Download Where is a Nose with Respect to a Foot? The Left

Cerebral Cortex December 2008;18:2879--2890
doi:10.1093/cercor/bhn046
Advance Access publication April 18, 2008
Where is a Nose with Respect to a Foot?
The Left Posterior Parietal Cortex
Processes Spatial Relationships among
Body Parts
Corrado Corradi-Dell’Acqua1,2, Maike D. Hesse2,3,4, Raffaella
I. Rumiati1 and Gereon R. Fink2,4,5
Neuropsychological studies suggest that patients with left parietal
lesions may show impaired localization of parts of either their own
or the examiner’s body, despite preserved ability to identify isolated
body parts. This deficit, called autotopagnosia, may result from
damage to the Body Structural Description (BSD), a representation
which codes spatial relationships among body parts. We used
functional magnetic resonance imaging to identify the neural
mechanisms underlying the BSD. Two human body or building parts
(factor: STIMULI) were shown to participants who either identified
them or evaluated their distance (factor: TASK). The analysis of the
interaction between STIMULI and TASK, which isolates the neural
mechanism underlying BSD, revealed an activation of left posterior
intraparietal sulcus (IPS) when the distance between body parts
was evaluated. The results show that the left IPS processes
specifically the information about spatial relationships among body
parts and thereby suggest that damage to this area may underlie
autotopagnosia.
Evidence of such body model is provided, for instance, by
studies investigating the perception of complex cutaneous
stimuli delivered on the body surface: different scholars
showed that these stimuli are processed according not only
to the tactile sensitivity of the body segments being stimulated,
but also to the spatial orientation of these body segments at the
time of the stimulation (Oldfield and Phillips 1983; Parsons and
Shimojo 1987; Haggard et al. 2006). Another example providing
evidence of a model of one’s own body orientation is given by
the rubber hand illusion (Botvinick and Cohen 1998; Tsakiris
and Haggard 2005); such illusion occurs when participants see
a fake but realistically appearing hand, whereas their real hand
is hidden from their sight: when the experimenter brushes
simultaneously both the real and the rubber fingers, the
majority of participants report to feel the brush strokes where
the fake and not the real hand is touched. Moreover, extensive
neurophysiological research in the macaque’s brain (Duhamel
et al. 1998; Graziano et al. 2000), subsequently confirmed in
humans using functional magnetic resonance imaging (fMRI)
(Bremmer et al. 2001; Grefkes et al. 2002), showed that
neurons in the premotor and parietal cortices respond both to
somatosensory stimulation delivered to a given body region and
to visual stimulation from the space adjacent to it, independent
of the respective body part’s position in space. Thus, multisensory integration about one’s own body does not take into
account only information arising from somatosensory and
motor maps, but also signals arising from the visual system.
This then raises question about how visual information of one’s
own body is organized and whether, as for the case of
somatosensory and motor modalities, the visual system is also
endowed with a specific body representation.
The most convincing finding in favor of a visual map of the
human body is offered by the observation of patients affected
by autotopagnosia (Pick 1922; Odgen 1985; Sirigu et al. 1991;
Buxbaum and Coslett 2001; Felician et al. 2003—see CorradiDell’Acqua and Rumiati 2007, as a review). These patients
typically fail to point to parts of a body irrespective of whether
this is their own, someone else’s, or a line drawing. Interestingly, the errors made by these patients in pointing are
more often spatial than categorical, in that they are more likely
to indicate by mistake those body parts that are spatially near to
the intended target (spatial error: hand / elbow) than those
that serve similar functions (categorical error: elbow / knee).
Furthermore, they are unable to construct, in a puzzle-like
fashion, a full body or a full face by means of tiles depicting
isolated parts. However, they also show a spared performance
when animals or manmade objects parts are used instead, thus
suggesting that their deficit cannot be considered to result
Keywords: autotopagnosia, body schema, body structural description,
extrastriate body area, intraparietal sulcus
Introduction
The human cerebral cortex is endowed with multiple maps of
the body. These maps vary according to the motor output, the
sensory channel from which they extract the information and
the reference frame in which this information is recoded. Using
electrical stimulation on canine (Walshe 1948) and human
(Foerster 1936; Penfield and Rasmussen 1950) cortex, scholars
discovered that the primary motor cortex is organized in
a somatotopically ordered map, in which each body part is
represented approximately proportionally to the number of
muscles innervated. Subsequently, studies using microelectrode stimulations (Mitz and Wise 1987; Luppino et al. 1991),
anatomical tracing methods (He et al. 1993) and positron
emission tomography (Fink et al. 1997) revealed the presence of several nonprimary motor areas, which are also
represented somatotopically. Neurophysiological studies also
showed the presence of a somatosensory map: Penfield and
Rasmussen (1950), for instance, by using cortical electrical
stimulation, discovered that the primary somatosensory cortex is organized somatotypically, in that each body part is
represented proportionally to the density of cutaneous
receptors.
The information coded by these multiple maps is integrated
in a higher order representation of the body that Head and
Holmes (1911) called ‘‘Body Schema,’’ which, in turn, provides
the information about one’s own body’s orientation in space.
Ó The Author 2008. Published by Oxford University Press. All rights reserved.
For permissions, please e-mail: [email protected]
1
Cognitive Neuroscience Sector, Scuola Internazionale
Superiore di Studi Avanzati (SISSA-ISAS), 34014 Trieste, Italy,
2
Institute of Neuroscience and Biophysics, Department of
Medicine, Research Center Jülich, 52425 Jülich, Germany,
3
Department of Neurology—Cognitive Neurology, University
Hospital Aachen, RWTH Aachen, 52074 Aachen, Germany,
4
Brain Imaging Center West, Research Center Jülich, 52425
Jülich, Germany and 5Department of Neurology, University
Hospital Cologne, Cologne University, 50924 Cologne, Germany
from a generalized impairment of spatial abilities. Moreover,
they are able to identify the body parts that they cannot locate,
suggesting that their deficit lies neither at the level of visual
processing of isolated body parts, nor at the level of the
semantic knowledge of the body.
Recent accounts postulate that patients affected by autotopagnosia have a damaged visuospatial map which computes the
spatial arrangement among parts of a standard body, the socalled ‘‘Body Structural Description’’ (BSD) (Buxbaum and
Coslett 2001; Schwoebel and Coslett 2005). Structural description models have 1st been employed in the domain of
object recognition (Marr and Nishihara 1978; Biederman 1987)
and suggest that an object is seen as the collection of its parts,
each of which is coded independently from the others. Thus,
the overall representation of the item can be seen as an
organized model which combines the spatial positions of parts
into a whole (see also Peissig and Tarr 2007, for a review). This
notion has been adopted by Buxbaum, Coslett, and Schwoebel
(Buxbaum and Coslett 2001; Schwoebel and Coslett 2005) to
describe the impaired process of the body representation
observed in autotopagnosic patients: even though these patients are able to process isolated body parts, they have trouble
in coding their spatial relations. Recently, Schwoebel and
Coslett (2005) had 70 stroke patients to perform a task requiring mental spatio-motor transformations of one’s own body (a
hand laterality task—Parsons 1987), a task in which patients’
real hand movements were compared with their imagined
movements in order to assess their ability to simulate actions,
and a pointing task similar to the one used to assess autotopagnosia. Moreover, patients choose from a set of body part
pictures either 1) the 1 that was closer to a target part in
a standard body, or 2) the 1 that served the same function of
a target part. Two patients were impaired in the pointing task
as well as in matching body parts for distance, but performed
flawlessly in all the other tasks. In contrast, 3 patients were
found selectively impaired at matching body parts by function,
whereas thirteen were impaired in either the hand laterality or
real/imagined hand movement task despite being still able to
perform normally on all other tasks. Thus, the coding of the
spatial relations among 2 visually presented body parts can be
dissociated from both the semantic knowledge about those
parts as well as the coding of one’s online posture. Schwoebel
and Coslett (2005), accordingly, advocated the psychological
validity of the BSD.
It remains, however, unclear how the BSD relates to the
brain. Localization based on neuropsychological studies of
patients has proved difficult due to lesion sizes; nevertheless, an
analysis of the lesions of patients with autotopagnosia indicates
that the left posterior parietal cortex is the area responsible for
this deficit (Odgen 1985; Semenza 1988; Denes et al. 2000;
Schwoebel et al. 2001; Guariglia et al. 2002; Felician et al. 2003),
even though a recent group study suggested that the left
middle temporal cortex might be involved as well (Schwoebel
and Coslett 2005). Moreover, fMRI studies associated both
the middle temporal cortex and the posterior parietal cortex of
the left hemisphere with tasks which may tap the BSD. For
instance, Downing et al. (2001) observed the posterior part of
the middle temporal gyrus bilaterally more active when
subjects perceived body parts relative to other kinds of stimuli,
and called this region extrastriate body area (EBA). EBA’s
sensitivity was also found to be larger when the body parts
were shown from an allocentric (e.g., looking at others’ parts)
2880 Neural Correlates of Body Structural Description
d
Corradi-Dell’Acqua et al.
rather than an egocentric (e.g., looking at one’s own toes) view
(Chan et al. 2004; Saxe et al. 2006), thus suggesting that this
region may be involved in a visual model of the body. With
regards to the left posterior parietal cortex, Felician et al.
(2004) engaged healthy participants in a pointing task similar
to the one used to diagnose autotopagnposia, and found an
activation in both the left intraparietal sulcus (IPS) and the left
superior parietal cortex (SPC). The same regions were
documented by Le Clec’H et al. (2000) who asked participants
to code the spatial relationship among parts of the human body
with respect to non-bodily control stimuli (e.g., for the body
parts ‘‘is this body part above or below a shoulder in a standing
body?’’, for the numbers ‘‘is this number greater or smaller
that 5?’’). These neuropsychological and imaging findings
suggest that the neural basis of the BSD may lie either in the
middle temporal cortex (presumably corresponding to EBA) or
in the left posterior parietal cortex (over or around the left
posterior IPS).
In the present study we aimed at directly specifying the
neural mechanisms underlying the BSD. We accordingly designed an fMRI study in which twenty participants performed
the following task: 2 images of body parts, vertically aligned at
various distances to each other, were presented on a screen for
750 ms, followed by a mask and the display of 2 vertical lines.
One of these lines represented the distance between the
midpoints of the 2 body parts on the screen. Participants were
asked to indicate the perceived distance by means of a forced
choice between these 2 lines (see Fig. 1).
As a nonspatial control task, vertically oriented names of
body parts replaced the lines. Participants indicated which
body part had been presented as an image (irrespective of their
distance) by means of a forced choice between the 2 names.
Thus, the neural correlates of the BSD can be identified by
disentangling those neural networks associated with coding
body parts’ spatial relations from those associated with coding
body parts per se. Moreover, such effect needs to be body-part
specific, that is, the neural correlates of BSD should not be
found significantly associated with coding the spatial relations
of 2 nonbody parts, relative to coding the nonbody parts per se.
Thus, as a nonbody control condition, the same tasks were
performed using building parts instead of body parts as stimuli.
This constitutes a 2 factorial design with the factors TASK
(spatial judgment vs. identification task) and STIMULI (human
body parts vs. building parts). The critical analysis relies on the
interaction, in which increases of neural activity associated with
judging the distance between body parts (relative to the
identification thereof) are controlled for task differences per se.
Experimental Procedures
Participants
Twenty subjects (11 males and 9 females, aged 20--34, average
age 23.7) participated in the study. None of the participants
had any history of neurological or psychiatric illness. Informed
consent was obtained from all subjects, who were naı̈ve to
the purpose of the experiment. The study was approved by the
local ethics committee and conducted in accordance with the
Declaration of Helsinki. One participant was excluded from
further analysis due to motion artifacts ( >4 mm) during the
fMRI scanning, thus reducing the number of participants
included in the analysis to nineteen.
Figure 1. Example trial sequences for both kinds of task. Participant were presented for 750 ms with either 2 human body or building parts that were vertically aligned, followed by
a flickering masking stimulus of 250 ms. In the Spatial Judgment task participants had to evaluate the distance between the images perceived, and to indicate which of the 2 subsequently
presented lines depicted this distance. In the Identification task participants were asked to identify the body/buildings parts and to indicate the name of 1 of these.
Stimuli
On each trial, 2 photographs were presented on a computer
screen as endpoints of a virtual vertical line of 3 different
lengths (6.14, 6.89, and 7.64 cm—corresponding to, for
a viewing distance of 29 cm, visual angles of 12.09°, 13.55°,
and 15.01°, respectively). Images depicted either parts of the
human body (i.e., nose, ear, belly-button, foot) or parts of
a building (i.e., chimney pot, roof tiles, window, door). Upper
parts of the body or building were always presented on the
screen above lower parts (e.g., the ear [or the chimney,
respectively] appeared always above the belly-button [or the
door, respectively]). All images had a vertical axis of 2.40 cm
(corresponding to a visual angle of 4.74°), were taken from
a frontal orientation, and presented in black and white on
a white background.
Seven pairs of photographs were chosen for both body and
building parts; these pairs were matched according to their
allocentric distance, that is, according to whether they were in
reality 1) highly distant to each other (e.g., the pairs ‘‘ear /
foot’’, that is the ear displayed above the foot as in the example
provided in Figure 1, and ‘‘nose / foot’’ for human body parts,
and ‘‘chimney / door’’ and ‘‘roof / door’’ for building parts),
or 2) intermediately distant to each other (e.g., ‘‘ear / bellybutton’’, ‘‘nose / belly-button’’ and ‘‘belly-button / foot’’ for
human body parts, and ‘‘chimney / window’’, ‘‘roof /
window’’ and ‘‘window / door’’ for the building parts), or 3)
close to each other (e.g., the pairs ‘‘ear / nose’’ and ‘‘nose /
ear’’ for human body parts, and ‘‘chimney / roof’’ and ‘‘roof /
chimney’’, which were shown in either of the 2 possible
combinations—that is, ear or nose on top of the other).
Experimental Set-Up
Participants lay supine with their head fixated by firm foam
pads and each hand placed on a button-box for manual
responses. Stimuli were presented using Presentation 9.0
(Neurobehavioral Systems, CA) and projected on a screen from
a distance of 29 cm through a mirror fixed to the head coil.
Each experimental trial consisted of the following events
(see Fig. 1): the 2 body/building parts appeared on the screen
for 750 ms, followed by a flickering checkerboard presented
Cerebral Cortex December 2008, V 18 N 12 2881
for 250 ms, thereby masking the image positions. Subsequently,
2 lines appeared 2.25 cm (corresponding to a visual angle of
4.44°) to the right and left of the center of the screen for 1000 ms
(spatial judgment task), followed by a 500 ms white screen.
In the control conditions, 2 words were presented vertically in
the same positions as the lines (identification task). Each trial
lasted 2500 ms. In the spatial judgment task, the length of 1
line corresponded to the distance between the 2 body/building
parts previously seen, whereas that of the other line was either
2.4 cm (corresponding to a visual angle of 4.74°) longer or
shorter (50% of the trials each). Subjects were asked to indicate
via button-press with the right or left hand (50% of the trials
each) which of the 2 lines was consistent with the distance
between the body/building images seen earlier. In the
identification task, the words were the German names of 2 of
the 4 body parts used as visual stimuli (i.e., Ohr [Ear, Lexical
Frequency: 292], Nase [Nose, Lex. Freq.: 211], Bauchnabel
[Belly-Button, Lex. Freq.: 4], and Fuß [Foot, Lex. Freq.: 591]) or
2 of the 4 building parts used as visual stimuli (i.e., Kamin
[Chimney, Lex. Freq.: 52], Dach [Roof, Lex. Freq.: 205], Fenster
[Window, Lex. Freq.: 534], and Tür [Door, Lex. Freq.:
68].—Lexical Frequency values were obtained by the CELEX
Lexical Database [Baayan et al. 1993]). One of the 2 words
referred to either of the 2 body/building parts previously seen;
the other referred to a part which was not shown (see Fig. 1).
Subjects were asked to indicate via button-press with the right
or left hand (50% of the trials each) if the name of 1 of the
parts previously seen was on the right (50% of the trials) or on
the left side (50% of the trials) of the screen. This constitutes
a 2 3 2 factorial design with the factors STIMULI (human body
vs. building parts) and TASK (spatial judgment [SJ] versus
identification [I]) with 4 conditions: 1) BodySJ, participants
performed the spatial judgment task with body parts; 2) BodyI,
identification task with body parts; 3) BuildingSJ, spatial
judgment task with building parts; and 4) BuildingI, identification task with building parts. Within each condition, 84 trials
(7 pairs 3 3 distances on the screen 3 4 repetitions) were
presented grouped into 12 blocks, each consisting of 7 trials.
The blocks were introduced by short instructions informing
the subjects about the upcoming task (1500 ms) and separated
by a blank screen interval of 17.5 s (implicit baseline). The
overall experiment consisted of 48 blocks (12 blocks per
condition 3 4 conditions). Both block order and trial order
within each block were randomized.
Image Acquisition
A Siemens Sonata 1.5-T whole-body MRI scanner was used to
acquire both T1-weighted anatomical images and gradient-echo
planar T2-weighted images with blood oxygenation level-dependent (BOLD) contrast. The scanning sequence was a
trajectory-based reconstruction sequence with a repetition
time (TR) of 3020 ms and an echo time (TE) of 66 ms. Each
volume comprised 30 axial slices with an in-plane resolution of
3.125 3 3.125 mm, a slice thickness of 4 mm, and 0.4 mm
interval between slices. For each subject, 586 volumes were
acquired during the whole experimental session (approximately 28 min). For the anatomical images the following
parameters were used: TR = 2.2 ms, TE = 3.93 ms, inversion
time = 1200 ms, field of view = 256 mm, number of sagittal
slices = 128, slice thickness = 1 mm, interslice gap 0.5 mm, flip
angle = 15°.
2882 Neural Correlates of Body Structural Description
d
Corradi-Dell’Acqua et al.
Image Processing
Statistical analysis was carried out using the general linear
model framework (Friston et al. 1995) implemented in the
SPM2 software package (Wellcome Department of Imaging
Neuroscience, London, UK). For each subject, the 1st 6
volumes were discarded. To correct for subject motion, the
functional images were realigned to a 1st functional image,
normalized to a template based on 152 brains from the
Montreal Neurological Institute (MNI), and then smoothed by
convolution with an 8-mm full-width at half-maximum (FWHM)
Gaussian kernel. A high-pass filter (using a cut-off of 128 s) was
finally applied.
On the 1st level, for each individual subject, we fitted a linear
regression model (general linear model) to the data, by
modeling each block condition of the 2 3 2 design with
a boxcar function convoluted with a canonical hemodynamic
response function. We also included the 6 differential realignment parameters as regressors to control for movementrelated variance. Additionally, in order to single-out artifacts
due to difficulty differences across task conditions, participants
average reaction times (RTs) and Accuracy for each block were
entered as additional regressors of no interest. Specific effects
were tested by applying appropriate contrasts to the parameters estimates for each condition.
On the 2nd level, 1-sample t-tests were performed on the
above contrast images of all subjects to give random-effects
statistical parametric maps (SPMs). Voxels were identified as
significant only if they passed a height threshold of t = 3.61
using an uncorrected voxel-level threshold of P < 0.001 and
a minimum cluster size of at least 140 activated voxels, corresponding to a P < 0.05 corrected at the cluster level.
Additionally, with regards to the interaction term (BodySJ –
BodyI) – (BuildingSJ – BuildingI), we restricted our analysis to
those voxels which exhibited the simple effect BodySJ – BodyI
greater than t = 1.73 (corresponding to an uncorrected voxellevel threshold of P < 0.05). Additionally, voxels that the MNI
tissue probability maps (Evans et al. 1994) reported to be more
probably associated to the white matter/cerebrospinal fluid
than to the gray matter were removed by the mask. Thus the
cluster threshold for the interaction term was of 85 activated
voxels, corresponding to a P < 0.05 corrected for the mask.
The localization of the functional activations with respect to
cytoarchitectonic areas was analyzed based on probabilistic
cytoarchitectonic maps derived from the analysis of cortical
areas in a sample of 10 human post-mortem brains, which were
subsequently normalized to the MNI reference space. The
significant results of the random-effects analysis were compared with the cytoarchitectonic maps using the SPM Anatomy
toolbox (Eickhoff et al. 2005).
Results
Behavioral Results
For each subject, and for each condition, the median value of RTs
of all correct trials and the Accuracies were calculated and used
in a 2 (TASK: spatial judgment and identification task) x 2
(STIMULI: human body and building parts) x 3 (SCREEN
DISTANCE: 6.14, 6.89, and 7.14 cm) x 3 (ALLOCENRIC
DISTANCE: close, intermediate, distant—see Methods section
for a more detailed description) repeated measures ANOVA.
Because SCREEN DISTANCE and ALLOCENTRIC DISTANCE had
more than 2 levels, polynomial contrasts were used in order to
assess linear effects across all factor levels. Statistical analysis was
carried out using SPSS 14.0 software (SPSS Inc, Chertsey, UK).
Overall conditions participants were 91.03% accurate (mean
squares [MSE] 566.83, F1,18 = 8604.56, P < 0.001). A main effect
of STIMULI was found to be significant (MSE = 0.87, F1,18 = 7.09,
P < 0.05), with building parts (89.91 ± 1.06%) eliciting more
errors than body parts (92.16 ± 1.03%). Moreover, a main effect
of SCREEN DISTANCE was found to be significant (MSE = 0.52,
F2,36 = 4.14, P < 0.01): participants were more accurate when
the parts were 6.89 cm apart (92.51 ± 1.04%), then when the
distance was 6.14 cm (91.09 ± 1.16%) or 7.64 cm (89.50 ±
1.25%). Finally, a main effect of ALLOCENTRIC DISTANCE was
found to be significant (MSE = 0.75, F2,36 = 6.51, P < 0.01):
participants were more accurate when the parts chosen were
close to each other in the real body/building (92.93 ±
0.85%—e.g., ‘‘ear’’ / ‘‘nose’’ or ‘‘chimney’’ / ‘‘tiles’’) and made
more errors if these had an intermediate (90.43 ± 1.02%—e.g.,
‘‘ear’’ / ‘‘belly-button’’ or ‘‘chimney’’ / ‘‘windows’’) or large
(89.74 ± 1.41%—e.g., ‘‘ear’’ / ‘‘foot’’ or ‘‘chimney’’ / ‘‘door’’)
distance in the real body/building. Polynomial contrasts insured
that the Accuracy Rates changed consistently with a linear
trend across all the levels of ALLOCENTRIC DISTANCE (MSE =
0.12, F1,18 = 10.56, P < 0.01). Figure 2A,B depicts Accuracy
Rates plotted across the 3 levels of allocentric distance for each
kind of task and stimulus: when human body parts were shown,
the Accuracy decreased linearly with the allocentric distance
irrespectively of the task employed (Fig. 2A), whereas when
building parts were shown the Accuracy decreases linearly
with the allocentric distance only when the identification task
was employed (Fig. 2B). No other main effects or interactions
were found to be significant.
With regards the RTs, over all conditions participants took
737.21 ms on average for delivering a correct response (MSE =
371741577.90, F1,18 = 2242.81, P < 0.001). A main effect of
TASK was found to be significant (MSE = 1933560.02, F1,18 =
50.95, P < 0.001), with participants being slower in the
identification (790.38 ± 15.03 ms) than in the spatial judgment
task (684.04 ± 19.23 ms). A main effect of STIMULI was also
found to be significant (MSE = 19669.38, F1,18 = 6.79, P < 0.05),
with participants being faster with body parts (731.85 ±
15.92 ms) than building parts (742.57 ± 15.84 ms). Finally, the
interaction STIMULI * TASK was found to be significant (MSE =
2402.84, F1,18 = 5.81, P < 0.05), in that during the identification
(but not during the spatial judgment) task, RTs to building parts
were longer than those to body parts (see Table 1).
Neural Activations
Those clusters which survived a threshold of P < 0.05
corrected for multiple comparisons are reported in Table 2.
Main Effect of TASK
Consistent with other studies (Fink et al. 2003), the spatial
judgment task, relative to the identification task [i.e., (BodySJ +
BuildingSJ) – (BodyI + BuildingI)], led to differential activation
of a predominantly right-hemisphere network comprising the
parietal cortex (extending from the postcentral gyrus to the
IPS and the supramarginal gyrus), the middle and the inferior
frontal gyri (extending into Brodmann area 44), the middle
occipital gyrus and the inferior temporal gyrus. Moreover, 2
additional clusters of neural activation were found in the left
postcentral and supramarginal gyri (see also Fig. 3A).
Consistent with Cohen et al. (2003), the identification task,
relative to the spatial judgment task [i.e., (BodyI + BuildingI) –
(BodySJ + BuildingSJ)], led to differential activations in the
lingual gyrus bilaterally, the inferior occipital gyrus (x = –44,
y = –84, z = –14) and extending anteriorly to about y = –42 (see
Fig. 3B), thus including the Visual Word Form Area (a left
hemisphere brain region which is supposed to hold a representation of abstract letter identities invariant of parameters
such as spatial position, size, or font). Additional clusters of
activation were also observed in the left thalamus, the left
inferior frontal gyrus (extending into areas 44 and 45), in the
middle and superior frontal gyri (extending into areas 6 and 4)
and the left middle temporal gyrus.
Main Effect of STIMULI
No differential activation was observed (when applying
a threshold of P < 0.05, corrected for multiple comparisons
for the whole brain) when body parts were presented, relative to buildings parts [i.e., (BodySJ + BodyI) – (BuildingSJ +
BuildingI)]. Following Downing et al. (2001), we had hypothesized that EBA should be activated in our experiment.
Accordingly, a region of interest analysis was conducted based
on the coordinates reported by Downing et al. (2001) (MNI
coordinates: x = ±51, y = –71, z = 1) using spherical regions with
a radius of 8 mm (corresponding to the FWHM Gaussian kernel
used to smooth functional images). This volume of interest
analysis revealed a significant increase in the BOLD-signal in the
left EBA [x = –51, y = –74, z = –2; t(18) = 4.15, Ke = 10, P < 0.05
small volume correction—see Fig. 3C], but not in the right EBA.
The opposite contrast [building parts relative to body
parts—that is, (BuildingSJ + BuildingI) – (BodySJ + BodyI)] led
to significant differential activations bilaterally of the parahippocampal and fusiform gyri (see Fig. 3D), consistent with
previous studies which described the parahippocampal gyrus
as responding strongly to a wide variety of stimuli depicting
places and spatial layouts (both outdoor and indoor scenes)
compared with various nonplace control stimuli (Aguirre et al.
1998; Epstein and Kanwisher 1998). Three additional clusters
of activation were observed: in the middle occipital gyrus
bilaterally and in the left superior occipital gyrus.
Interactions
In order to identify those regions specific for the BSD, we
tested the interaction term (BodySJ – BodyI) – (BuildingSJ –
BuildingI). Moreover, we used a mask which included in the
analysis only those voxels which exhibited a simple effect
BodySJ – BodyI greater than t = 1.73 (corresponding to an
uncorrected voxel-level threshold of P < 0.05). This procedure
allowed us to roule out those brain regions which exhibit the
same BOLD response in the 2 tasks when body parts are used,
thus reflecting uniquely an effect of identifying (but not
judging the distance among) building parts, which is not
central to the main purpose of the study. We found a significant
increase of the BOLD-signal in the caudal portion of the SPC
(x = –22, y = –74, z = 46; Ke = 88, P < 0.05 corrected for the
mask), extending vertically from z = 60 to z = 40, thus including
to the caudal part of the IPS (see Fig. 4A,B) as well as the dorsal
portion of a region in the left superior occipital gyrus which
has been associated with the main effect of building parts
(see Figs 3D and 4C). A percentage signal change analysis
showed that this interaction resulted from an increased neural
activity when judging body parts’ distances (relative to the
Cerebral Cortex December 2008, V 18 N 12 2883
Figure 2. Behavioral results. Accuracies (A, B) and RTs (C, D) with standard error of the mean (SEM) bars plotted against the 3 levels of the Allocentric distance. Black circles
depict the Spatial Judgment Task, whereas white triangles over dashed lines depict the Identification Task. The left column (A, C) represents those conditions in which body parts
are shown, whereas the right column (B, D) represents those conditions in which building parts are shown.
Table 1
Mean RTs for correct responses and mean accuracies as a function of TASK and STIMULUS
(values in parentheses are standard errors)
RTs (ms)
Spatial judgment
Identification
Accuracy (%)
Human body
Building
Human body
Building
684.50 (19.39)
779.19 (16.28)
683.58 (19.84)
801.56 (14.53)
91.45 (1.73)
92.87 (1.10)
90.48 (1.62)
89.34 (1.57)
identification thereof), whereas with building parts no significant difference between the 2 tasks was visible (see Fig. 4D).
The inverse interaction, that is, spatial judgment task
(relative to the identification task) in the building parts
condition, contrasted with the spatial judgment task (relative
to the identification task) in the body parts condition [i.e.,
(BuildingL – BuildingN) – (BodyL – BodyN)], did not reveal any
significant differential increase of brain activity.
Discussion
Coding Space between Screen-Parts, Object-Parts, and
Body-Parts
Our behavioral results suggest a perceptual grouping effect, in
that participants’ assessment of both the identity and the spatial
relations of body parts stimuli was easier when these belong to
body portions which are close to one another (e.g., an ear and
a nose), and become progressively harder the more distant the
body parts were (e.g., an ear and a foot—see Fig. 2), even if
such information is irrelevant for the main purpose of the task.
Perceptual grouping effects have been frequently documented
in experimental psychology. For instance, Baylis and Driver
(1993) had participants assessing the spatial relationship
2884 Neural Correlates of Body Structural Description
d
Corradi-Dell’Acqua et al.
between object parts on a video display, and found that
participants’ response times were faster at judging the distance
between parts of the same object than parts of 2 different
objects. Likewise, Behrmann et al. (1998) and Zemel et al.
(2002) had participants performing a same-different task based
on shape properties of object parts, and found that comparisons among parts of the same object were easier than
comparisons among parts of different objects. Moreover,
Vecera et al. (2001) had participants reporting attributes that
appeared on the same part or on different parts of a single
multipart object and found them being more accurate in
reporting the attributes on the same part than attributes on
different parts. All the studies reviewed above show how
attentional selection does not rely only on scene-based
properties of the display, but also on the representation of
the object structure. Also in our experiment, the images
adopted as stimuli are not processed only according to their
position within the scene, but also within the human body
structure.
Conceptually there may be at least 3 kinds of regions
involved in our study: those involved in coding the spatial
relation between parts of the video display (screen-parts);
those involved in coding the position that the parts used as
stimuli have with respect to the whole object (object-parts);
and those involved in coding the spatial relations between
object-parts, but only when this object is a human body (bodyparts). In our study we can disentangle between regions
coding the relations between body-parts from regions coding
the relations between screen-parts and object-parts. Indeed,
whereas the last 2 sets of regions can be associated with by the
main effect of TASK (spatial judgment vs. identification task),
the former is tested by the interaction term, which verifies
whether the effect of TASK is limited to the case in which the
stimuli are body parts. We found that several regions in the
Table 2
Significant clusters showing an increase in activation during 1) the spatial judgment task relative to the identification task, 2) the identification task relative to the spatial judgment task, 3) perception of
building parts relative to human body parts, and 4) the interaction STIMULI * TASK
Region
Cytoarchitectonic
probabilistic maps
Side
Coordinates
Ke
x
y
z
1) Spatial judgment task versus identification task: (BodySJ þ BuildingSJ) (BodyI þ BuildingI)
Postcentral gyrus
Area 1 [80--100%]
R
Supramarginal gyrus
Area 2 [10--60%]
Midd. frontal gyrus
R
Inf. frontal gyrus (p. orb.)
Inf. frontal gyrus (p. orb.)
Area 44 [40--80%]
R
Inf. temporal gyrus
R
Postcentral gyrus
Area 2 [0--50%]
L
Midd. occipital gyrus
R
Postcentral gyrus
Area 1 [30--80%]
L
Supramarginal gyrus
Area 2 [0--60%]
48
60
44
46
56
64
62
44
44
40
36
28
46
42
10
52
26
82
40
40
64
46
12
2
16
12
44
22
64
50
2) Identification task versus spatial judgment task: (BodyI þ BuildingI) (BodySJ þ BuildingSJ)
Lingual gyrus
Area 18 [0--10%]
R
Inf. occipital gyrus
L
Putamen
L
Thalamus
Inf. frontal gyrus
Area 44 [30--60%]
L
Inf. frontal gyrus (p. tri.)
Area 45 [30--50%]
Sup. frontal gyrus
Area 6 [40--80%]
L
Middle temporal gyrus
L
Precentral gyrus
Area 6 [10--20%]
L
Area 4a [10--30%]
Caudate nucleus (head)
R
16
44
22
20
46
46
6
52
50
42
12
50
84
0
28
16
28
2
42
4
12
2
2
14
66
0
18
28
66
12
50
46
22
3) Building versus human body parts: (BuildingSJ þ BuildingI) (BodySJ þ BodyI)
Parahippocampal gyrus
Fusiform gyrus
Parahippocampal gyrus
Fusiform gyrus
Midd. occipital gyrus
Midd. occipital gyrus
Area 18 [0--20%]
Precuneus
Sup. occipital gyrus
26
28
30
26
40
26
16
20
40
32
48
28
84
94
64
74
20
22
20
28
14
12
42
34
22
74
46
4) Interaction: (BodySJ BodyI) (BuildingSJ BuildingI)
Sup. parietal cortex
R
L
R
L
L
L
2921§
1327§
594§
313§
299§
276{
224{
19 354§
2101§
1880§
938§
750§
661§
171||
1105§
1018§
526§
417§
308§
88**
Note: All clusters survive a threshold of P \ 0.05 corrected for multiple comparisons at the cluster level for the whole brain or for the mask, respectively. Coordinates (in standard MNI space) refer to
maximally activated foci as indicated by the highest T score within an area of activation: x 5 distance (mm) to right (þ) or left () of the midsagittal line; y 5 distance anterior (þ) or posterior () to
the vertical plane through the anterior commissure (AC); z 5 distance above (þ) or below () the intercommissural (AC--PC) line. L and R refer to left and right hemispheres, respectively. §P \ 0.001,
{
P \ 0.01, ||P \ 0.05, and **P \ 0.05 corr. for the mask.
right hemisphere, including the postcentral gyrus, the supramarginal and the intraparietal cortex were activated by spatial
processing of stimuli irrespective of their identity, whereas the
left posterior SPC extending to the IPS was found active only
when the stimuli were body parts. The involvement of regions
in the right hemisphere for scene-based spatial judgments is
consistent with previous fMRI studies which had participants
carrying out perceptual tasks on egocentric (rather than
allocentric) spatial localizations (Vallar et al. 1999; Galati et al.
2000; Committeri et al. 2004; Neggers et al. 2006), and with
patient-studies which reported that lesions in the right
posterior parietal cortex, as well as in the right premotor
cortex, may induce unilateral neglect which, in most cases,
involves a disruption of egocentric representations of space
(Vallar and Perani 1986; Leibovitch et al. 1998; Halligan et al.
2003). Right hemisphere regions were also associated with
object-based spatial judgments. For instance, studies using the
mental rotation paradigm (Shepard and Metzler 1971) found
the right SPC involved in object-based judgments of oblique
stimuli irrespective of whether these were human bodies
(Zacks et al. 2003) or alphabetic stimuli (Harris and Miniussi
2003; see also Parsons 2003). Moreover, Behrmann et al. (2006)
reported the case of a patient with integrative agnosia and
a lesion of the right temporal cortex, who after being trained to
identify abstract objects (each of which was composed of
subparts), was unable to report the correct arrangement of the
parts in the target object, despite being still able to identify
them when presented together with other stimuli.
The Neural Basis of the BSD
Structural description models (Marr and Nishihara 1978;
Biederman 1987) imply that object processing occurs in 2
stages: 1) processing the image features and 2) coding their
spatial relationships. Recent clinical observations suggest that
a human shape may undergo the same 2-stage-processing, only
the latter of which seems to be damaged in autotopagnosic
patients (Sirigu et al. 1991; Buxbaum and Coslett 2001;
Schwoebel and Coslett 2005). Our study extends these clinical
findings by distinguishing between those areas associated with
processing human body parts, and activated in both the spatial
judgment and the identification tasks, and those areas
associated with coding their spatial relations, and activated
only in the spatial judgment, but not in the identification, task.
We found that left EBA was active during visual processing of
body parts irrespective of the task performed (see also Urgesi
et al. 2007, for a similar argument). This result is consistent
with previous imaging studies which described EBA responding
to visual stimuli depicting human bodies or body parts in
Cerebral Cortex December 2008, V 18 N 12 2885
Figure 3. Cortical activations associated with the TASK (A, B) and STIMULI (C, D) main effects: (A) glass-brain images and surface rendering of functional contrasts testing the
spatial judgment task relative to the identification task; (B) surface rendering of functional contrasts testing the identification task relative to the spatial judgment task; (C) surface
rendering, axial (z 5 2) and sagittal (x 5 53) sections of the functional contrast testing human body parts relative to building parts; (D) glass-brain images and surface
rendering of functional contrasts testing building parts relative to human body parts.
a category selective fashion (Downing et al. 2001; Downing,
Chan, et al. 2006). Moreover, repetitive transcranial magnetic
stimulation over EBA has been shown to impair identification of
human body parts (but not face or object parts—Urgesi et al.
2004), thus showing that EBA is not only correlated with, but
also causally involved in visual processing of isolated body
parts, which is a necessary prerequisite for building a structural
description of a human shape. However, the functional pattern
exhibited by EBA has been shown to go beyond the
identification of a specific category of visual stimuli: indeed,
2886 Neural Correlates of Body Structural Description
d
Corradi-Dell’Acqua et al.
this region was found to be involved also in processing either
biological motion (Grossman and Blake 2002; Peelen et al.
2006) or static pictures depicting human bodies in action
(Downing, Peelen, et al. 2006), as well as in blind execution
(either real or imaginary) of simple visually guided movements
(Astafiev et al. 2004).
We also found the posterior portion of the left SPC and IPS
being more active while assessing the spatial relationships
among body parts rather during the identification thereof. This
activation was body parts specific, in that no task difference
Figure 4. Cortical activations associated with the interaction STIMULI * TASK. (A, B) Sagittal (x 5 22) and axial (z 5 40, 46, 52, 58) sections reporting an activation on the
caudal portion of the superior parietal lobe. The activated area extends vertically from z 5 40 to z 5 60, and involves the caudal part of the left SPC as well as the caudal part of
the intraparietal sulcus. (C) Surface rendering of the functional contrasts testing the interaction term (depicted in red) and the main effect building relative to human body parts
(green). (D) The percentage signal changes associated with 2 overlapping regions found significant for the interaction term (left graph—local maxima: 22, 74, 46) and for the
main effect of STIMULI (right graph—local maxima: 16, 64, 42) are also displayed together with standard error of the mean (SEM) bars: turquoise lines refer to human body
parts visual stimuli, whereas red dashed lines refer to building parts. (E) Differential percentage signal changes describing, for each region and for each kind of stimulus, the
Spatial Judgment task minus the Identification task, are displayed together with SEM bars.
was found when nonbody parts were used, and, therefore, it
cannot be considered due to coding the spatial relations among
screen-parts or object-parts. This result is consistent with
previous clinical studies which document patients with lesion
in the left hemisphere (Odgen 1985; Buxbaum and Coslett
2001; Schwoebel et al. 2001) and, more specifically, in the left
posterior parietal cortex (Semenza 1988; Denes et al. 2000;
Guariglia et al. 2002), suffering from autotopagnosia. One of the
striking characteristics of these patients is their preserved
ability to reproduce, in a puzzle-like fashion, the correct
arrangement of sub-components of any kind of objects but
faces (Odgen 1985; Guariglia et al. 2002) and human bodies
(Guariglia et al. 2002). Similarly, the left posterior IPS has been
implicated also in a pointing task similar to the 1 used to
diagnose autotopagnosia (Felician et al. 2004), or in mentally
browsing a visual model of the human body (Le Clec’H et al.
2000). Our results converge, and also extend, the findings
reviewed above by specifying, both anatomically and functionally, a region over and around the left caudal IPS as involved in
the BSD and which, if damaged, may lead to autotopagnosia.
One could argue that the functional pattern of the left
posterior SPC and IPS is inconsistent with these regions being
involved in spatial assessment of body parts, in that building
parts never elicit a weaker response than human body parts. As
shown in Figure 4C, a region in the left superior occipital gyrus,
which has been significantly associated to the main effect of
STIMULI, extends to the caudal IPS partially overlapping the
region which has been associated with the interaction term.
The functional pattern of these 2 overlapping regions is quite
similar in that they both respond more to building than to
human body parts (see Fig. 4D), thus suggesting the neural
activity measured in the SPC and IPS to be confounded by the
activity measured in the neighboring (and overlapping) region.
However, the left caudal SPC and IPS (but not the left superior
occipital gyrus) responds also more when assessing the spatial
relations among human body parts relative to their identification (see Fig. 4E, in which confounds due to a general effect of
STIMULI are hidden); such signal increase is significantly larger
than 0 (as insured by the masking procedure) as well as
significantly larger than the signal increase associated with task
differences using building parts as stimuli (as insured by the
significant interaction effect). Thus, the left caudal SPC and IPS
alone respond more to the spatial judgment than to the
identification task, but only when the stimuli used are body parts.
Do the Left Posterior SPC and IPS Respond to the
Attentional Demands of the Employed Tasks?
The regions in the posterior portion of the left SPC and IPS
found significantly associated with the interaction term have
also been described to activate proportionally to the attentional
demands of visuospatial tasks (Corbetta et al. 2000; Hahn et al.
2006—see also Corbetta and Shulman 2002, as a review).
Cerebral Cortex December 2008, V 18 N 12 2887
Moreover, regions in the IPS have been documented active
when participants performed nonspatial, selective attention
tasks (Coull and Frith 1998; Wojciulik and Kanwisher 1999;
Marois et al. 2000—see also Husain and Nachev 2007, as
a review). One could argue that the sensitivity of these regions
to the interaction term may reflect differential attentional
demands in the 4 conditions. However, this interpretation
seems unlikely. First of all, the design matrix in the 1st level
analysis included participants average RTs and accuracy rates
(which are our only available predictors for testing this
alternative hypothesis) as additional regressors of no interest;
thus, each brain region whose activity was modulated uniquely
by task difficulty should be explained by these regressors and,
therefore, not be revealed by the analysis of the factors
STIMULI and TASK. Second, the functional pattern of the left
posterior SPC and IPS (described in Fig. 4D), is inconsistent
with task difficulty of the 4 block conditions (described by
Table 1). In particular, the Identification Task delays participants significantly more than the Spatial Judgment Task,
especially when building parts are shown; this is inconsistent
with the functional pattern of the left posterior SPC and IPS in
which the Spatial Judgment task is associated with a higher
neural activity than the Identification task when body parts
(but not building parts) are shown.
BSD and One’s Own Body
The functional pattern of the left posterior IPS contrasts
noticeably with previous imaging studies in which regions in
both the anterior and ventral IPS (either bilaterally or in the left
hemisphere only) were found active when participants
experienced the rubber hand illusion (Ehrsson et al. 2004,
2005; Tsakiris et al. 2007), as well as when they carried out
tasks which required them to process inputs arising from
multiple sensory modalities (Bremmer et al. 2001; Grefkes et al.
2002). Taken together these results suggest that, whereas
regions in the anterior and ventral IPS bilaterally take part in
the integration of multimodal information about one’s own
body and body posture (i.e., the body schema), regions in the
left posterior IPS are involved in processing the spatial relations
among parts of an abstract body (i.e., the BSD).
Lesions in the left hemisphere of right-handed patients may
induce ideomotor apraxia, which results in the inability to
imitate meaningless gestures even by using the nonparetic limb
ispilateral to the lesion (De Renzi et al. 1980; De Renzi and
Faglioni 1999; Goldenberg and Strauss 2002). Goldenberg and
Karnath (2006) have recently found a double dissociation
between patients who are unable to imitate hand postures
(which show a lesion over the left temporo-parietal junction
extending to the inferior parietal lobe) and patients who are
unable to imitate finger postures (which show a lesion over the
left inferior frontal gyrus). They suggested that hands (but not
finger) postures can be coded as simple spatial relationships
between a limited set of discrete body parts (e.g. the position of
the hand with respect to eyes, chin, shoulders, etc.—Goldenberg 2002), and that such spatial information may be stored in
the left parietal cortex. Our results converge with this theory
by identifying a region in the left caudal IPS in which spatial
relations among visually presented body parts are specifically
processed. It is, therefore, possible that, during the perception
of a meaningless gesture, the visual information about body
parts spatial relations may be coded by this region, and that
imitation may result in applying this spatial information over
2888 Neural Correlates of Body Structural Description
d
Corradi-Dell’Acqua et al.
portions of one’s own body which are homologous to the ones
seen.
Funding
Deutsche Forschungsgemeinschaft (KFO 112, TP1) to G.F.R.
Notes
We are grateful to out colleagues of the Institute of Neuroscience and
Biophysics, Department of Medicine, Research Center Jülich. Conflict
of Interest : None declared.
Address correspondence to Dr. Corrado Corradi-Dell’Acqua. Cognitive
Neuroscience Sector, Scuola Internazionale Superiore di Studi Avanzati
(SISSA/ISAS), via Beirut 2-4, 34014 Trieste, Italy. Email: [email protected].
References
Aguirre GK, Zarahn E, D’Esposito M. 1998. An area within human
ventral cortex sensitive to ‘‘building’’ stimuli: evidence and implications. Neuron. 21:373--383.
Astafiev SV, Stanley CM, Shulman GL, Corbetta M. 2004. Extrastriate
body area in human occipital cortex responds to the performance
of motor actions. Nat Neurosci. 7:542--548.
Baayan RH, Piepenbrock R, Gulikers L. 1993. The CELEX lexical
database (CD-ROM). Philadelphia (PA): Linguistic Data Consortium,
University of Pennsylvania.
Baylis GC, Driver J. 1993. Visual attention and objects: evidence for
hierarchical coding of location. J Exp Psychol Hum Percept Perform.
19:451--470.
Behrmann M, Peterson MA, Moscovitch M, Suzuki S. 2006. Independent
representation of parts and the relations between them: evidence from
integrative agnosia. J Exp Psychol Hum Percept Perform. 32:1169--1184.
Behrmann M, Zemel RS, Mozer MC. 1998. Object-based attention and
occlusion: evidence from normal participants and a computational
model. J Exp Psychol Hum Percept Perform. 24:1011--1036.
Biederman I. 1987. Recognition-by-components: a theory of human
image understanding. Psychol Rev. 94:115--147.
Botvinick M, Cohen J. 1998. Rubber hands ‘feel’ touch that eyes see.
Nature. 391:756.
Bremmer F, Schlack A, Shah NJ, Zafiris O, Kubischik M, Hoffmann K,
Zilles K, Fink GR. 2001. Polymodal motion processing in posterior
parietal and premotor cortex: a human fMRI study strongly implies
equivalencies between humans and monkeys. Neuron. 29:287--296.
Buxbaum LJ, Coslett HB. 2001. Specialised structural description of
human body parts: Evidence from autotopagnosia. Cogn Neuropsychol. 14:289--306.
Chan AW, Peelen MV, Downing PE. 2004. The effect of viewpoint on
body representation in the extrastriate body area. Neuroreport.
15:2407--2410.
Cohen L, Martinaud O, Lemer C, Lehéricy S, Samson Y, Obadia M,
Slachevsky A, Dehaene S. 2003. Visual word recognition in the left
and right hemispheres: anatomical and functional correlates of
peripheral alexias. Cereb Cortex. 13:1313--1333.
Committeri G, Galati G, Paradis A, Pizzamiglio L, Berthoz A, LeBihan D.
2004. Reference frames for spatial cognition: different brain areas
are involved in viewer-, object-, and landmark-centered judgments
about object location. J Cogn Neurosci. 16:1517--1535.
Corbetta M, Kincade JM, Ollinger JM, McAvoy MP, Shulman GL. 2000.
Voluntary orienting is dissociated from target detection in human
posterior parietal cortex. Nat Neurosci. 3:292--297.
Corbetta M, Shulman GL. 2002. Control of goal-directed and stimulusdriven attention in the brain. Nat Rev Neurosci. 3:201--215.
Corradi-Dell’Acqua C, Rumiati RI. 2007. What the brain knows about
the body: evidence for dissociable representations. Brain development in learning environments: embodied and perceptual advancements. Newcastle (UK): Cambridge Scholars Publishing. p. 50--64.
Coull JT, Frith CD. 1998. Differential activation of right superior parietal
cortex and intraparietal sulcus by spatial and nonspatial attention.
Neuroimage. 8:176--187.
De Renzi E, Faglioni P. 1999. Apraxia. In: Denes G, Pizzamiglio L, editors.
Handbook of clinical and experimental neuropsychology. Hove
(UK): Psychology Press. p. 421--440.
De Renzi E, Motti F, Nichelli P. 1980. Imitating gestures. A quantitative
approach to ideomotor apraxia. Arch Neurol. 37:6--10.
Denes G, Cappelletti JY, Zilli T, Dalla Porta F, Gallana A. 2000. A
category-specific deficit of spatial representation: the case of
autotopagnosia. Neuropsychologia. 38:345--350.
Downing PE, Chan AW, Peelen MV, Dodds CM, Kanwisher N. 2006.
Domain specificity in visual cortex. Cereb Cortex. 16:1453--1461.
Downing PE, Jiang Y, Shuman M, Kanwisher N. 2001. A cortical area
selective for visual processing of the human body. Science.
293:2470--2473.
Downing PE, Peelen MV, Wiggett AJ, Tew BD. 2006. The role of
the extrastriate body area in action perception. Soc Neurosci.
1:52--62.
Duhamel JR, Colby CL, Goldberg ME. 1998. Ventral intraparietal area
of the macaque: congruent visual and somatic response properties.
J Neurophysiol. 79:126--136.
Ehrsson HH, Holmes NP, Passingham RE. 2005. Touching a rubber hand:
feeling of body ownership is associated with activity in multisensory
brain areas. J Neurosci. 25:10564--10573.
Ehrsson HH, Spence C, Passingham RE. 2004. That’s my hand! Activity in
premotor cortex reflects feeling of ownership of a limb. Science.
305:875--877.
Eickhoff SB, Stephan KE, Mohlberg H, Grefkes C, Fink GR, Amunts K,
Zilles K. 2005. A new SPM toolbox for combining probabilistic
cytoarchitectonic maps and functional imaging data. Neuroimage.
25:1325--1335.
Epstein R, Kanwisher N. 1998. A cortical representation of the local
visual environment. Nature. 392:598--601.
Evans AC, Kamber M, Collins DL, MacDonald D. 1994. An MRI based
probabilistic atlas of neuroanatomy. In: Shorvon S, Fish D, Andermann
F, Byder GM, editors. Magnetic resonance scanning and epilepsy. New
York: Plenum. p. 263--274.
Felician O, Ceccaldi M, Didic M, Thinus-Blanc C, Poncet M. 2003.
Pointing to body parts: a double dissociation study. Neuropsychologia. 41:1307--1316.
Felician O, Romaiguère P, Anton J, Nazarian B, Roth M, Poncet M, Roll J.
2004. The role of human left superior parietal lobule in body part
localization. Ann Neurol. 55:749--751.
Fink GR, Frackowiak RS, Pietrzyk U, Passingham RE. 1997. Multiple
nonprimary motor areas in the human cortex. J Neurophysiol.
77:2164--2174.
Fink GR, Marshall JC, Weiss PH, Stephan T, Grefkes C, Shah NJ, Zilles K,
Dieterich M. 2003. Performing allocentric visuospatial judgments
with induced distortion of the egocentric reference frame: an fMRI
study with clinical implications. Neuroimage. 20:1505--1517.
Foerster O. 1936. Motorische Felder und Bahnen. Berlin: Springer.
Friston KJ, Holmes AP, Poline JB, Grasby PJ, Williams SC, Frackowiak RS,
Turner R. 1995. Analysis of fMRI time-series revisited. Neuroimage.
2:45--53.
Galati G, Lobel E, Vallar G, Berthoz A, Pizzamiglio L, Le Bihan D. 2000.
The neural basis of egocentric and allocentric coding of space in
humans: a functional magnetic resonance study. Exp Brain Res.
133:156--164.
Goldenberg G. 2002. Body perception disorders. In: Ramachandran VS,
editor. Encyclopedia of the human brain. San Diego (CA): Academic.
p. 443--458.
Goldenberg G, Karnath H. 2006. The neural basis of imitation is body
part specific. J Neurosci. 26:6282--6287.
Goldenberg G, Strauss S. 2002. Hemisphere asymmetries for imitation of
novel gestures. Neurology. 59:893--897.
Graziano MS, Cooke DF, Taylor CS. 2000. Coding the location of the arm
by sight. Science. 290:1782--1786.
Grefkes C, Weiss PH, Zilles K, Fink GR. 2002. Crossmodal processing of
object features in human anterior intraparietal cortex: an fMRI study
implies equivalencies between humans and monkeys. Neuron.
35:173--184.
Grossman ED, Blake R. 2002. Brain areas active during visual perception
of biological motion. Neuron. 35:1167--1175.
Guariglia C, Piccardi L, Puglisi Allegra MC, Traballesi M. 2002. Is
autotopoagnosia real? EC says yes. A case study. Neuropsychologia.
40:1744--1749.
Haggard P, Kitadono K, Press C, Taylor-Clarke M. 2006. The brain’s
fingers and hands. Exp Brain Res. 172:94--102.
Hahn B, Ross TJ, Stein EA. 2006. Neuroanatomical dissociation between
bottom-up and top-down processes of visuospatial selective
attention. Neuroimage. 32:842--853.
Halligan PW, Fink GR, Marshall JC, Vallar G. 2003. Spatial cognition:
evidence from visual neglect. Trends Cogn Sci. 7:125--133.
Harris IM, Miniussi C. 2003. Parietal lobe contribution to mental rotation demonstrated with rTMS. J Cogn Neurosci. 15:315--323.
He SQ, Dum RP, Strick PL. 1993. Topographic organization of
corticospinal projections from the frontal lobe: motor areas on
the lateral surface of the hemisphere. J Neurosci. 13:952--980.
Head H, Holmes G. 1911. Sensory disturbances from cerebral lesions.
Brain. 34:102--254.
Husain M, Nachev P. 2007. Space and the parietal cortex. Trends Cogn
Sci. 11:30--36.
Le Clec’H G, Dehaene S, Cohen L, Mehler J, Dupoux E, Poline JB,
Lehéricy S, van de Moortele PF, Le Bihan D. 2000. Distinct cortical
areas for names of numbers and body parts independent of language
and input modality. Neuroimage. 12:381--391.
Leibovitch FS, Black SE, Caldwell CB, Ebert PL, Ehrlich LE, Szalai JP.
1998. Brain-behavior correlations in hemispatial neglect using CT
and SPECT: the Sunnybrook Stroke Study. Neurology. 50:901--908.
Luppino G, Matelli M, Camarda RM, Gallese V, Rizzolatti G. 1991.
Multiple representations of body movements in mesial area 6 and
the adjacent cingulate cortex: an intracortical microstimulation
study in the macaque monkey. J Comp Neurol. 311:463--482.
Marois R, Chun MM, Gore JC. 2000. Neural correlates of the attentional
blink. Neuron. 28:299--308.
Marr D, Nishihara HK. 1978. Representation and recognition of the
spatial organization of three-dimensional shapes. Proc R Soc Lond B
Biol Sci. 200:269--294.
Mitz AR, Wise SP. 1987. The somatotopic organization of the
supplementary motor area: intracortical microstimulation mapping.
J Neurosci. 7:1010--1021.
Neggers SFW, Van der Lubbe RHJ, Ramsey NF, Postma A. 2006.
Interactions between ego- and allocentric neuronal representations
of space. Neuroimage. 31:320--331.
Odgen J. 1985. Autotopagnosia: occurrence in a patent without
nominal aphasia and with intact ability to point to parts of animals
and objects. Brain. 108:1009--1022.
Oldfield SR, Phillips JR. 1983. The spatial characteristics of tactile form
perception. Perception. 12:615--626.
Parsons LM. 1987. Imagined spatial transformations of one’s hands and
feet. Cognit Psychol. 19:178--241.
Parsons LM. 2003. Superior parietal cortices and varieties of mental
rotation. Trends Cogn Sci. 7:515--517.
Parsons LM, Shimojo S. 1987. Perceived spatial organization of cutaneous patterns on surfaces of the human body in various positions.
J Exp Psychol Hum Percept Perform. 13:488--504.
Peelen MV, Wiggett AJ, Downing PE. 2006. Patterns of fMRI activity
dissociate overlapping functional brain areas that respond to
biological motion. Neuron. 49:815--822.
Peissig JJ, Tarr MJ. 2007. Visual object recognition: do we know more
now than we did 20 years ago? Annu Rev Psychol. 58:75--96.
Penfield W, Rasmussen T. 1950. The cerebral cortex of man. A clinical
study of localization of function. New York: Macmillan.
Pick A. 1922. Störung der Orientierung am eigenen Körper. Beitrag zur
Lehre von Bewußtsein des eigenen Körpers. Psychol Forsch.
2:303--318.
Saxe R, Jamal N, Powell L. 2006. My body or yours? The effect of visual
perspective on cortical body representations. Cereb Cortex.
16:178--182.
Schwoebel J, Coslett HB. 2005. Evidence for multiple, distinct
representations of the human body. J Cogn Neurosci. 17:543--553.
Schwoebel J, Coslett HB, Buxbaum LJ. 2001. Compensatory coding of
body part location in autotopagnosia: evidence for extrinsic
egocentric coding. Cogn Neuropsychol. 18:363--381.
Cerebral Cortex December 2008, V 18 N 12 2889
Semenza C. 1988. Impairment in localization of body parts following
brain damage. Cortex. 24:443--449.
Shepard RN, Metzler J. 1971. Mental rotation of three-dimensional
objects. Science. 171:701--703.
Sirigu A, Grafman J, Bressler K, Sunderland T. 1991. Multiple
representations contribute to body knowledge processing. Evidence from a case of autotopagnosia. Brain. 114(Pt 1B):629--642.
Tsakiris M, Haggard P. 2005. The rubber hand illusion revisited:
visuotactile integration and self-attribution. J Exp Psychol Hum
Percept Perform. 31:80--91.
Tsakiris M, Hesse MD, Boy C, Haggard P, Fink GR. 2007. Neural
signatures of body ownership: a sensory network for bodily selfconsciousness. Cereb Cortex. 17:2235--2244.
Urgesi C, Berlucchi G, Aglioti SM. 2004. Magnetic stimulation of
extrastriate body area impairs visual processing of nonfacial body
parts. Curr Biol. 14:2130--2134.
Urgesi C, Calvo-Merino B, Haggard P, Aglioti SM. 2007. Transcranial
magnetic stimulation reveals two cortical pathways for visual body
processing. J Neurosci. 27:8023--8030.
2890 Neural Correlates of Body Structural Description
d
Corradi-Dell’Acqua et al.
Vallar G, Lobel E, Galati G, Berthoz A, Pizzamiglio L, Le Bihan D. 1999. A
fronto-parietal system for computing the egocentric spatial frame of
reference in humans. Exp Brain Res. 124:281--286.
Vallar G, Perani D. 1986. The anatomy of unilateral neglect after righthemisphere stroke lesions. A clinical/CT-scan correlation study in
man. Neuropsychologia. 24:609--622.
Vecera SP, Behrmann M, Filapek JC. 2001. Attending to the parts of
a single object: part-based selection limitations. Percept Psychophys. 63:308--321.
Walshe FMR. 1948. Critical studies in neurology. London: E. and S.
Livingstone, Ltd.
Wojciulik E, Kanwisher N. 1999. The generality of parietal involvement
in visual attention. Neuron. 23:747--764.
Zacks JM, Gilliam F, Ojemann JG. 2003. Selective disturbance of
mental rotation by cortical stimulation. Neuropsychologia. 41:
1659--1667.
Zemel RS, Behrmann M, Mozer MC, Bavelier D. 2002. Experiencedependent perceptual grouping and object-based attention. J Exp
Psychol Hum Percept Perform. 28:202--217.