frontal cortex that are associated with
Mental rotation has been controversial since the first motor planning and execution (Cohen & Bookheimer,
reports. The experimental results have been taken as 1994). These findings and others have led researchers to
evidence for the existence of analog spatial representa- ask whether motor simulation plays an important role in
tions in the mind/brain (Shepard & Cooper, 1982); mental rotation tasks. Neuroimaging studies can also con-
however, alternative interpretations have attempted to tribute to answering this question. To do so, it is im-
explain the behavioral patterns in mental imagery ex- portant to dissociate activity in motor regions of the brain
periments without recourse to analog representations due to motor demands of a task (e.g., pressing a response
(Pylyshyn, 1981). Neuropsychological and neuroimaging button) from activity related to spatial reasoning.
data have provided further evidence in the larger im- There are now several dozen neuroimaging studies of
mental rotation (see Appendix A), so one might hope that
the available data would help settle these outstanding
Washington University in Saint Louis controversies: Does mental rotation depend on analog

D 2008 Massachusetts Institute of Technology Journal of Cognitive Neuroscience 20:1, pp. 1–19
spatial representations, and does mental rotation de- activation was converted to the stereotactic coordinate
pend on motor simulation? Unfortunately, however, the system of Talairach and Tournoux (1988) and entered
body of extant data is difficult to navigate: Foci of brain into a database. When a paper reported multiple analy-
activity associated with mental rotation have been re- ses that reused the same degrees of freedom, the most
ported in every lobe of the cerebrum and in the cere- tightly controlled task comparison was used. For results
bellum. The range of experimental and control tasks reported in figures or described verbally, the stereotactic
has varied widely, which may contribute to this variability. coordinates of each focus of activation were recorded
In such situations, the statistical technique of meta- using CARET (Van Essen et al., 2001; http://brainmap.
analysis is particularly valuable (Turkeltaub, Eden, Jones, wustl.edu/caret) and the PALS atlas (Van Essen, 2005;
& Zeffiro, 2002). In the rest of this article, I describe http:// brainmap.wustl.edu:8081/sums/directory.
a meta-analytic study of neuroimaging studies of men- do?id=679528). (This required subjective judgments as
tal rotation, delving into more detail on neuroimaging to the centroid of each focus based on the reported
and other data relevant to the motor imagery question, data.) A total of 320 activation foci were identified, 213
and discuss the implications of these data for the two transformation-specific foci and 107 omnibus foci. These
controversies. are listed in Appendix B.
Volume-wise probability maps were generated using
the method and software described by Turkeltaub et al.
(2002). Each focus of activation was located in a com-
SELECTION OF ARTICLES AND
mon stereotactic space (Talairach & Tournoux, 1988
META-ANALYSIS PROCEDURE
space) and convolved with a Gaussian kernel (SD =
Articles were identified by searching the PsycInfo (www. 6.0 mm) to model the spatial uncertainty associated with
apa.org/psycinfo) and MedLine (www.nlm.nih.gov/pubs/ the focus’s location. The resulting Gaussian distributions
factsheets/medline.html) databases in July 2006 for articles were summed, projected onto a 2-mm isotropic lattice,
whose titles, abstracts, or keywords included the term and thresholded to control the voxelwise false-positive
‘‘mental rotation’’ and any of the following terms: ‘‘PET,’’ rate at p  .001, based on the Monte Carlo simulations
‘‘positron emission tomography,’’ ‘‘fMRI,’’ ‘‘MRI,’’ or ‘‘neu- reported by Turkeltaub et al. Three maps were generat-
roimaging.’’ This returned 68 documents. Documents were ed: one based on the transformation-specific foci, one
excluded if they were not peer-reviewed journal articles based on the omnibus foci, and one based on all of the
publishing new data (e.g., dissertations, review articles, foci. After thresholding, contiguous clusters of above-
n = 8), did not report a positron emission tomography threshold voxels were identified. Above-threshold voxels
(PET) or functional magnetic resonance imaging (fMRI) within the cortex were visualized using the PALS atlas
study (e.g., behavioral experiments, n = 13), did not and using CARET. The complete dataset, including the
study mental rotation of visual stimuli (e.g., spatial navi- individual foci and probability maps, is available at http://
gation or mental rotation of haptic stimuli, n = 6), or if sumsdb.wustl.edu/sums/directory.do?id=6617254&dir_
they did not present the data in a form from which a name=JCogNeuro_07.
stereotactic coordinate could be extracted (e.g., only
sulci/gyri or Brodmann’s areas reported, or a small num-
ber of transverse slice images, n = 9). The remaining 32
EFFECTS OF MENTAL ROTATION TASKS
articles are listed in Appendix A. In all experiments,
ON LOCAL BRAIN ACTIVITY
participants either decided whether two objects matched
or decided whether an object matched a prototype (e.g., The results of the meta-analysis are shown in Table 1
whether an alphanumeric character was facing forward). and Figure 1. Brain regions that were consistently ac-
The majority of experiments (n = 19) used Shepard and tivated included the superior parietal, frontal, and in-
Metzler (1971) objects or similar stimuli. Other stimulus ferotemporal cortex. Activity was observed bilaterally in
sets included alphanumeric characters (n = 7), drawings most areas; however, in the parietal cortex, activity was
or photographs of objects (n = 4), drawings or photos of somewhat more consistently observed in the right hemi-
hands (n = 4), abstract 2-D line figures (n = 3), abstract sphere, whereas in the frontal cortex, activity was more
3-D cubes (n = 3), and drawings of bodies (n = 1). consistently observed in the left hemisphere.
For each article, the task comparisons reported were
categorized as transformation-specific or omnibus.
Posterior Parietal/Occipital Cortex
Transformation-specific contrasts isolated within-task
effects of mental rotation, for example, by comparing a A large number of studies reported foci of activation in
condition involving stimuli with large rotations to stimuli the superior parietal cortex and adjacent areas, leading
with small rotations during the same task. Task com- to a large significant region of activation centered in the
parisons were identified as omnibus if they compared intraparietal sulcus and extending into the transverse
a mental rotation task to a loose control, for example, occipital sulcus, including Brodmann’s areas 7, 19, 39,
looking at a fixation crosshair. Each reported focus of and 40. The consistent activation of the superior parie-

2 Journal of Cognitive Neuroscience Volume 20, Number 1

Table 1. Brain Regions That Were Consistently Activated in a Meta-analysis of Neuroimaging Studies of Mental Rotation
Talairach Coordinates

x y z Volume (cm3) Brodmann’s Area Description

All Task Comparisons

23.6 64.8 42.7 29.35 7/19 Intraparietal sulcus
27.3 63.6 41.3 21.45 7/19 Intraparietal sulcus, temporal–occipital sulcus
38.3 5.2 34.7 8.94 6 Precentral sulcus, superior frontal sulcus
36.0 83.4 4.2 3.28 19 Lateral occipital sulcus
42.6 15.3 21.0 2.42 44/46 Inferior frontal sulcus, inferior precentral sulcus
29.8 7.5 46.9 2.26 4/6 Precentral sulcus, superior frontal sulcus
0.1 15.2 42.3 0.83 8/9/32 Superior frontal gyrus
15.3 17.6 35.8 0.78 24/32 Anterior cingulate sulcus
0.7 5.5 53.7 0.12 6 Cingulate sulcus
8.1 89.9 0.9 0.06 17 Calcarine fissure

Omnibus Task Comparisons

24.9 66.7 47.1 5.26 7/19 Superior parietal lobule
31.3 79.0 31.9 2.66 19 Intraparietal sulcus, temporal–occipital sulcus
40.4 2.1 42.1 1.83 6 Precentral sulcus
39.1 24.7 35.9 1.75 9/46 Inferior frontal sulcus
39.8 17.0 14.7 1.39 13 Insula
35.8 0.1 41.9 1.32 6 Precentral sulcus
19.5 64.5 51.7 1.28 7 Intraparietal sulcus
38.3 20.6 26.5 1.06 44/46 Inferior frontal sulcus, inferior precentral sulcus
0.6 2.7 53.2 0.60 6 Medial precentral gyrus
1.5 14.3 46.9 0.42 8 Superior frontal gyrus
44.0 33.8 56.7 0.27 2 Postcentral sulcus
50.9 20.5 5.0 0.15 44/45 Frontal operculum

Transformation-specific Task Comparisons

23.2 64.7 41.6 26.74 7/19 Intraparietal sulcus, temporal–occipital sulcus
29.3 64.6 34.8 21.77 19 Intraparietal sulcus, temporal–occipital sulcus,
lateral occipital sulcus
30.4 3.7 43.7 2.06 6 Precentral sulcus
43.9 19.2 22.0 1.70 44/45/46 Superior frontal sulcus
27.7 9.7 49.1 1.58 6/8 Precentral sulcus
15.0 17.8 35.4 1.18 24/32 Anterior cingulate sulcus
45.8 11.3 23.2 0.43 44/45/46 Inferior frontal sulcus, inferior precentral sulcus
47.5 59.5 10.0 0.18 37 Posterior inferior temporal sulcus

Coordinates are for the center of mass of a contiguous above-threshold probability cluster.

Zacks 3
 Thus, the results of the meta-analysis satisfy two key
criteria for supporting the hypothesis that mental rota-
tion depends on analog spatial representations. First,
activity was observed in a brain area known to imple-
ment spatial maps, namely, the superior parietal cortex.
Second, activity in this region was consistently related
to the amount of mental rotation performed in a trial or
block of trials; activations were observed in transformation-
specific comparisons, not only in omnibus comparisons.
This result reduces the likelihood that activity in the pos-
terior parietal cortex during mental rotation tasks re-
flects incidental task features such as encoding a visual
stimulus or making a manual response.

Motor Regions in the Precentral Cortex

The top pane of Figure 1 indicates that a number of
studies reported activity during mental rotation tasks in
the precentral sulcus, bilaterally. Regions in the precen-
tral cortex are associated with motor planning and
Figure 1. Meta-analysis of neuroimaging studies of mental rotation. execution, and have been identified with mental rotation
The top panel shows the regions that responded above chance
since the first neuroimaging studies of mental rotation
( p  .001, corrected for multiple comparisons) in mental rotation
task comparisons across all studies. Brighter colors indicate stronger (Cohen & Bookheimer, 1994). It has been suggested
responses. The bottom panel compares two more specific types of that precentral activity may reflect the use of motor sim-
task contrast ( p < .001, corrected for multiple comparisons): Red: ulation (computations that map to the specifics of joint
transformation-specific contrasts compared different degrees of mental angles and/or torques; Michelon, Vettel, & Zacks, 2006)
rotation demand within a single task. Green: omnibus contrasts
to solve mental rotation problems. In other words, par-
compared mental rotation tasks to loose control tasks, for example,
fixating on a crosshair. Yellow: regions that were significantly ticipants may simulate moving objects with their hands.
associated with both transformation-specific and omnibus task However, an alternative is that activation in motor areas
comparison. In both panels, the meta-analysis results are projected may be caused by incidental features of the tasks used—
on medial, oblique lateral, and posterior views of the cortical surface. in particular, the need to make a manual response in
nearly every study. Thus, does ‘‘motor’’ activity during
mental rotation tasks reflect the use of motor simulation
tal cortex during mental rotation tasks converges with to solve the problems, or does it simply reflect the de-
neuropsychological data (Ratcliff, 1979) in indicating mands of planning and executing a motor response?
an important role for this region in visuospatial image Inspection of Figure 1 and Table 1 provides part of an
transformations. As can be seen in the bottom of Fig- answer to this question. Foci of activation during trans-
ure 1, activity in this region was consistently observed in formation-specific contrasts were confined largely to the
transformation-specific comparisons. medial superior portion of the precentral sulcus. This
The superior parietal cortex is known to implement region is described in the motor control literature as the
maps of space that code the locations of targets of in- supplementary motor area (SMA; Picard & Strick, 2001).
tended actions. This can be seen most clearly in electro- The SMA is known to be responsive during motor imagery
physiological studies of eye movements and reaching in tasks and projects to both the primary motor cortex (M1)
the monkey (Andersen & Buneo, 2002; Colby & Goldberg, and the spinal cord, placing it in a good position to play a
1999). Individual cells in monkey superior parietal cortex role in motor control and simulation. Foci of activation
frequently represent locations in eye-centered coordi- during the omnibus contrasts were largely confined to the
nates, but these eye-centered responses are modulated lateral inferior portion of the precentral sulcus. These acti-
by the monkey’s head or body position (Snyder, Grieve, vations likely overlapped both M1 and lateral premotor
Brotchie, & Andersen, 1998), making it possible to read off cortex (PM).1 Like SMA, M1 and PM both contain projec-
location in body- or world-centered reference frames. tions to the spinal cord, but these are denser in M1 (Dum
Converging with the monkey data, patients with neglect & Strick, 2002). PM appears to code for actions at a more
syndrome, in which one part of space is ignored, often abstract level, representing the distal targets of actions as
have lesions in the parietal cortex (Robertson, 2004; but well as the proximal effectors (Schubotz & von Cramon,
see Karnath, Ferber, & Himmelbach, 2001), and this ne- 2003). One possibility is that the activation of these areas
glect can manifest itself in eye-, body-, or world-centered during omnibus contrasts reflects the planning and exe-
reference frames. cution of the task’s motor response. In transformation-

4 Journal of Cognitive Neuroscience Volume 20, Number 1

specific contrasts, these components are better controlled motor simulation strategy. This supports the hypothesis
because the tasks’ motor response is constant throughout that activity in these regions during mental rotation re-
the within-task conditions being compared. Thus, the me- flects motor simulation.
dial superior activations, likely corresponding to activity in
the SMA, may well be due to mental rotation.
A series of PET experiments by Kosslyn, DiGirolamo, Prefrontal Cortex
Thompson, and Alpert (1998) have probed the condi-
tions under which mental rotation tasks activate motor The lateral inferior prefrontal cortex has not typically
areas. Their results converge in suggesting that, under been highlighted in discussions of mental rotation data;
some conditions, people are more likely to engage in however, eight different studies (studies 5, 9, 13 15, 23,
motor simulation to solve mental rotation problems, and and 30–31 in Appendix A) identified foci of activation
that when people engage this strategy, medial/superior in this area in transformation-specific contrasts and the
motor areas are activated. The first experiment com- responses were tightly clustered, leading to significant
pared a mental rotation task involving hands with a task regions of activation in the meta-analysis. In two of these
using abstract figures (Kosslyn et al., 1998). Mental ro- studies (Kosslyn et al., 1998, 2001), activation in the left
tation of hands produced greater activity in the left pre- inferior frontal cortex was specific to mental rotation
central gyrus. Another study used all abstract figures tasks that were designed to encourage motor simulation
but manipulated participants’ training before scanning (see above). As can be seen in Figure 1, the overlapping
(Kosslyn, Thompson, Wraga, & Alpert, 2001). In one locations were in BA 44 and 45. This region has long
condition, participants were trained to imagine an object been associated with speech production; more recently,
rotated by an external force; in the other condition they it has been implicated in motor control and imitation
imagined turning an object with their right hand. Ac- (Rizzolatti, Fadiga, Gallese, & Fogassi, 1996). Therefore,
tivity in a region of the precentral cortex was reliably one possibility is that this activation, like that in SMA,
greater when participants imagined turning the object. reflects the use of motor simulation in some mental ro-
(The authors identify this region as M1, but its ste- tation tasks.
reotactic coordinates place it in the precentral sulcus,
within the medial superior cluster shown in Figure 1.
This may correspond to SMA rather than M1.) A third
IMPLICATIONS
study (Wraga, Thompson, Alpert, & Kosslyn, 2003) used a
transfer paradigm, in which participants first performed The present meta-analysis allows for two strong conclu-
mental rotation of either hands or abstract objects, and sions regarding the neural substrate of mental rotation.
then all participants performed mental rotation of ab- First, the posterior parietal cortex (as well as regions ex-
stract objects during scanning. Those participants who tending down into the superior posterior occipital cor-
had previously rotated hands showed greater activity in tex) is consistently activated during mental rotation
the precentral cortex bilaterally than those who had not. across a range of tasks, imaging modalities, and statisti-
A recent event-related fMRI study compared a task re- cal analysis strategies. This region is therefore a good
quiring right–left judgments about pictures of hands to candidate for implementing the transformation-specific
right–left judgments about alphanumeric characters (de computations required to carry out mental rotation
Lange, Hagoort, & Toni, 2005). This experiment identified tasks. This finding converges with data from neuropsy-
two sets of regions in the precentral sulcus whose activ- chological studies (Farah, 1989) and from a transcranial
ity was positively correlated with the degree of rotation magnetic stimulation study (Harris & Miniussi, 2003).
required on each trial. One set increased in activity with The fact that these areas are known to contain multiple
increasing rotation during mental rotation of both hands spatial maps provides support for the view that mental
and characters, whereas the other set increased with in- rotation is based on analog spatial representations.
creasing rotation only during hand rotation. Interestingly, However, such inferences from where brain activity is
the regions with hand-selective rotation responses also observed in a task to what that brain region is doing are
appeared to respond to the motor demands of the task, relatively weak, particularly as in this case when the
which further strengthen the interpretation that they areas in question respond during tasks unrelated to the
performed motor simulations. One block-design fMRI task of interest (Poldrack, 2006). The inference can be
study that compared mental rotation of hands and tools strengthened in this case, however, because activation in
(Vingerhoets, de Lange, Vandemaele, Deblaere, & Achten, this area was found in transformation-specific as well as
2002) failed to find significant differences between the omnibus comparisons. Moreover, several studies have
two in motor areas, although there was a trend toward conducted parametric analyses of brain activity as a
greater activity in the right precentral cortex during hand function of response time, accuracy, proportion of
rotation. In sum, these data indicate that activity in parts rotated stimuli, or rotation angle, and have observed
of the precentral sulcus during mental rotation tasks graded effects in these areas (Gauthier et al., 2002;
is greatest when the task situation affords the use of a Podzebenko, Egan, & Watson, 2002; Zacks, Ollinger,

Zacks 5
Sheridan, & Tversky, 2002; Harris et al., 2000; Tagaris, depend (Zacks & Michelon, 2005). Performing most
Kim, Strupp, & Andersen, 1996). (This does not, how- mental rotation tasks requires coordinating two different
ever, necessitate that mental rotation operate holistically reference frames: an object-centered reference frame
on whole objects rather than on object parts; see, e.g., and an environmental reference frame. Object-centered
Just & Carpenter, 1985.) Together with the meta-analytic reference frames locate things relative to the intrinsic
result, these provide relatively good support for the axes of objects. For example, a car has a well-defined top
hypothesis that mental rotation is a continuous trans- (roof ), bottom (undercarriage), front (grille), and back
formation performed on analog spatial representations. (trunk). Environmental reference frames locate things
Second, motor areas in the posterior frontal cortex are relative to the larger environment. For example, a movie
clearly activated during many mental rotation paradigms. theater has a well-defined front (screen), back (exits),
However, these activations appear to reflect at least two top (ceiling), and bottom (floor). For visually presented
different processes. Activity in lateral inferior precentral stimuli, both of these reference frames must be com-
regions, overlapping M1 and PM, likely reflects incidental puted by the brain from information initially coded in an
features of some mental rotation tasks such as the need eye-centered reference frame. One possibility is that
to execute a motor response. Activity in medial superior object-centered and environmental reference frames in
regions, most likely in the SMA, probably reflects com- the superior parietal cortex are distributed representa-
putational processes that are specific to the mental image tions constructed in the superior parietal cortex by gain
transformations. In particular, activity in these regions may fields, which modulate the eye-centered responses of
reflect the use of motor simulation strategies when the individual cells based on signals coding the current
task affords it. Studies using transcranial magnetic stimu- position of the head and body (Andersen & Buneo,
lation have provided converging evidence consistent with 2002). When mentally rotating an object, a person
this hypothesis, although they do not have the spatial updates the relationship between the object’s object-
resolution to distinguish between M1, the SMA, and the centered reference frame and an environmental frame. If
premotor cortex. Three studies have reported that tem- the task requires one to compare two objects to decide
porarily lesioning the left motor cortex impairs mental if they are identical, updating is used to align the two
rotation of hands, but not of alphanumeric characters or objects’ object-centered reference frames, so that the
feet (Tomasino, Borroni, Isaja, & Ida Rumiati, 2005; Ganis, locations of object features can be compared in a
Keenan, Kosslyn, & Pascual-Leone, 2000). common environmental reference frame. Interactions
Modulation of activity in the precentral cortex may among cell populations in the superior parietal cortex
reflect the degree that participants adopt a motor sim- may be the mechanism by which these two frames are
ulation strategy to solve a mental rotation task. (The brought into a particular relation. Such interactions may
lateral inferior prefrontal cortex and the superior parie- be governed by associations learned from a lifetime of
tal cortex also play important roles in motor planning, experience observing rotational motion.
and thus, may also be modulated by the use of motor In some situations, it may be possible to bring to bear
simulation.) That activity also may reflect the use of a yet another reference frame when performing mental
motor transformation strategy, which could be imple- rotation tasks: effector-centered reference frames, which
mented by transformations of a neural representation of are defined relative to one’s hands or feet. When one
a reaching movement (Georgopoulos, Lurito, Petrides, grasps an object, an effector-based reference frame be-
Schwartz, & Massey, 1989). The hypothesis that motor comes coupled to the object’s object-based reference
simulation sometimes interacts with mental rotation frame. When one moves the object by hand, this coupled
receives further support from behavioral studies that reference frame is updated relative to the larger environ-
have identified interactions between mental and manual mental reference frame. It is well established that people
rotation (Wexler, Kosslyn, & Berthoz, 1998; Wohlschläger can simulate the updating of effector-based references
& Wohlschläger, 1998). In these experiments, partici- frames; this type of updating is at the core of motor im-
pants were asked to mentally rotate a picture in order to agery ( Jeannerod & Frak, 1999). Adult humans have ex-
make a spatial judgment, while turning a knob or tensive experience with grasping and carrying. For some
joystick. Manual rotation in the same direction as the mental rotation tasks, people may tend to bring this ex-
mental rotation sped response times, whereas manual perience to bear, updating not just an object-based refer-
rotation in the opposite direction slowed response ence frame but also a coupled object- and effector-based
times. This interaction between motor performance frame. This may be advantageous because it allows one to
and mental rotation indicates that when motor simula- bring additional computational resources to bear. Large
tion arises in mental imagery tasks, it is not merely numbers of cells in the precentral cortex are known to
epiphenomenal; instead, motor processes interact with code locations in effector-based coordinates (Colby,
visuospatial processes. 1998). Therefore, during some mental imagery tasks,
How are we to understand such interactions compu- people may update effector-based reference frames in
tationally? One way is to analyze mental imagery tasks in addition to object-based reference frames, leading to ac-
terms of the spatial reference frames on which they tivation in the precentral cortex.2 In short, the nervous

6 Journal of Cognitive Neuroscience Volume 20, Number 1

system has specialized mechanisms for updating effector- 14. Kosslyn, S. M., Thompson, W. L., Wraga, M., &
centered reference frames. When those mechanisms can Alpert, N. M. (2001). Imagining rotation by
be used to help solve mental rotation problems, this in- endogenous versus exogenous forces: Distinct
creases activity in motor regions of the brain. neural mechanisms. NeuroReport, 12, 2519–2525.
15. Ng, V. W., et al. (2001). Identifying rate-limiting
nodes in large-scale cortical networks for visuospatial
processing: An illustration using fMRI. Journal of
APPENDIX A Cognitive Neuroscience, 13, 537–545.
16. Podzebenko, K., Egan, G. F., & Watson, J. D.
Articles in the meta-analysis
(2002). Widespread dorsal stream activation during a
1. Alivisatos, B., & Petrides, M. (1997). Functional parametric mental rotation task, revealed with functional
activation of the human brain during mental magnetic resonance imaging. Neuroimage, 15, 547–558.
rotation. Neuropsychologia, 35, 111–118. 17. Podzebenko, K., Egan, G. F., & Watson, J. D. G.
2. Barnes, J., et al. (2000). Cortical activity during (2005). Real and imaginary rotary motion processing:
rotational and linear transformations. Functional parcellation of the human parietal lobe
Neuropsychologia, 38, 1148–1156. revealed by fMRI. Journal of Cognitive Neuroscience,
3. Berns, G. S., Chappelow, J. D., & Zink, C. F. (2005). 17, 24–36.
Neurobiological correlates of social conformity and 18. Richter, W., et al. (2000). Motor area activity during
independence during mental rotation. Biological mental rotation studied by time-resolved single-trial fMRI.
Psychiatry, 58, 245–253. Journal of Cognitive Neuroscience, 12, 310–320.
4. Carpenter, P. A., Just, M. A., Keller, T. A., Eddy, W., 19. Richter, W., Ugurbil, K., Georgopoulos, A., & Kim,
& Thulborn, K. (1999). Graded functional activation S.-G. (1997). Time-resolved fMRI of mental rotation.
in the visuospatial system with the amount of task NeuroReport, 8, 3697–3702.
demand. J Cognitive Neuroscience, 11, 9–24. 20. Silk, T., et al. (2005). Fronto-parietal activation in
5. Cohen, M. S., et al. (1996). Changes in cortical attention-deficit hyperactivity disorder, combined type:
activity during mental rotation. A mapping study using Functional magnetic resonance imaging study.
functional MRI. Brain, 119, 89–100. British Journal of Psychiatry, 187, 282–283.
6. Dietrich, T., et al. (2001). Effects of blood estrogen 21. Tagaris, G. A., Kim, S.-G., Strupp, J. P., Andersen,
level on cortical activation patterns during cognitive P., et al. (1997). Mental rotation studied by functional
activation as measured by functional MRI. magnetic resonance imaging at high field (4 Tesla):
Neuroimage, 13, 425–432. Performance and cortical activation. Journal of
7. Halari, R., et al. (2006). Comparable fMRI activity Cognitive Neuroscience, 9, 419–432.
with differential behavioural performance on mental 22. Tagaris, G. A., et al. (1998). Functional magnetic
rotation and overt verbal fluency tasks in healthy men resonance imaging of mental rotation and memory
and women. Experimental Brain Research, 169, 1–14. scanning: A multidimensional scaling analysis of brain
8. Harris, I. M., et al. (2000). Selective right parietal activation patterns. Brain Research Reviews, 26, 106–112.
lobe activation during mental rotation: A parametric 23. Thomsen, T., et al. (2000). Functional magnetic
PET study. Brain, 123, 65–73. resonance imaging (fMRI) study of sex differences in a
9. Johnston, S., Leek, E. C., Atherton, C., Thacker, N., mental rotation task. Medical Science Monitor, 6, 1186–1196.
& Jackson, A. (2004). Functional contribution of medial 24. Vanrie, J., Bâeatse, E., Wagemans, J., Sunaert, S.,
premotor cortex to visuo-spatial transformation in & Van Hecke, P. (2002). Mental rotation versus invariant
humans. Neuroscience Letters, 355, 209–212. features in object perception from different viewpoints:
10. Jordan, K., Heinze, H. J., Lutz, K., Kanowski, M., & An fMRI study. Neuropsychologia, 40, 917–930.
Jancke, L. (2001). Cortical activations during the mental 25. Vingerhoets, G., de Lange, F. P., Vandemaele,
rotation of different visual objects. Neuroimage, 13, 143–152. P., Deblaere, K., & Achten, E. (2002). Motor imagery in
11. Just, M. A., Carpenter, P. A., Maguire, M., Diwadkar, mental rotation: An fMRI study. Neuroimage, 17, 1623–1633.
V., & McMains, S. (2001). Mental rotation of objects 26. Vingerhoets, G., et al. (2001). Regional brain activity
retrieved from memory: A functional MRI study of spatial during different paradigms of mental rotation in healthy
processing. Journal of Experimental Psychology: General, volunteers: A positron emission tomography study.
130, 493–504. Neuroimage, 13, 381–391.
12. Just, M. A., et al. (2001). Interdependence of 27. Weiss, E., et al. (2003). Sex differences in brain
nonoverlapping cortical systems in dual cognitive tasks. activation pattern during a visuospatial cognitive task:
Neuroimage, 14, 417–426. A functional magnetic resonance imaging study in
13. Kosslyn, S. M., Digirolamo, G. J., Thompson, W. L., healthy volunteers. Neuroscience Letters, 344, 169–172.
& Alpert, N. M. (1998). Mental rotation of objects 28. Windischberger, C., et al. (2003). Fuzzy cluster
versus hands: Neural mechanisms revealed by positron analysis of high-field functional MRI data. Artificial
emission tomography. Psychophysiology, 35, 151–161. Intelligence in Medicine, 29, 203–223.

Zacks 7
 29. Windischberger, C., Lamm, C., Bauer, H., & Moser, bodies: An fMRI investigation. Neuropsychologia,
E. (2003). Human motor cortex activity during mental 37, 1029–1040.
rotation. Neuroimage, 20, 225–232. 32. Zitzmann, M., Weckesser, M., Schober, O., &
30. Wraga, M., Thompson, W. L., Alpert, N. M., & Nieschlag, E. (2001). Changes in cerebral glucose
Kosslyn, S. M. (2003). Implicit transfer of motor strategies metabolism and visuospatial capability in hypogonadal
in mental rotation. Brain and Cognition, 52, 135–143. males under testosterone substitution therapy.
31. Zacks, J., Rypma, B., Gabrieli, J. D., Tversky, B., Experimental and Clinical Endocrinology and
& Glover, G. H. (1999). Imagined transformations of Diabetes, 109, 302–304.

APPENDIX B
Activation foci entered in the meta-analysis. (See Appendix A for ID numbers. All coordinates reported in Talairach
’88 space. T = transformation-specific; O = omnibus. This table can be downloaded from http://sumsdb.wustl.edu/
sums/directory.do?id=6617254&dir_name=JCogNeuro_07.)
Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

1 115 Table 3 Mental rotation minus mirror image task T N 39 54 45

1 115 Table 3 Mental rotation minus mirror image task T N 13 68 48
1 115 Table 3 Mental rotation minus mirror image task T N 38 32 3
1 115 Table 3 Mental rotation minus mirror image task T N 20 10 47
1 115 Table 3 Mental rotation minus mirror image task T N 8 17 6
2 1152 Table 1 Rotational transformation O N 17 64 53
2 1152 Table 1 Rotational transformation O N 26 72 31
2 1152 Table 1 Rotational transformation O N 40 39 59
2 1152 Table 1 Rotational transformation O N 43 0 42
2 1152 Table 1 Rotational transformation O N 17 61 59
2 1152 Table 1 Rotational transformation O N 38 61 2
3 249 Figure 4 Baseline mental rotation (red areas) O Y 22 1 70
3 249 Figure 4 Baseline mental rotation (red areas) O Y 31 51 70
3 249 Figure 4 Baseline mental rotation (red areas) O Y 23 1 70
3 249 Figure 4 Baseline mental rotation (red areas) O Y 27 48 70
3 249 Figure 4 Baseline mental rotation (red areas) O Y 0 3 55
3 249 Figure 4 Baseline mental rotation (red areas) O Y 47 31 55
3 249 Figure 4 Baseline mental rotation (red areas) O Y 22 69 55
3 249 Figure 4 Baseline mental rotation (red areas) O Y 28 59 55
3 249 Figure 4 Baseline mental rotation (red areas) O Y 45 1 34
3 249 Figure 4 Baseline mental rotation (red areas) O Y 46 12 34
3 249 Figure 4 Baseline mental rotation (red areas) O Y 25 72 34
3 249 Figure 4 Baseline mental rotation (red areas) O Y 30 73 34
3 249 Figure 4 Baseline mental rotation (red areas) O Y 27 90 8
3 249 Figure 4 Baseline mental rotation (red areas) O Y 31 91 8
3 249 Figure 4 Baseline mental rotation (red areas) O Y 0 16 8
4 14 Figure 4 Mental rotation in 40-degree increments O Y 0 9 55
4 14 Figure 4 Mental rotation in 40-degree increments O Y 22 30 57

8 Journal of Cognitive Neuroscience Volume 20, Number 1

APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

4 14 Figure 4 Mental rotation in 40-degree increments O Y 19 28 57

4 14 Figure 4 Mental rotation in 40-degree increments O Y 15 84 55
4 14 Figure 4 Mental rotation in 40-degree increments O Y 12 72 55
4 14 Figure 4 Mental rotation in 40-degree increments O Y 37 81 18
4 14 Figure 4 Mental rotation in 40-degree increments O Y 30 80 18
5 94 Figure 5 Mental rotation minus control for single T Y 46 11 10
subject
5 94 Figure 5 Mental rotation minus control for single T Y 21 98 26
subject
5 94 Figure 5 Mental rotation minus control for single T Y 44 77 26
subject
5 94 Figure 5 Mental rotation minus control for single T Y 43 18 26
subject
5 94 Figure 5 Mental rotation minus control for single T Y 26 39 35
subject
5 94 Figure 5 Mental rotation minus control for single T Y 20 36 35
subject
5 94 Figure 5 Mental rotation minus control for single T Y 17 56 35
subject
5 94 Figure 5 Mental rotation minus control for single T Y 14 60 35
subject
5 94 Figure 5 Mental rotation minus control for single T Y 25 94 35
subject
5 94 Figure 5 Mental rotation minus control for single T Y 12 64 39
subject
5 94 Figure 5 Mental rotation minus control for single T Y 0 97 39
subject
5 94 Figure 5 Mental rotation minus control for single T Y 48 21 39
subject
5 94 Figure 5 Mental rotation minus control for single T Y 30 41 39
subject
5 94 Figure 5 Mental rotation minus control for single T Y 39 11 47
subject
5 94 Figure 5 Mental rotation minus control for single T Y 42 44 47
subject
5 94 Figure 5 Mental rotation minus control for single T Y 34 41 54
subject
5 94 Figure 5 Mental rotation minus control for single T Y 41 17 54
subject
5 94 Figure 5 Mental rotation minus control for single T Y 24 56 68
subject
5 94 Figure 5 Mental rotation minus control for single T Y 27 47 58
subject
6 429 Text Mental rotation O Y 38 76 36

Zacks 9
APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

6 429 Text Mental rotation O Y 38 76 36

7 6 Table 2 Males, rotation minus control T N 34 76 22
7 6 Table 2 Males, rotation minus control T N 32 82 2
7 6 Table 2 Males, rotation minus control T N 34 70 22
7 6 Table 2 Males, rotation minus control T N 36 4 52
7 6 Table 2 Males, rotation minus control T N 32 62 52
7 6 Table 2 Males, rotation minus control T N 26 54 62
7 6 Table 2 Females, rotation minus control T N 28 78 12
7 6 Table 2 Females, rotation minus control T N 38 78 10
7 6 Table 2 Females, rotation minus control T N 34 78 2
7 6 Table 2 Females, rotation minus control T N 38 40 56
7 6 Table 2 Females, rotation minus control T N 12 52 62
7 6 Table 2 Females, rotation minus control T N 16 58 58
8 68 Text Mental rotation O N 30 68 44
9 210 Text Mental rotation minus control T N 5.94 57.3976 51.669
9 210 Text Mental rotation minus control T N 1.98 16.0472 44.627
9 210 Text Mental rotation minus control T N 41.58 17.5304 0.805
10 149 Table 3 3-D object rotation versus controls T N 28 64 48
10 149 Table 3 3-D object rotation versus controls T N 36 52 52
10 149 Table 3 Abstract object rotation versus controls T N 24 64 48
10 149 Table 3 Abstract object rotation versus controls T N 24 56 56
10 149 Table 3 Abstract object rotation versus controls T N 32 48 44
10 149 Table 3 Abstract object rotation versus controls T N 32 4 40
10 149 Table 3 Letter rotation versus controls T N 24 64 48
10 149 Table 3 Letter rotation versus controls T N 24 52 56
11 422 Table 1 Mental rotation O N 52 19 6
11 422 Table 1 Mental rotation O N 51 21 5
11 422 Table 1 Mental rotation O N 29 52 44
11 422 Table 1 Mental rotation O N 29 54 42
11 422 Table 1 Mental rotation O N 3 72 8
11 422 Table 1 Mental rotation O N 26 70 3
11 422 Table 1 Mental rotation O N 32 67 4
11 422 Table 1 Mental rotation O N 34 25 35
11 422 Table 1 Mental rotation O N 32 33 35
11 422 Table 1 Mental rotation O N 0 13 35
11 422 Table 1 Mental rotation O N 37 18 18
11 422 Table 1 Mental rotation O N 40 21 13
11 422 Table 1 Mental rotation O N 40 1 45

10 Journal of Cognitive Neuroscience Volume 20, Number 1

APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

11 422 Table 1 Mental rotation O N 38 2 41

11 422 Table 1 Mental rotation O N 1 16 48
11 422 Table 1 Mental rotation O N 35 8 49
11 422 Table 1 Mental rotation O N 37 5 45
11 422 Table 1 Mental rotation O N 1 2 56
11 422 Table 1 Mental rotation O N 1 55 12
12 497 Table 1 Mental rotation O N 17 60 47
12 497 Table 1 Mental rotation O N 21 60 48
12 497 Table 1 Mental rotation O N 35 53 40
12 497 Table 1 Mental rotation O N 38 47 42
12 497 Table 1 Mental rotation O N 39 17 20
12 497 Table 1 Mental rotation O N 40 18 19
12 497 Table 1 Mental rotation O N 36 27 34
12 497 Table 1 Mental rotation O N 32 24 42
12 497 Table 1 Mental rotation O N 52 29 3
12 497 Table 1 Mental rotation O N 53 18 8
12 497 Table 1 Mental rotation O N 35 1 47
12 497 Table 1 Mental rotation O N 35 3 46
13 157 Table 1 Rotation of Shepard–Metzler cubes minus T N 27 85 28
baseline condition
13 157 Table 1 Rotation of Shepard–Metzler cubes minus T N 29 71 36
baseline condition
13 157 Table 1 Rotation of Shepard–Metzler cubes minus T N 15 68 44
baseline condition
13 157 Table 1 Rotation of Shepard–Metzler cubes minus T N 33 42 40
baseline condition
13 157 Table 1 Rotation of Shepard–Metzler cubes minus T N 39 90 8
baseline condition
13 157 Table 1 Rotation of Shepard–Metzler cubes minus T N 27 79 36
baseline condition
13 157 Table 1 Rotation of Shepard–Metzler cubes minus T N 12 65 48
baseline condition
13 157 Table 1 Rotation of Shepard–Metzler cubes minus T N 35 46 40
baseline condition
13 158 Table 2 Rotation of hand stimuli minus baseline T N 22 83 28
condition
13 158 Table 2 Rotation of hand stimuli minus baseline T N 20 68 44
condition
13 158 Table 2 Rotation of hand stimuli minus baseline T N 38 37 44
condition
13 158 Table 2 Rotation of hand stimuli minus baseline T N 50 32 40
condition

Zacks 11
APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

13 158 Table 2 Rotation of hand stimuli minus baseline T N 29 21 52

condition
13 158 Table 2 Rotation of hand stimuli minus baseline T N 34 2 36
condition
13 158 Table 2 Rotation of hand stimuli minus baseline T N 26 15 4
condition
13 158 Table 2 Rotation of hand stimuli minus baseline T N 20 49 20
condition
13 158 Table 2 Rotation of hand stimuli minus baseline T N 4 74 8
condition
14 2522 Table 2 IA baseline T N 26 60 48
14 2522 Table 2 IA baseline T N 22 2 52
14 2522 Table 2 IA baseline T N 24 78 32
14 2522 Table 2 IA baseline T N 36 80 12
14 2522 Table 2 IA baseline T N 36 88 4
14 2522 Table 2 IA baseline T N 32 32 4
14 2522 Table 2 IA baseline T N 28 36 16
14 2522 Table 2 IA baseline T N 44 20 24
14 2522 Table 2 IA baseline T N 44 58 8
14 2522 Table 2 EA baseline T N 26 56 48
14 2522 Table 2 EA baseline T N 44 22 24
14 2522 Table 2 EA baseline T N 32 76 32
14 2522 Table 2 EA baseline T N 28 76 32
14 2522 Table 2 EA baseline T N 20 4 52
14 2522 Table 2 EA baseline T N 34 90 4
14 2522 Table 2 EA baseline T N 48 56 8
15 540 Table 1 Mental rotation T N 14 64 42
15 540 Table 1 Mental rotation T N 6 64 48
15 540 Table 1 Mental rotation T N 38 42 48
15 540 Table 1 Mental rotation T N 32 72 20
15 540 Table 1 Mental rotation T N 35 72 4
15 540 Table 1 Mental rotation T N 52 11 26
15 540 Table 1 Mental rotation T N 6 58 37
15 540 Table 1 Mental rotation T N 17 69 37
15 540 Table 1 Mental rotation T N 26 78 7
15 540 Table 1 Mental rotation T N 8 70 55
16 552 Table 1 Mental rotation region correlated with T N 30 82 36
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 22 78 54
proportion of stimuli rotated

12 Journal of Cognitive Neuroscience Volume 20, Number 1

APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

16 552 Table 1 Mental rotation region correlated with T N 42 54 46

proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 36 70 32
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 34 56 54
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 36 90 4
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 32 12 56
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 44 12 8
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 4 64 64
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 28 54 44
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 44 82 14
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 42 86 10
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 24 84 42
proportion of stimuli rotated
16 552 Table 1 Mental rotation region correlated with T N 36 20 4
proportion of stimuli rotated
17 31 Table 3 Rotating alphanumerics minus T N 26 86 38
stationary alphanumerics
17 31 Table 3 Rotating alphanumerics minus T N 54 82 10
stationary alphanumerics
17 31 Table 3 Rotating alphanumerics minus T N 52 76 8
stationary alphanumerics
17 31 Table 3 Rotating alphanumerics minus T N 38 90 0
stationary alphanumerics
17 31 Table 3 Stationary alphanumerics minus T N 4 90 10
rotating alphanumerics
17 31 Table 3 Rotating abstracts minus stationary T N 28 60 64
abstracts
17 31 Table 3 Rotating abstracts minus stationary T N 54 82 8
abstracts
17 31 Table 3 Rotating abstracts minus stationary T N 52 80 6
abstracts
17 31 Table 3 Rotating abstracts minus stationary T N 40 88 2
abstracts
18 3700 Figure 2 Rotation-control contrast T Y 21 94 21
18 3700 Figure 2 Rotation-control contrast T Y 22 94 14
18 3700 Figure 2 Rotation-control contrast T Y 15 94 3

Zacks 13
APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

18 3700 Figure 2 Rotation-control contrast T Y 24 94 2

19 312 Figure 3 Mental rotation T Y 25 20 57
19 312 Figure 3 Mental rotation T Y 36 26 57
19 312 Figure 3 Mental rotation T Y 0 7 57
19 312 Figure 3 Mental rotation T Y 32 61 57
19 312 Figure 3 Mental rotation T Y 33 58 57
19 312 Figure 3 Mental rotation T Y 6 66 57
20 283 Text Mental rotation minus baseline, ADHD O N 27 9 54
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 36 9 18
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 45 18 21
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 15 90* 24
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 21 6 51
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 15 75 42
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 30 84 3
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 48 39 3
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 3 27 39
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 51 3 24
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 39 12 80
subjects
20 283 Text Mental rotation minus baseline, ADHD O N 24 27 18
subjects
21 423 Figure 5 Mental rotation O Y 44 27 40
21 423 Figure 5 Mental rotation O Y 12 24 58
21 423 Figure 5 Mental rotation O Y 33 82 31
21 423 Figure 5 Mental rotation O Y 39 15 62
21 423 Figure 5 Mental rotation O Y 42 30 62
21 423 Figure 5 Mental rotation O Y 12 24 58
21 423 Figure 5 Mental rotation O Y 13 59 4
21 423 Figure 5 Mental rotation O Y 44 27 40
21 423 Figure 5 Mental rotation O Y 0 30 16
21 423 Figure 5 Mental rotation O Y 13 59 4
21 423 Figure 5 Mental rotation T Y 35 52 56

14 Journal of Cognitive Neuroscience Volume 20, Number 1

APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

21 423 Figure 5 Mental rotation T Y 33 82 31

21 423 Figure 5 Mental rotation T Y 19 97 2
21 423 Figure 5 Mental rotation T Y 19 97 2
21 423 Figure 5 Mental rotation T Y 35 52 56
21 423 Figure 5 Mental rotation T Y 42 30 62
21 423 Figure 5 Mental rotation T Y 39 15 62
22 112 Figure 1 Visual mental rotation O Y 16 66 60
22 112 Figure 1 Visual mental rotation O Y 25 55 62
22 112 Figure 1 Visual mental rotation O Y 18 61 59
22 112 Figure 1 Visual mental rotation O Y 44 21 57
22 112 Figure 1 Visual mental rotation O Y 49 3 43
23 1192 Table 4 MR task minus control task: males only T N 22 69 42
23 1192 Table 4 MR task minus control task: males only T N 27 50 43
23 1192 Table 5 MR task minus control task: females only T N 22 77 38
23 1192 Table 5 MR task minus control task: females only T N 40 24 22
23 1192 Table 5 MR task minus control task: females only T N 20 58 46
23 1192 Table 5 MR task minus control task: females only T N 38 20 22
24 922 Table 1 Rotation-control contrast T N 27 1 54
24 922 Table 1 Rotation-control contrast T N 51 39 27
24 922 Table 1 Rotation-control contrast T N 25 57 57
24 922 Table 1 Rotation-control contrast T N 37 67 24
24 922 Table 1 Rotation-control contrast T N 13 60 4
24 922 Table 1 Rotation-control contrast T N 39 36 39
24 922 Table 1 Rotation-control contrast T N 34 78 16
24 922 Table 1 Rotation-control contrast T N 3 30 45
25 387 Table 1 Mental rotation T N 42 6 48
25 387 Table 1 Mental rotation T N 15 72 63
25 387 Table 1 Mental rotation T N 18 21 39
25 387 Table 1 Mental rotation T N 18 21 39
25 387 Table 1 Mental rotation T N 6 87 0
25 387 Table 1 Mental rotation T N 57 15 3
25 387 Table 1 Mental rotation T N 3 39 45
25 387 Table 1 Mental rotation T N 9 69 3
25 387 Table 1 Mental rotation T N 12 27 72
25 387 Table 1 Mental rotation T N 37.8 9.1 41.5
25 387 Table 1 Mental rotation T N 14 73.2 51.4
25 387 Table 1 Mental rotation T N 15 17 34.9
25 387 Table 1 Mental rotation T N 15 17 34.9

Zacks 15
APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

25 387 Table 1 Mental rotation T N 6.1 87.7 4.8

25 387 Table 1 Mental rotation T N 51 17.9 0.4
25 387 Table 1 Mental rotation T N 3.4 34.5 41.1
25 387 Table 1 Mental rotation T N 8.7 70.3 6.5
25 387 Table 1 Mental rotation T N 11.4 29.5 61.6
26 1628 Table 2 EH CH T N 17 89 19
26 1628 Table 2 EH CH T N 25 48 31
26 1628 Table 2 EH CH T N 29 50 50
26 1628 Table 2 EH CH T N 24 9 50
26 1628 Table 2 EH CH T N 26 9 52
26 1628 Table 2 EH CH T N 33 85 12
26 1628 Table 2 ET CT T N 4 31 25
26 1628 Table 2 ET CT T N 1 40 27
26 1628 Table 2 ET CT T N 24 69 23
26 1628 Table 2 ET CT T N 24 60 32
26 1628 Table 2 ET CT T N 29 58 53
26 1628 Table 2 ET CT T N 17 67 48
26 1628 Table 2 ET CT T N 25 3 51
26 1628 Table 2 ET CT T N 22 79 22
26 1628 Table 2 ET CT T N 24 81 25
27 171 Text Mental rotation minus control T N 32 24 4
27 171 Text Mental rotation minus control T N 52 4 32
27 171 Text Mental rotation minus control T N 28 8 52
27 171 Text Mental rotation minus control T N 28 8 64
27 171 Text Mental rotation minus control T N 0 16 40
27 171 Text Mental rotation minus control T N 36 44 40
27 171 Text Mental rotation minus control T N 40 40 52
28 217 Figure 9 Mental rotation versus rest condition O Y 29 88 33
28 217 Figure 9 Mental rotation versus rest condition O Y 14 92 33
28 217 Figure 9 Mental rotation versus rest condition O Y 56 2 12
28 217 Figure 9 Mental rotation versus rest condition O Y 45 31 9
28 217 Figure 9 Mental rotation versus rest condition O Y 48 16 20
28 217 Figure 9 Mental rotation versus rest condition O Y 31 79 38
28 217 Figure 9 Mental rotation versus rest condition O Y 31 79 38
29 231 Figure 5 Activation during task processing O Y 42 18 32
29 231 Figure 5 Activation during task processing O Y 39 23 32
29 231 Figure 5 Activation during task processing O Y 45 59 46
29 231 Figure 5 Activation during task processing O Y 33 69 50

16 Journal of Cognitive Neuroscience Volume 20, Number 1

APPENDIX B (continued)

Coordinates
ID Page Table or Figure Condition/Contrast Class Hand-entered? x y z

29 231 Figure 5 Activation during task processing O Y 0 16 50

29 231 Figure 5 Activation during task processing O Y 21 4 55
29 231 Figure 5 Activation during task processing O Y 27 15 55
30 141 Table 2 Hand baseline T N 38 44 12
30 141 Table 2 Hand baseline T N 40 0 28
30 141 Table 2 Hand baseline T N 32 12 4
30 141 Table 2 Hand baseline T N 8 42 16
30 141 Table 2 Hand baseline T N 22 40 24
30 141 Table 2 Object baseline T N 22 64 44
30 141 Table 2 Object baseline T N 46 12 24
30 141 Table 2 Object baseline T N 28 74 28
30 141 Table 2 Object baseline T N 20 62 48
30 141 Table 2 Object baseline T N 34 24 12
30 141 Table 2 Object baseline T N 48 62 8
30 141 Table 2 Object HO baseline T N 10 76 36
30 141 Table 2 Object HO baseline T N 0 62 4
30 141 Table 2 Object HO baseline T N 26 12 52
30 141 Table 2 Object OO baseline T N 24 62 44
30 141 Table 2 Object OO baseline T N 30 70 32
30 141 Table 2 Object OO baseline T N 44 10 24
30 141 Table 2 Object OO baseline T N 36 40 40
31 1035 Figure 4 Inverted upright T Y 45 26 20
31 1035 Figure 4 Inverted upright T Y 44 10 32
31 1035 Figure 4 Inverted upright T Y 48 19 32
31 1035 Figure 4 Inverted upright T Y 3 30 32
31 1035 Figure 4 Inverted upright T Y 19 12 50
31 1035 Figure 4 Inverted upright T Y 43 57 50
31 1035 Figure 4 Inverted upright T Y 12 83 50
31 1035 Figure 4 Inverted upright T Y 12 83 50
31 1035 Figure 4 Inverted upright T Y 12 83 50
31 1035 Figure 4 Inverted upright T Y 12 83 50
31 1035 Figure 4 Inverted upright T Y 12 83 50
32 304 Table 1 Mental rotation O N 28 92 8

MR = mental rotation; IA = internal action; EA = external action; ADHD = attention deficit disorder; EH = experimental hands; CH = control
hands; ET = experimental tools; CT = control tools; HO = hand-object; OO = object-object.
*This coordinate appears in the original publication as 15, 90, 24; this erratum was corrected for the analyses.

Zacks 17
Acknowledgments to mental rotation demonstrated with rTMS. Journal of
Cognitive Neuroscience, 15, 1–9.
Ty Fagan and Corey Maley assisted with the collection and clas- Jeannerod, M., & Frak, V. (1999). Mental imaging of motor
sification of the data for the meta-analysis. John Harwell and activity in humans. Current Opinion in Neurobiology, 9,
David Van Essen provided valuable assistance with making the 735–739.
foci data available online. I thank them, and also Larry Snyder for Just, M. A., & Carpenter, P. A. (1985). Cognitive coordinate
helpful discussion. systems: Accounts of mental rotation and individual
Reprint requests should be sent to Jeffrey M. Zacks, Departments differences in spatial ability. Psychological Review, 92,
of Psychology and Radiology, Washington University in Saint 137–172.
Louis, St. Louis, MO, or via e-mail: [email protected]. Karnath, H.-O., Ferber, S., & Himmelbach, M. (2001). Spatial
awareness is a function of the temporal not the posterior
parietal lobe. Nature, 411, 950–952.
Notes Kosslyn, S. M. (1994). Image and brain: The resolution of
the imagery debate. Cambridge: MIT Press.
1. Using this meta-analysis technique, it is possible that such Kosslyn, S. M., DiGirolamo, G. J., Thompson, W. L., & Alpert,
patterns can arise due to incidental differences between a small N. M. (1998). Mental rotation of objects versus hands: Neural
number of studies, if each study contributes several nearby foci mechanisms revealed by positron emission tomography.
to the analysis. This was not the case in the present instance: In Psychophysiology, 35, 151–161.
the left hemisphere, 12 separate studies were represented; in Kosslyn, S. M., Thompson, W. L., Wraga, M., & Alpert, N. M.
the right hemisphere, 8 separate studies were represented. (2001). Imagining rotation by endogenous versus exogenous
2. The parietal cortex and the frontal cortex contain cells that forces: Distinct neural mechanisms. NeuroReport, 12,
code information that can be read out in effector-based coor- 2519–2525.
dinates (Andersen & Buneo, 2002; Colby, 1998). Therefore, it is Michelon, P., Vettel, J. M., & Zacks, J. M. (2006). Lateral
possible that some of the variations in activation in parietal somatotopic organization during imagined and prepared
regions across mental rotation paradigms reflect the engage- movements. Journal of Neurophysiology, 95, 811–822.
ment of these representations. Picard, N., & Strick, P. L. (2001). Imaging the premotor
areas. Current Opinion in Neurobiology, 11, 663–672.
Podzebenko, K., Egan, G. F., & Watson, J. D. (2002).
REFERENCES Widespread dorsal stream activation during a parametric
mental rotation task, revealed with functional magnetic
Andersen, R. A., & Buneo, C. A. (2002). Intentional maps in resonance imaging. Neuroimage, 15, 547–558.
posterior parietal cortex. Annual Review of Neuroscience, Poldrack, R. A. (2006). Can cognitive processes be inferred
25, 189–220. from neuroimaging data? Trends in Cognitive Sciences,
Cohen, M. S., & Bookheimer, S. Y. (1994). Localization of 10, 59–63.
brain function using magnetic resonance imaging. Trends Pylyshyn, Z. W. (1981). The imagery debate: Analogue media
in Neurosciences, 17, 268–277. versus tacit knowledge. Psychological Review, 88, 16–45.
Colby, C. L. (1998). Action-oriented spatial reference frames Pylyshyn, Z. W. (2003). Return of the mental image: Are
in cortex. Neuron, 20, 15–24. there really pictures in the brain? Trends in Cognitive
Colby, C. L., & Goldberg, M. E. (1999). Space and attention Sciences, 7, 113–118.
in parietal cortex. Annual Review of Neuroscience, 22, Ratcliff, G. (1979). Spatial thought, mental rotation and the
319–349. right cerebral hemisphere. Neuropsychologia, 17, 49–54.
Cooper, L. A., & Shepard, R. N. (1973). The time required Rizzolatti, G., Fadiga, L., Gallese, V., & Fogassi, L. (1996).
to prepare for a rotated stimulus. Memory & Cognition, Premotor cortex and the recognition of motor actions.
1, 246–250. Cognitive Brain Research, 3, 131–141.
de Lange, F. P., Hagoort, P., & Toni, I. (2005). Neural Robertson, L. C. (2004). Space, objects, minds and brains.
topography and content of movement representations. New York: Psychology Press.
Journal of Cognitive Neuroscience, 17, 97–112. Schubotz, R. I., & von Cramon, D. Y. (2003). Functional–
Dum, R. P., & Strick, P. L. (2002). Motor areas in the frontal anatomical concepts of human premotor cortex: Evidence
lobe of the primate. Physiology and Behavior, 77, 677–682. from fMRI and PET studies. Neuroimage, 20(Suppl. 1),
Farah, M. J. (1989). The neuropsychology of mental imagery. S120–S131.
In F. Boller & J. Grafman (Eds.), Handbook of Shepard, R. N., & Cooper, L. A. (1982). Mental images and
neuropsychology (pp. 395–413). Amsterdam: Elsevier. their transformations. Cambridge: MIT Press.
Ganis, G., Keenan, J. P., Kosslyn, S. M., & Pascual-Leone, A. Shepard, R. N., & Metzler, J. (1971). Mental rotation of
(2000). Transcranial magnetic stimulation of primary three-dimensional objects. Science, 171, 701–703.
motor cortex affects mental rotation. Cerebral Cortex, 10, Snyder, L. H., Grieve, K. L., Brotchie, P., & Andersen,
175–180. R. A. (1998). Separate body- and world-referenced
Gauthier, I., Hayward, W. G., Tarr, M. J., Anderson, A. W., representations of visual space in parietal cortex.
Skudlarski, P., & Gore, J. C. (2002). BOLD activity during Nature, 394, 887–891.
mental rotation and viewpoint-dependent object Tagaris, G. A., Kim, S.-G., Strupp, J. P., & Andersen, P. (1996).
recognition. Neuron, 34, 161–171. Quantitative relations between parietal activation and
Georgopoulos, A. P., Lurito, J. T., Petrides, M., Schwartz, A. B., performance in mental rotation. NeuroReport, 7, 773–776.
& Massey, J. T. (1989). Mental rotation of the neuronal Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxic
population vector. Science, 243, 234–236. atlas of the human brain: 3-Dimensional proportional
Harris, I. M., Egan, G. F., Sonkkila, C., Tochon-Danguy, system, an approach to cerebral imaging. Stuttgart:
H. J., Paxinos, G., & Watson, J. D. G. (2000). Selective Thieme.
right parietal lobe activation during mental rotation: Tomasino, B., Borroni, P., Isaja, A., & Ida Rumiati, R. (2005).
A parametric PET study. Brain, 123, 65–73. The role of the primary motor cortex in mental rotation:
Harris, I. M., & Miniussi, C. (2003). Parietal lobe contribution A TMS study. Cognitive Neuropsychology, 22, 348–363.

18 Journal of Cognitive Neuroscience Volume 20, Number 1

Turkeltaub, P. E., Eden, G. F., Jones, K. M., & Zeffiro, T. A. Wexler, M., Kosslyn, S. M., & Berthoz, A. (1998). Motor
(2002). Meta-analysis of the functional neuroanatomy of processes in mental rotation. Cognition, 68, 77–94.
single-word reading: Method and validation. Neuroimage, Wohlschläger, A., & Wohlschläger, A. (1998). Mental and
16, 765–780. manual rotation. Journal of Experimental Psychology:
Van Essen, D. C. (2005). A Population-Average, Landmark- Human Perception and Performance, 24, 397–412.
and Surface-based (PALS) atlas of human cerebral cortex. Wraga, M., Thompson, W. L., Alpert, N. M., & Kosslyn, S. M.
Neuroimage, 28, 635–662. (2003). Implicit transfer of motor strategies in mental
Van Essen, D. C., Drury, H. A., Dickson, J., Harwell, J., Hanlon, rotation. Brain and Cognition, 52, 135–143.
D., & Anderson, C. H. (2001). An integrated software suite Zacks, J. M., & Michelon, P. (2005). Transformations
for surface-based analyses of cerebral cortex. Journal of the of visuospatial images. Behavioral and Cognitive
American Medical Informatics Association, 8, 443–459. Neuroscience Reviews, 4, 96–118.
Vingerhoets, G., de Lange, F. P., Vandemaele, P., Deblaere, K., Zacks, J. M., Ollinger, J. M., Sheridan, M. A., & Tversky, B.
& Achten, E. (2002). Motor imagery in mental rotation: (2002). A parametric study of mental spatial transformations
An fMRI study. Neuroimage, 17, 1623–1633. of bodies. Neuroimage, 16, 857–872.

Zacks 19