Cerebral Cortex January 2009;19:1--12
doi:10.1093/cercor/bhn067
Advance Access publication April 29, 2008
FEATURE ARTICLE
The Human Dorsal Action Control System
Develops in the Absence of Vision
The primate dorsal pathway has been proposed to compute vision
for action. Although recent findings suggest that dorsal pathway
structures contribute to somatosensory action control as well, it is
yet not clear whether or not the development of dorsal pathway
functions depends on early visual experience. Using functional
magnetic resonance imaging, we investigated the pattern of
cortical activation in congenitally blind and matched blindfolded
sighted adults while performing kinesthetically guided hand movements. Congenitally blind adults activated similar dorsal pathway
structures as sighted controls. Group-specific activations were
found in the extrastriate cortex and the auditory cortex for
congenitally blind humans and in the precuneus and the
presupplementary motor area for sighted humans. Dorsal pathway
activity was in addition observed for working memory maintenance
of kinesthetic movement information in both groups. Thus, the
results suggest that dorsal pathway functions develop in the
absence of vision. This favors the idea of a general mechanism of
movement control that operates regardless of the sensory input
modality. Group differences in cortical activation patterns imply
different movement control strategies as a function of visual
experience.
Keywords: congenitally blind humans, dorsal stream, haptic, neural
plasticity, working memory
Movements are guided by visual, auditory, kinesthetic, and tactile
information. To date, it is still an open question where and how
these different types of information are integrated for coherent
action control. For visual information processing, 2 anatomically
and functionally distinct cortical pathways have been proposed:
a dorsal pathway, which projects from early visual areas to the
posterior parietal cortex and uses vision for space perception,
and a ventral pathway, which projects from early visual areas to
the inferotemporal cortex and computes vision for object
perception (Ungerleider and Mishkin 1982). That is, the 2
pathways encode ‘‘where’’ and ‘‘what’’ an object is. More recently,
the functional role of the dorsal pathway has been reinterpreted
by suggesting that it is involved not in spatial processing per se,
but in the visual guidance and control of actions, specifically
determining ‘‘how’’ to interact with an object (Goodale and
Milner 1992). Both pathways project to the prefrontal cortex
constituting a frontoparietal network for movement control.
An equivalent pathway model has been proposed for the
somatosensory domain stating that somatosensory information
is also segregated along 2 pathways subserving action
(posterior parietal cortex) and perception (posterior parietal
cortex and insula) (Dijkerman and DeHaan 2007). Recently, we
tested this theoretical framework by means of functional
magnetic resonance imaging (fMRI) using a kinesthetic moveÓ The Author 2008. Published by Oxford University Press. All rights reserved.
For permissions, please e-mail: [email protected]
Katja Fiehler1, Michael Burke1, Siegfried Bien2, Brigitte Röder3
and Frank Rösler1
1
Department of Experimental and Biological Psychology,
Department of Neuroradiology, Philipps-University Marburg,
D-35032 Marburg, Germany and 3Department of Biological
Psychology and Neuropsychology, University of Hamburg,
D-20146 Hamburg, Germany
2
ment task (Fiehler et al. 2007, 2008). The results demonstrated
that kinesthetic information is processed in brain regions of
the dorsal pathway, thus, confirming the assumptions of the
somatosensory pathway model. Moreover, we extended the
originally proposed function of the dorsal pathway in online
action control to mnemonic processes related to action control.
In accordance with previous findings on visual (Connolly et al.
2003; Curtis 2006) and tactile (Stoeckel et al. 2003; Kaas et al.
2007) action--related working memory, we found dorsal pathway
activity associated with working memory maintenance of
kinesthetic information. Together, these findings provide converging evidence that the dorsal pathway supports action
control not only on the basis of vision but also more generally
by integrating all movement-relevant signals. It is, however, an
unresolved issue whether visual experience is indispensable to
the development of dorsal pathway functions.
Can the dorsal pathway evolve even in the absence of visual
information? Congenitally blind humans who lack in visual
experience during ontogenesis provide an exceptional opportunity to address this question. Several authors have reported
activity in dorsal pathway structures in blind and sighted
participants in spatial imagery tasks, that is, tasks requiring
space perception (Röder et al. 1997; Vanlierde et al. 2003).
However, to our knowledge, brain activity related to action
control has not yet been investigated systematically in humans
lacking any visual input since birth.
Here, fMRI was applied to decide whether or not the
development of dorsal pathway functions that contribute to
kinesthetic action control depends on intact vision. We
specifically examined if areas of the dorsal pathway are recruited
in the blind in the same manner as previously shown in sighted
individuals (Fiehler et al. 2008) when hand movements are
performed on the basis of kinesthetic information. As early vision
alters sensory perception, like kinesthesia (Imbiriba et al. 2006)
and touch (Röder et al. 2004), as well as cognitive processing
(Cattaneo et al. 2007), we tested congenitally blind humans only.
We found a large activation overlap in areas of the dorsal
pathway for congenitally blind and sighted participants during
performance and working memory maintenance of kinesthetically guided hand movements. Our results imply that the
development of the dorsal action control system is not restricted
to visual experience during ontogenesis. Thus, we suggest
a multimodal cortical network of movement control that evolves
on the basis of visual and nonvisual sensory input signals.
Materials and Methods
Participants
Twelve congenitally blind (6 females; mean age 25 years, range 20--38
years) and twelve sighted (8 females; mean age 25 years, range
19--35 years) adults participated. Written informed consent according
to procedures approved by the local research ethics committee was
obtained from each volunteer. All 12 blind participants were congenitally blind as assessed in an interview and had major retinal
damage, and their blindness was not due to a progressive neurological
disease. Sighted participants were part of a larger sample whose results
had been published by Fiehler et al. (2008) elsewhere. All participants reported no history of medical, neurological, or psychiatric
disorders and were free from prescription medication. Participants
were matched by age (maximum difference ± 3 years) and years of
education (MW ± SD, sighted: 17.3 ± 4.5 years; blind: 17.6 ± 2.7 years).
Handedness was assessed by the Edinburgh Handedness Inventory
(Oldfield 1971) using the adapted version for the blind volunteers
(sighted participants: 12 right handed; congenitally blind participants:
10 right handed, 2 ambidextrous). All participants were blindfolded
throughout the scan.
Stimuli and Experimental Setup
During the whole experiment, congenitally blind (CB) and sighted
control (SC) participants lay supine in the scanner with their upper arm
next to the body and their right forearm flexed to reach the
experimental apparatus with the stylus held in the right hand (Fig. 1B
left). The stimulus set comprised of 29 different line patterns engraved
in small plastic cards (95 3 150 mm; width 3 height) (Fig. 1B right).
There were 2 types of line patterns: one circular and 28 multisegment
patterns. Line length was kept constant for all patterns (120 mm). To
ensure comparable task complexity, each multisegment pattern was
constructed with the same features: 4 segments (1 short, 2 medium,
and 1 long segment of 15, 30, and 45 mm length, respectively)
connected by one acute (45°), one right (90°), and one obtuse (135°)
angle. To ease tracing requirements, the beginning of each line pattern
was located at the left side of the card at a constant height of 75 mm
and marked by a small plastic pin. Three line pattern cards were
attached to a cardholder that was uprightly fixed on a base. The base
was attached comfortably to the abdomen of the participants by 2
hook-and-loop strips. The cardholder was adjustable in 3 directions by
a 120° turn on its own axis so that 3 line patterns could be successively
presented per trial. The cardholder was turned by the experimenter
who stood next to the participant while 2 research assistants sorted and
attached the cards to a card holder in the previously defined trial order
and handed it to the experimenter. Cardholders were exchanged
between trials and during the delay epoch.
Due to the small size of the line patterns participants mainly used finger,
hand, and wrist movements while the upper arm remained stationary.
In addition, head and shoulder motion was restrained by a Tempur foam
head cushion (Tempur Deutschland GmbH, Steinhagen, Germany) and
4 additional arm cushions mounted to the right and left arm.
Figure 1. Experimental protocol, experimental setup, task conditions, and behavioral results. (A) The experimental protocol used in the experiment. Three different line patterns
had to be traced with a stylus and maintained across a variable delay of 7.5--8.5 s. Then, a probe stimulus had to be traced and a motor response had to be given indicating
whether the probe matched one of the stimulus from the previous stimulus set. ITI: inter-trial interval. (B) The experimental setup (left). Task stimuli were different line patterns
engraved in plastic cards. The cards were attached to a triangular card holder which could be turned such that one side was oriented towards the hand of the participant.
Blindfolded participants traced the line patterns with a stylus with their right hand and performed task-related button presses with their left hand. The four task conditions (right).
Task stimuli presented for movement performance contained zero (Set size 0), one (Set size 1), two (Set size 2), or three (Set size 3) multi-segment line patterns and three, two,
one, or zero circular line patterns, respectively. An increasing number of multi-segment line patterns led to an increase in task difficulty. During recognition, one probe stimulus
was presented which matched a stimulus from the previous stimulus set in 50% of the trials. (C) Behavioral data on inverse efficiency scores (the average reaction time divided
by the proportion correct) for the four task conditions (SS0: Set size 0; SS1: Set size 1; SS2: Set size 2; SS3: Set size 3). The behavioral results revealed a task difficulty effect for
congenitally blind (CB, n 5 12) and sighted control (SC, n 5 12) participants, i.e., increased inverse efficiency scores with an increasing number of multi-segment line patterns
(5 Set size).
2 Movement Processing in the Blind
d
Fiehler et al.
Experimental Protocol
The performance of kinesthetically guided hand movements was
embedded in a delayed-recognition task that reliably activates cortical
areas of the dorsal pathway in sighted participants during movement
execution without vision (Fiehler et al. 2008). The time course of the
delayed-recognition task is depicted in Figure 1A. Each trial started with
a variable interval of 0.25 -- 1.00 s followed by a tone (ca. 1 s) indicating
the beginning of the movement-encoding period which always lasted
14 s. During this period, participants traced 3 line patterns with a stylus
in their right hand, one after the other. After movement encoding,
a second tone (ca. 1 s) signaled the beginning of the delay period.
Participants were instructed to maintain a mental representation of the
hand movements across a variable 7.5--8.5 s delay period while the right
hand rested on their abdomen. The jitter at the beginning of each trial
and of the delay period was randomly varied in 250-ms steps. Following
the delay, a third tone (ca. 1 s) signaled the start of the movement
recognition period. Participants traced one probe line pattern and finally
pressed one of two buttons on a response box with their left hand to
indicate whether the last line pattern matched one of the stimulus set
items. The presentation of the probe was followed by a variable intertrial
interval (ITI) together lasting 7.5--9.5 s depending on the length of the 2
applied jitters. The total trial length was 34 s.
The stimulus sets always consisted of 3 stimuli with either none (set size
0), 1 (set size 1), 2 (set size 2), or 3 (set size 3) multisegment patterns and
3, 2, 1, or 0 circles, respectively (Fig. 1B, right). None of the multisegment
patterns was repeated within the same trial. The order of the different
stimulus sets was randomized so that participants did not know how many
complex line patterns were presented during the movement period.
During the movement recognition period, a probe stimulus was presented
which was a multisegment pattern in the set size 1, 2, and 3 conditions
and either a multisegment or a circular pattern in the set size 0 condition.
There were 12 trials per condition and run, and 2 runs per participant,
resulting in 96 trials per participant. To guarantee comparable task
requirements, 2 filler trials per set size 1, 2, and 3 conditions and runs
were inserted with a circle presented as probe. Thus, there was a total of
102 trials per participant. Trials were balanced for the number and
position of complex line patterns and the position of the match stimulus
within the stimulus set. Furthermore, the number of match/nonmatch
motor responses and the direction of the match/nonmatch response
buttons were counterbalanced across the experiment.
Prior to fMRI scanning, participants were trained in the experimental
procedure outside the scanner while lying in a supine position. The
practice session lasted 30--60 min depending on the participants’
performance. After training, participants were able to reliably trace the
3 line pattern cards within 14 s, to inhibit any arm movement during the
maintenance period and to accurately recognize the movement during
the recognition period. Only participants who successfully performed the
practice session participated in the fMRI experiment. Performance criteria
were a percent correct rate of more than 90% for the set size 0 condition
and more than 60 % for the set size 3 condition. Stimuli presented during
the practice session were not used during the scanning session.
Functional MRI Acquisition
Imaging was performed on a 1.5 T GE Signa Scanner (GE Medical Systems,
Milwaukee, WI) equipped with a standard quadrature bird-cage head coil.
A high-resolution 3D set of anatomical images was acquired in the axial
orientation for each participant using a T1-weighted spoiled gradient
echo sequence (SPGR, field of view [FOV] = 240 3 180 mm2, matrix size
256 3 192, time echo [TE] = 6 ms, time repetition [TR] = 33 ms)
covering the whole brain. Functional images were acquired with a T2*weighted gradient echo-planar imaging (EPI) sequence sensitive to blood
oxygenation level--dependent (BOLD) contrasts (TR = 2000 ms; TE =
60 ms; flip angle = 80°; 64 3 64-pixel matrix; voxel size = 3.75 3 3.75 3
5 mm; FOV = 240 3 240 mm2), with each volume consisting of 19
axial slices (inter slice distance = 1 mm).
Data analysis
Behavioral Data
Reaction times and percent correct rates were measured for each
experimental condition (task difficulty: set size 0, 1, 2, 3) in both groups
of participants (CB and SC). Inverse efficiency scores were calculated
by dividing the average reaction time by the proportion correct for
each condition (cf., Townsend and Ashby 1993). This procedure
eliminates possible speed accuracy trade-offs. For statistical analysis,
a repeated-measures analysis of variance (ANOVA) with the withinparticipant factor task difficulty and the between-participant factor
group was computed. Significance level was set at P < 0.001.
Functional MRI Data
The Statistic Parametric Mapping (SPM) package was used for image
processing and statistical analyses (SPM2; http://www.fil.ion.ucl.ac.uk/
spm; Welcome Department of Imaging Neuroscience, London, UK).
The first 4 volumes of each fMRI run were discarded to allow for signal
equilibration. Data preprocessing comprised the following steps:
images of each participant were corrected for motion artifacts, corrected
for slice time acquisition differences using a sinc-interpolation
algorithm, spatially normalized to a standard EPI template as defined
by the Montreal MNI and spatially smoothed using a 4 mm full
width at half maximum (FWHM) Gaussian kernel. The voxel
dimensions of each reconstructed scan were 3 3 3 3 3 mm.
A first-level analysis was performed individually for each participant
based on the general linear model (GLM). For each participant, BOLD
signals increasing with the level of task difficulty (set size 3 > set size
2 > set size 1 > set size 0) were assessed by a parametric analysis,
including both correct and incorrect trials. To this end, the predictor of
interest (movement encoding) was weighted by the logarithm of 2 (set
size 0), 3 (set size 1), 4 (set size 2), and 5 (set size 3) to test which
voxels in the brain positively correlate with the four levels of task
difficulty (parameter). The resultant logarithmic function provides
a better fit to the observed performance data (F = 42.03, P < 0.05; cf.,
Fig. 1C) than a linear function (F = 10.40, P = 0.08). This parametric
variation of movement complexity substantiates the impact of an
observed activation pattern: areas exhibiting increased activity in
response to an increasing number of the to be encoded stimuli are very
likely specifically involved in processing that stimulus (e.g., Druzgal and
D’Esposito 2001). In addition to the predictor of interest (movement
encoding), the GLM model comprised 5 predictors of no interest for
the remaining task periods (movement maintenance and recognition)
and for the filler trials of each task period. Regression coefficients for all
predictors were estimated using least squares within SPM2 (Friston
et al. 1995). Low-frequency signals were suppressed by applying a
1/128 Hz highpass filter.
Following first-level analyses, images of parameter estimates for the
contrast of interest (task difficulty during movement encoding) were
entered into a second-level one-sample t-test (random effects analysis)
for each group separately (within-group analysis). Significance level was
set at P < 0.001, uncorrected. To correct for multiple comparisons
(type I error), a spatial cluster filter technique (Forman et al. 1995)
implemented in AlphaSim was applied to the statistical parametric
maps to produce a corrected significance level of P < 0.05. This
resulted in a minimum cluster size of 464 mm3 for the SC group and
472 mm3 for the CB group. To identify cortical regions significantly
activated in both CB and SC participants, we conducted a conjunction
analysis. Significant group differences in the BOLD signal were detected
by a two-sample t-test (between-group analysis). The resulting
statistical parametric maps were thresholded at P < 0.005, uncorrected,
combined with a spatial cluster threshold of 248 mm3 (Forman et al.
1995) to achieve a corrected significance level of P < 0.05.
Analogous to the data analysis of the movement-encoding period,
cortical regions involved in working memory maintenance of kinesthetic information were assessed by a parametric analysis that
contained the predictor of interest (movement maintenance) weighted
by the logarithm of 2 (set size 0), 3 (set size 1), 4 (set size 2), and 5 (set
size 3). Thus, an increase in the BOLD signal should reflect increasing
memory load. The GLM also comprised 5 predictors of no interest
containing the remaining task periods (movement encoding and
recognition) and the filler trials of each task period. Given estimates
of the temporal smoothness of the hemodynamic response, the
covariate modeling of the first seconds of the delay period would be
contaminated by hemodynamic activity from the movement-encoding
period. To temporally separate the BOLD responses of the movement
encoding and the delay period, the first 3 s of the delay period were not
Cerebral Cortex January 2009, V 19 N 1 3
entered into the analyses (cf. Postle et al. 2000). Regression coefficients
for all predictors were estimated using least squares within SPM2, and
low-frequency signals were suppressed by a 1/128 Hz highpass filter.
Within- and between-group parametric maps were calculated using
random-effects analysis setting the significance level to P < 0.005,
uncorrected. To correct for multiple comparisons, the statistical
parametric maps were thresholded by a minimum cluster size of
1072 mm3 for the SC group and 896 mm3 for the CB group for the
within-group analyses to achieve comparable false-positive probabilities
(P < 0.05, corrected; Forman et al. 1995). The cluster threshold (P <
0.05, corrected) for the between-group analyses was 336 mm3.
The areas showing significant changes in BOLD signal were
characterized in terms of MNI coordinates (x, y, z), cluster size
(number of voxels per cluster), and maximal Z scores. Anatomically
distinct brain areas within volumes of interest were classified by means
of the Anatomy toolbox for SPM2 (Eickhoff et al. 2005; http://www.fzjuelich.de/inb/inb-3//spm_anatomy_toolbox) that is based on histological processing and cytoarchitectonic analyses of 10 postmortem
human brains. The resulting cytoarchitectural areas are probability
maps. This means, volumes of interest are defined statistically; as
a consequence they are not only restricted to one cortical area only but
do also extend into neighboring areas. The probability for a specific
volume of interest, however, is always larger than for the neighboring
ones. Weighted contrast values were calculated for selected volumes of
interest using the MarsBar toolbox for SPM2 (Brett et al. 2002; http://
marsbar.sourceforge.net/).
We investigated the pattern of cortical activation in 12 CB and
12 SC participants while performing kinesthetically guided
hand movements under 4 levels of task difficulty. Task difficulty
was determined by the number of distinct movement
sequences per trial (Fig. 1B). Both groups executed the
movements with their right hand without any visual feedback.
The performance was embedded in a delayed recognition task
where participants had to indicate whether a performed hand
movement matched one of the 3 previous movements (Fig. 1A).
To explore whether the development of the dorsal pathway
functions associated with action control depends on vision, we
first focus on the results of the movement-encoding period.
Then we report the findings of the delay period uncovering the
role of the dorsal pathway in kinesthetic working memory
maintenance and the impact of early visual experience on that
process.
Behavior
Participants’ performance was measured in terms of inverse
efficiency scores (for a definition, see method section). Performance data were clearly affected by task difficulty (Table 1). The
Table 1
Task performance as indicated by inverse efficiency scores for the 4 task conditions in CB and SC
participants
CB
size
size
size
size
0
1
2
3
Cortical Activation
To determine which brain areas are involved in kinesthetic
movement processing, we first focused on the encoding period.
A parametric analysis (set size 3 > set size 2 > set size 1 > set
size 0) for each group was calculated to reveal brain regions
systematically responding to movement difficulty. We then
compared the activation patterns of the CB and the SC group in
order to identify cortical areas recruited in a similar or different
manner for kinesthetic movement control. Secondly, we
investigated brain regions involved in working memory
maintenance of kinesthetic information. To this end, we
calculated an additional parametric analysis (set size 3 > set
size 2 > set size 1 > set size 0) to detect cortical areas showing
an activation increase with increasing memory load.
Brain Regions Active During Execution of Kinesthetically
Guided Hand Movements
Results
Set
Set
Set
Set
repeated-measures ANOVA with the within-participant factor
task difficulty and the between-participant factor group revealed
a significant main effect of task difficulty (F3,66 = 125.89, P <
0.0001) indicating decreasing performance with increasing task
difficulty (Fig. 1C). Task efficiency of CB and SC participants did
not differ (group effect, F1,22 = 0.02, P = 0.88; group 3 task
difficulty interaction, F3,66 = 2.18, P = 0.10).
SC
Mean
SEM
Mean
SEM
21.4
47.2
53.6
53.8
2.1
4.6
4.9
4.4
24.6
41.5
52.4
54.9
2.0
3.3
2.9
3.0
Note: Inverse efficiency scores were calculated by dividing the average reaction time by the
proportion correct. Increasing set size reflects increased level of task difficulty. SEM, standard
error of mean.
4 Movement Processing in the Blind
d
Fiehler et al.
CB Group
CB participants strongly activated the postcentral gyrus in both
hemispheres, mainly covering area 2 of the primary somatosensory cortex. In contrast to the left hemispheric activation,
the right postcentral gyrus activation spread into the secondary
somatosensory cortex and the inferiorly adjacent auditory
cortex. The widely distributed activation of the postcentral
gyrus extended caudally into the anterior part of the intraparietal sulcus, a terminal point of the dorsal pathway. Dorsal
pathway activity was additionally found in the left superior
parietal lobe. Interestingly, CB participants also recruited
a brain region located within the ventral perceptual pathway,
the right inferior temporal gyrus (Table 2 and Fig. 2 upper
row).
SC Group
SC participants showed a more distributed activation pattern
than CB participants. The most prominent activation occurred
in the left and right postcentral gyrus, predominantly in area 2
of the primary somatosensory cortex. Similar to the CB group,
we found dorsal pathway activation in the anterior portion of
the intraparietal sulcus and the superior parietal lobe bilaterally. Kinesthetic movement processing further involved
medial and lateral premotor areas in the left and right
hemisphere, the right middle occipital gyrus, and the cerebellar
vermis (Table 2 and Fig. 2 middle row).
Group Comparison
In the group comparison, we first concentrated on brain areas
commonly activated by CB and SC participants by applying
a conjunction analysis. This yielded the same cortical regions
that were observed in the CB group by the parametric analysis
(regions marked with an asterisk in Table 2). Thus, all cortices
identified for kinesthetic movement processing in the CB
group were also activated in the SC group during this task
(Fig. 2 lower row). There was a significant activation overlap in
Table 2
Cortical activation for kinesthetically guided hand movements responding to task difficulty (P \
0.05, corrected)
Region
CB participants
Postcentral gyrus (SI)/aIPS
Postcentral gyrus/superior
temporal gyrus (AC)
Superior parietal lobe
Inferior temporal gyrus
SC participants
Dorsal premotor cortex (PMd)
Ventral premotor cortex (PMv)
Supplementary motor area
(SMA)/pre-SMA
Ventrolateral prefrontal cortex
Postcentral gyrus (SI)
Postcentral gyrus (SI)/aIPS/angular gyrus
Superior parietal lobe/aIPS
Superior parietal lobe
Middle occipital gyrus
Cerebellar vermis
Coordinates
Cluster
Z
x
y
L
R
L
30
51
51
45
24
24
66
42
15
370
301
39
5.11*
4.99*
3.97*
L
R
12
51
75
63
54
6
18
41
3.58*
4.31*
L
39
27
21
51
45
6
12
3
3
3
6
6
60
60
57
27
27
66
35
222
95
48
81
58
4.29*
4.20*
4.60*
3.87
4.43
4.82*
30
60
33
27
12
36
3
18
27
57
57
75
84
66
6
45
48
66
54
33
30
35
45
191
67
56
23
31
4.14
4.06*
4.43*
4.79*
3.97
3.84
4.67
R
L
R
L
L
L
R
L
R
R
z
Note: Areas showing significant changes in BOLD signal are characterized in terms of peak
coordinates (x, y, z) referring to the MNI stereotactic space, cluster size (number of voxels per
cluster), and maximal Z scores. The asterisk indicates cortical regions that were also detected by
the conjunction analysis (P \ 0.05, corrected) reflecting a significant overlap of the activation in
sighted and congenitally blind participants. SI, primary somatosensory cortex; aIPS, anterior
intraparietal sulcus; AC, auditory cortex; L, left; R, right.
the right inferior temporal gyrus between CB and SC participants but no significant activation in the SC group alone. This
can be ascribed to the fact that SC participants activated
a similar brain region (x/y/z, 51/–57/–15; P < 0.001, uncorrected), but the corresponding cluster size was marginally
smaller than the cluster filter threshold.
Brain areas activated either in the CB or in the SC group
were delineated by a two-sample t-test (Table 3). This analysis
revealed stronger activation in CB participants than in SC
participants in 2 sensory areas, namely, in the left extrastriate
cortex (BA18) and in the posterior portion of the auditory
cortex along the superior temporal gyrus (BA 22) of the left
hemisphere (Fig. 3). The auditory cortex activation was further
classified by means of statistical probability maps (Anatomy
toolbox for SPM2, http://www.fz-juelich.de/inb/inb-3//spm_
anatomy_toolbox). The observed activation cluster belonged to
area TE 1.0 (Morosan et al. 2001), a subregion of the auditory
cortex (Fig. 3). Note that the accuracy of the anatomical
classification is limited by the spatial resolution of the present
fMRI data. Stronger activation for SC than CB participants was
found in the rostral presupplementary motor area and in the
right parieto--occipital fissure extending into the precuneus
(Fig. 3).
Brain Regions Active during Working Memory
Maintenance Of Kinesthetically Guided Hand Movements
The results of the maintenance period are summarized in Table
4 and Figure 4. The parametric analysis used to isolate cortical
areas responding to memory load revealed significant signal
changes for the SC group in the left polar prefrontal cortex (BA
10), the left middorsolateral prefrontal cortex (BA 46), and the
anterior intraparietal sulcus bilaterally with the center of
gravity in the left hemisphere. In the CB group, activity was also
found in the anterior intraparietal sulcus in both hemispheres,
only the left activation cluster, however, reached the cluster
size threshold whereas the right cluster slightly missed it.
Activations in congenitally blind and sighted volunteers
significantly overlapped in the left anterior intraparietal sulcus
(x/y/z, –48/–48/45; P < 0.05, corrected), especially in human
intraparietal area 1 (hIP1; Choi et al. 2006). The left intraparietal activation cluster of the SC group extended more
anteriorly compared to the CB group as evidenced by
a significantly stronger activation in human intraparietal area
2 (hIP2; Choi et al. 2006) for SC participants.
Discussion
The present study investigated whether dorsal pathway
functions develop in the absence of visual experience. We
observed that CB humans activated dorsal pathway structures
while performing kinesthetically guided hand movements. This
dorsal pathway activation substantially overlapped with the
activation pattern observed in SC participants. We also found
group-specific brain activity: the SC group showed stronger
activation than the CB group in the presupplementary motor
area and the right precuneus extending into the parieto-occipital fissure, whereas the CB group showed stronger
activation than the SC group in the left extrastriate cortex
and the left auditory cortex. In both groups, working memory
maintenance of kinesthetic information activated areas of the
dorsal pathway, in particular the left anterior intraparietal
sulcus. In the following, we will first discuss the functional
properties of the cortices involved in kinesthetic movement
control and will then elaborate the present findings on
kinesthetic working memory.
Neural Mechanisms Of Kinesthetic Movement Control
In the present task, complex hand movements were substantially guided by kinesthetic input signals from the hand and
fingers. Additionally, tactile feedback from the fingertips
holding the stylus were used, although most likely to a small
degree. In line with functional imaging studies on tactile object
exploration (Stoeckel et al. 2003) and kinesthetic movement
control (Fiehler et al. 2008), we observed a large bilateral
activation cluster in the primary somatosensory cortex (SI),
especially in area 2, in both CB and SC participants. This agrees
with previous observations showing that area 2 processes
primarily kinesthetic information and, to a lesser extent,
complex cutaneous stimuli, for example, surface curvature
(Bodegard et al. 2001). As expected, kinesthetically guided
movements further activated the superior parietal cortex in
both groups. Research in human (Gerardin et al. 2000) and
nonhuman (Sakata et al. 1973) primates has demonstrated that
the superior parietal cortex implements a more elaborative
processing of kinesthetic movement information. In contrast to
the primary somatosensory cortex, neurons in the superior
parietal cortex respond to multiple joint interactions and
integrate both tactile and kinesthetic information. Dense
corticocortical connections (Jones and Powell 1969) and
similar receptor distributions (Scheperjans et al. 2005) facilitate intense mutual interactions between the primary somatosensory cortex and the superior parietal cortex and their
exchange of information. Thus, area 2 and the superior parietal
Cerebral Cortex January 2009, V 19 N 1 5
Figure 2. Cortical activation during kinesthetic hand movements in congenitally blind and sighted control participants. Group averaged statistical parametric maps showed
a large activation overlap in dorsal and ventral pathway structures between congenitally blind (CB; upper row) and sighted control (SC; middle row) participants. This result was
confirmed by the conjunction analysis (lower row) detecting overlapping brain regions significantly activated in each group (activation of the CB group \ activation of the SC
group). L left; R right.
Table 3
Group comparison for kinesthetically guided hand movements (P \ 0.05, corrected)
Region
Coordinates
CB [ SC
Extrastriate cortex
Auditory cortex
SC [ CB
Presupplementary motor area (pre-SMA)
Precuneus/parieto--occipital fissure
Cluster
Z
27
9
19
15
3.87
3.47
57
27
10
42
2.91
3.58
x
y
z
L
L
9
48
87
21
L
R
6
18
18
69
Note: Areas showing significant changes in BOLD signal are characterized in terms of peak
coordinates (x, y, z) referring to the MNI stereotactic space, cluster size (number of voxels per
cluster), and maximal Z scores. These cortical areas were also detected by the parametric
analysis calculated separately for each group. L, left; R, right.
cortex seem to be involved in the processing of kinesthetic and
tactile information in the present task.
The superior parietal cortex located within the dorsal
pathway is known to be involved in online action control as
well. In previous imaging studies, the superior parietal cortex
has been observed to be active during hand and finger
movements requiring kinesthetic control (Gerardin et al.
2000). Moreover, lesions of posterior parietal areas result in
a severe impairment of exploratory hand movements despite
intact elementary sensory or motor functions, termed tactile
6 Movement Processing in the Blind
d
Fiehler et al.
apraxia (Binkofski et al. 2001). In addition, we observed activity
in a second dorsal pathway region, the anterior part of the
intraparietal sulcus. The anterior intraparietal sulcus has been
discussed as an area that transforms sensory input into motor
commands and continuously updates the ongoing movement
with respect to the aspired movement goal (Desmurget
and Grafton 2000). As a consequence, this area allows for
a flexible control of actions. Due to strong cortical connections
(Rushworth et al. 2006), the information is further transferred
to the ventral premotor cortex important for performing of
precise hand movements. Congruously, we found motorrelated activation in that area.
The localization of the anterior intraparietal sulcus activation
for kinesthetic hand movement control in the present task
significantly overlapped with brain areas reported for visually
guided hand movements (Binkofski et al. 1998; Culham et al.
2003; Frey et al. 2005). Thus, it seems that the anterior
intraparietal sulcus controls action on the basis of both
kinesthetic and visual information. This is in accordance with
the visual (Goodale and Milner 1992) and the somatosensory
(Dijkerman and DeHaan 2007) pathway models which both
hypothesize that the dorsal action stream terminates in the
posterior parietal cortex. Our results together with previous
findings suggest a modality-independent function of the dorsal
pathway, in particular the anterior intraparietal sulcus, in hand
movement control. Future research could further substantiate
Figure 3. Group-specific cortical activation during kinesthetic hand movements. (A) Group averaged statistical parametric maps from the between-group contrasts (two-sample
t-test) revealed higher activation for sighted control (SC) than congenitally blind (CB) participants in the parieto-occipital fissure and the pre-supplementary motor area (upper
row). Congenitally blind compared to sighted control participants showed increased activity in the extrastriate cortex and the auditory cortex. Averaged parameter estimates are
depicted for the group-specific regions. (B) Single-participant fMRI activation in the group-specific regions. The magnitude of activation is illustrated for each one of the
participants for the four group-specific brain areas. (C) Stronger auditory cortex activation in congenitally blind than sighted participants during kinesthetic movement control.
Cortical maps from the between-group analysis (two-sample t-test) are superimposed on a brain map parcellated into different anatomical regions based on histological
processing and cytoarchitectonic analyses of ten post-mortem human brains (http://www.fz-juelich.de/ime/spm_anatomy_toolbox). Based on this brain map, the observed
activation cluster is located in area TE 1.0, a subregion of the auditory cortex (cf., Morosan et al. 2001). L left; R right.
Cerebral Cortex January 2009, V 19 N 1 7
Table 4
Cortical activation for kinesthetic movement maintenance responding to memory load (P \ 0.05,
corrected)
Region
CB participants
Anterior intraparietal sulcus
Thalamus
SC participants
Anterior intraparietal sulcus
Polar prefrontal cortex
Middorsolateral prefrontal cortex
SC [ CB
Anterior intraparietal sulcus
Anterior insula
Coordinates
Cluster
Z
x
y
z
L
R
L
36
39
27
51
63
33
45
45
12
33
21b
45
3.62*
4.46
4.11
L
R
L
L
42
45
48
48
63
63
42
21
36
36
6
36
407
112
56
89
4.48*
4.59
3.91
3.84
L
L
54
30
39
9
42
12
20
25
3.99
3.52
Note: Areas showing significant changes in BOLD signal are characterized in terms of peak
coordinates (x, y, z) referring to the MNI stereotactic space, cluster size (number of voxels per
cluster), and maximal Z scores. The right anterior intraparietal sulcus activation in congenitally
blind participants, marked with b, reached a significance level of P \ 0.001, uncorrected, but
slightly missed the cluster filter threshold. The asterisk indicates cortical regions that were also
detected by the conjunction analysis (P \ 0.05, corrected) reflecting a significant overlap of the
activation in sighted and congenitally blind participants. L, left; R, right.
the assumption of multimodality by directly comparing
different input modalities (e.g., visual vs. kinesthetic) used for
action guidance under specific task instructions within one
experiment.
The key finding of the present study is the substantial
overlap of activation of dorsal pathway structures, the anterior
intraparietal sulcus and the superior parietal cortex, in CB and
SC participants. Because vision is the dominant sense in sighted
humans providing the most reliable spatial information, goaldirected actions are most often predominantly visually guided.
Therefore, it is plausible to assume that early visual experience
is indispensable for the development of parietal systems
responsible for movement control. An effect of early vision
on multisensory action control has recently been demonstrated
by Röder et al. (2004, 2007) in behavioral experiments. They
found that spatial reference frames used for auditory action
control differed for CB and SC participants. However, their data
do not exclude a role of the parietal cortex for action control in
the blind. The present fMRI results suggest that movement
control is mediated by the dorsal pathway regardless of the
sensory input modalities available during development. This is
in accordance with recent findings by Garg et al. (2007) who
observed activity enhancement in a vision-related cortical
region of the dorsal fronto-parietal pathway, the frontal eye
fields, in both congenitally blind and sighted participants
performing a spatial attention task. The present dorsal pathway
activity in congenitally blind adults also supports the argument
that activation of the dorsal pathway observed in haptic
movement tasks in sighted individuals is not solely caused by
visual imagery.
In addition to vision, touch, and kinesthesia, auditory
information contribute to movement control as well. Analogous
to the ‘‘what’’ and ‘‘where’’ segregation of visual information
processing (Ungerleider and Mishkin 1982), a similar functional
organization has been proposed for the auditory modality (for
evidence in monkeys, see, Romanski et al. 1999; Rauschecker
and Tian 2000; for evidence in humans, see, Alain et al. 2001;
Clark et al. 2002). There is evidence that sound localization and
sound identification are functionally distinct along dorsal and
8 Movement Processing in the Blind
d
Fiehler et al.
ventral pathways, respectively. Consequently, greatest spatial
selectivity was found in the caudal aspects of the superior
temporal gyrus, considered as the origin of the dorsal ‘‘where’’
pathway. This area has dense cortical connections to the
posterior parietal cortex (Romanski et al. 1999), again in line
with its role in spatial processing. Thus, multiple sensory inputs
about spatial relations converge in the posterior parietal cortex
and are integrated into a coherent space representation used
for action guidance. This furthermore implies that the dorsal
pathway has access to multimodal information. Evidence for
the involvement of the auditory ‘‘where’’ pathway and its
parietal projection areas in congenitally blind people would
provide further evidence for a modality-independent development of dorsal pathway functions.
Interestingly, in CB participants we found activation in an
area of the ventral pathway, in the right inferior temporal gyrus.
A similar activation was observed in the SC group but the
corresponding cluster size slightly missed the cluster filter
threshold. The existence of reliable inferior temporal cortex
activity in both groups, was substantiated by a significant
activation overlap (x/y/z, 48/–57/–15; P < 0.05, corrected)
detected by the conjunction analysis. This activation partly
overlapped with the rostral part of the object sensitive area
termed lateral occipital complex (LOC; Malach et al. 1995). The
functional specialization of LOC in object recognition was
confirmed by a lesion study that reported visual form agnosia in
a patient with a severe damage in the ventral pathway (James
et al. 2003). Recent imaging experiments have demonstrated
that a region within LOC (LOtv) is not only involved in visual
but also in tactile object recognition (Amedi et al. 2001; Pietrini
et al. 2004). Object-related auditory information failed to
activate LOtv (Amedi et al. 2002). LOtv responded, however, if
shape information was preserved in the auditory signal by using
visual-to-auditory sensory substitution soundscapes (Amedi
et al. 2007). Apparently, LOtv functions as a multisensory
object-selective network that is driven by the presence of
shape information. Based on these findings, we assume that
ventral pathway activation in CB and SC participants is
associated with the encoding and memorizing of the geometrical shape of the movement path in order to successfully
perform the delayed-recognition task. While SC participants
might have used both visual imagery and haptic input signals to
extract shape information, it is evident that CB participants
must have relied on haptic input only supporting the role of
LOtv in visual and haptic object recognition.
Recruitment of Presupplementary Motor Area and
Precuneus in Sighted Humans
Although the dorsal pathway network engaged in the present
task was similar in CB and SC participants, we observed brain
regions selectively recruited by either the one or the other
group as well. In comparison to the CB group, SC participants
showed stronger activation in the pre-supplementary motor
area (pre-SMA) and in the right precuneus extending into the
parieto-occipital fissure. Anatomical studies revealed reciprocal
connections between these two areas (Wise et al. 1997).
Together with the pre-SMA, the precuneus is part of a neural
network functionally specialized for the process of spatially
guided behavior (Selemon and Goldman-Rakic 1988). In
particular, parieto-occipital regions are supposed to compute
egocentric and allocentric spatial coordinates used for body
movement control. This finding was supported by a recent
Figure 4. Cortical activation during working memory maintenance of kinesthetic information in congenitally blind and sighted control participants. (A) Delay-related activity was
found in the anterior intraparietal sulcus (1) bilaterally in congenitally blind (CB) and sighted (SC) participants. Sighted participants additionally activated the left middorsolateral
prefrontal cortex (2) and the left polar prefrontal cortex (3). (B) The conjunction analysis (activation of the CB group \ activation of the SC group) detected a significant activation
overlap in the left anterior intraparietal sulcus (aIPS), in particular in human intraparietal area 1 (hIP1; cf., Choi et al. 2006). (C) Group averaged statistical parametric maps from
the between-group contrasts (two-sample t-test) revealed higher activation for sighted control than congenitally blind participants in the anteriorly located human intraparietal
area 2 (hIP2; cf., Choi et al. 2006). Average parameter estimates are depicted for both intraparietal regions. Cortical maps from the conjunction and between-group analysis (twosample t-test) are superimposed on a brain map parcellated into different anatomical regions based on histological processing and cytoarchitectonic analyses of ten post-mortem
human brains (http://www.fz-juelich.de/ime/spm_anatomy_toolbox). L left; R right.
study in patients with lesions in the occipito-parietal junction
and adjoining precuneus who suffered from optic ataxia
(Karnath and Perenin 2005). Moreover, tactile object localization in contrast to tactile object identification was associated
with greater activation in the precuneus region (Reed et al.
2005). The pre-SMA is proposed to contribute to this process
by implementing higher order aspects of motor control, that is,
updating of motor plans and control of complex movement
sequences (Picard and Strick 1996). Activity in the pre-SMA
systematically varied with movement difficulty, i.e., the
complexity in movement transitions was positively correlated
with the strength of cortical activation (Harrington et al. 2000).
Consistent with this view, we found that three different
complex hand movements (Set size 3) compared to three
simple circular hand movements (Set size 0) led to an increase
in the BOLD signal in the pre-SMA (x/y/z, cluster size, Z score:
–6/21/36, 91 voxels, Z = 3.85, P < 0.05, corrected) for the SC
but not for the CB participants. A similar result was observed
for the occipito-parietal region (x/y/z, cluster size, Z score: 18/
–57/21, 21 voxels, Z = 3.86, P < 0.05, corrected). The pre-SMA,
in contrast to the occipito-parietal region, also showed a linear
increase in activation strength with an increase in movement
difficulty in the SC group, favoring its role in sensorimotor
control processes.
In addition to their function in active movement guidance,
both precuneus and pre-SMA seem to be important for motor
imagery (Gerardin et al. 2000). It has been argued that these
cortical regions are constituent parts of a distinct neural
network of motor imagery engaged in the construction of
internal spatial representations for movement control. Activity
in this network has been found to increase when demands of
locomotor imaging tasks require increasing cognitive and
sensory information processing (Malouin et al. 2003).
Following this argument, the higher activation in the preSMA and occipito-parietal areas in SC compared to CB
participants might be due to higher task-related cognitive and
sensorimotor demands. Because of the dominant visual sense,
sighted individuals are not as experienced as blind humans in
kinesthetic movement control. Therefore, the execution of the
present task might have required more effort for the SC group
reflected by increased activity in brain regions associated with
spatial movement control. Support for this assumption was also
provided by a comparable performance level of both groups
(see behavioral data) and significantly less training time for CB
participants. Alternatively, the group difference might be
attributed to the use of different strategies for movement
processing as congenitally blind in contrast to sighted humans
have been shown to prefer egocentric rather than external
allocentric reference frames for motor control (Röder et al.
2007).
Recruitment of the Extrastriate and Auditory Cortex in
Congenitally Blind Humans
Kinesthetic movement processing elicited stronger activation
in the extrastriate cortex in CB compared to SC participants.
This activation of the occipital cortex in the blind is in
accordance with previous findings showing activation in striate
and extrastriate areas in a variety of non-visual perceptual tasks,
for example, tactile discrimination (Sadato et al. 1996, 2004;
Cohen et al. 1997) and auditory localization (Weeks et al. 2000;
Gougoux et al. 2005), as well as cognitive tasks, for example,
Cerebral Cortex January 2009, V 19 N 1 9
speech processing (Röder et al. 2002) and verbal memory
(Amedi et al. 2003). The present data add to these reports by
demonstrating extrastriate cortex activity in the blind also for
kinesthetic information processing used for action guidance.
Visual deafferentation leads to cross-modal cortical reorganization of the occipital cortex (Röder and Rösler 2004;
Pascual-Leone et al. 2005). Neural transplantation (Schlaggar
and O’Leary 1991) and rewiring studies (Sur and Leamey 2001)
suggest that the occipital cortex tissue can process input from
non-visual modalities, that is, somatosensory and auditory
information. Even transient visual deprivation (five days) in
sighted humans can already induce an activation of the primary
visual cortex for tactile and auditory processing (Pascual-Leone
and Hamilton 2001). Since transcranial magnetic stimulation
(TMS) over the occipital cortex seems to interfere with
perceptual and cognitive processing of blind people (Cohen
et al. 1997; Amedi et al. 2004), it has been argued that this area
may take over non-visual functions in the blind. Although, visual
cortex activity emerges within a few days after blindfolding and
has been found in a great variety of perceptual and cognitive
tasks, its functional significance is still not fully understood.
Moreover, CB participants in contrast to SC participants
exhibited increased activation in the auditory cortex. Somatosensory input to the auditory cortex has recently been
demonstrated in human and non-human primates. Intracranial
recordings in monkeys have shown activation in the caudomedial auditory cortex (CM) triggered by repetitive electrical
stimulation of the median nerve (Schroeder et al. 2001), by
cutaneous stimulation at the head/neck and hand, and by
kinesthetic stimulation at the elbow (Fu et al. 2003). The
timing and laminar activation for somatosensory and auditory
inputs are nearly the same in monkey area CM supporting
somatosensory-auditory convergence in this area. Anatomical
reconstructions, however, indicated that the somatosensory
input region includes but may not be restricted to area CM
(Schroeder et al. 2001). Using fMRI, a putative human
homologue of area CM has been identified located in a subregion of human auditory cortex along the superior temporal
gyrus (Foxe et al. 2002). This area substantially overlaps with
the region activated in the present study. Comparable to the
monkey data, the subregion of the human auditory cortex
responded to both auditory and somatosensory stimuli. The
activation strength was larger for multisensory than summed
unisensory stimulation suggesting multisensory integration in
that convergence zone. This result was confirmed by a recent
electrophysiological study that demonstrated early activation of
auditory cortical areas (around 100 ms after stimulus onset) in
response to vibrotactile stimulation (Caetano and Jousmäki
2006). Thus, multisensory input signals are integrated early in
the cortical processing hierarchy, in other words, in brain areas
traditionally considered as unisensory. A direct comparison of
brain areas contributing to auditory and vibrotactile processing
revealed a shared neural substrate in the posterior auditory belt
area (Schürmann et al. 2006). These results were interpreted
such that vibrotactile stimuli elicit a perception of a sound via
tactile sense due to the sound-like temporal patterns in
vibration. However, single tactile pulses that resulted in
a sensation of a brief touch without any vibration also activated
parts of the posterior auditory belt region (Schürmann et al.
2006). Therefore, it is more likely that the posterior auditory
belt region represents a multisensory convergence zone that
subserves processing of composite audiotactile events.
10 Movement Processing in the Blind
d
Fiehler et al.
While in sighted individuals the guidance of hand movements is primarily based on vision, blind people have to rely on
kinesthetic, tactile, and auditory information. The greater
experience in and stronger reliance of movement control on
composite somatosensory and auditory feedback might have
led to a stronger excitability of the multisensory convergence
zone in the auditory cortex in the blind. Thus, visual
deprivation results in both intermodal changes (i.e., visual
cortex activation) and compensatory changes in the intact
sensory cortices (i.e., auditory cortex activation) due to
intramodal plasticity (cf., Röder and Rösler 2004).
Neural Mechanisms of Kinesthetic Working Memory
Consistent with previous findings in tactile working memory
(Stoeckel et al. 2003; Kaas et al. 2007), the present study
identified the anterior intraparietal sulcus as being responsible
for working memory maintenance of kinesthetic information
used to guide hand actions. Activity in this region systematically
varied with memory load, i.e., the hemodynamic response
increased with increasing mnemonic demands. Importantly, we
demonstrated that the anterior intraparietal sulcus was not only
recruited by sighted but also by congenitally blind participants.
This finding further supports the assumption that dorsal
pathway functions related to the control of immediate and
delayed actions do not depend on early visual input.
Based on the anatomically defined maps by Choi et al.
(2006), activation overlapped significantly in both groups in the
left posterior portion of the anterior intraparietal sulcus, more
precisely, in human intraparietal area 1 (hIP1), but differed in
the left anteriorly located human intraparietal area 2 (hIP2).
Since, to our knowledge, no studies have examined the distinct
functional roles of hIP1 and hIP2 so far, an interpretation of the
observed group difference in hIP2 must be speculative and is
left for future research.
Kinesthetic working memory is moreover associated with
greater activity in prefrontal regions. Research in non-human
primates using the tactile flutter discrimination task, which
requires the comparison of two frequencies separated by
a short delay, suggests that the prefrontal cortex is part of
a network involved in tactile working memory (Romo and
Salinas 2003). In accordance with the monkey data, prefrontal
activity related to tactile working memory has been reported in
humans using fMRI (Stoeckel et al. 2003; Kaas et al. 2007) as
well. Across different modalities, the prefrontal cortex seems to
be involved in mnemonic control processes, whereas the
posterior association cortices directly operate on the storage
buffer (cf., Postle et al. 1999). Our data extend previous
findings on tactile working memory to the kinesthetic
modality.
Conclusion
In sum, congenitally blind and sighted humans recruited similar
networks of the dorsal action control system for kinesthetic
movement processing. This result demonstrates that dorsal
pathway functions develop even in the absence of vision.
Group differences in cortical activation patterns imply different
movement control strategies depending on visual experience
during ontogenesis.
Funding
German Research Foundation (Ro 529/18).
Notes
We thank Franz-Josef Visse and Judith Tacke for their help in recruiting
blind participants and Annerose Engel for thoughtful discussions about
the experimental design. Conflict of Interest: None declared.
Address correspondence to Katja Fiehler. Email: fiehler@staff.
uni-marburg.de.
References
Alain C, Arnott SR, Hevenor S, Graham S, Grady CL. 2001. What’’ and
‘‘where’’ in the human auditory system. Proc Natl Acad Sci USA.
98:12301--12306.
Amedi A, Floel A, Knecht S, Zohary E, Cohen LG. 2004. Transcranial
magnetic stimulation of the occipital pole interferes with verbal
processing in blind subjects. Nat Neurosci. 7:1266--1270.
Amedi A, Jacobson G, Hendler T, Malach R, Zohary E. 2002.
Convergence of visual and tactile shape processing in the human
lateral occipital complex. Cereb Cortex. 12:1202--1212.
Amedi A, Malach R, Hendler T, Peled S, Zohary E. 2001. Visuo-haptic
object-related activation in the ventral visual pathway. Nat Neurosci.
4:324--330.
Amedi A, Raz N, Pianka P, Malach R, Zohary E. 2003. Early ‘visual’ cortex
activation correlates with superior verbal memory performance in
the blind. Nat Neurosci. 6:758--766.
Amedi A, Stern WM, Camprodon JA, Bermpohl F, Merabet L, Rotman S,
Hemond C, Meijer P, Pascual-Leone A. 2007. Shape conveyed by
visual-to-auditory sensory substitution activates the lateral occipital
complex. Nat Neurosci. 10:687--689.
Binkofski F, Dohle C, Posse S, Stephan KM, Hefter H, Seitz RJ, et al. 1998.
Human anterior intraparietal area subserves prehension: A combined
lesion and functional MRI activation study. Neurology. 50:1253--1259.
Binkofski F, Kunesch E, Classen J, Seitz RJ, Freund H-J. 2001. Tactile
apraxia: Unimodal apractic disorder of tactile object exploration
associated with parietal lobe lesions. Brain. 124:132--144.
Bodegard A, Geyer S, Grefkes C, Zilles K, Roland PE. 2001. Hierarchical
processing of tactile shape in the human brain. Neuron. 31:317--328.
Brett M, Anton J-L,Valabregue R, Poline J-B. 2002. Region of interest
analysis using an SPM toolbox [abstract]. Presented at the 8th
International Conferance on Functional Mapping of the Human
Brain June 2-6 2002 Sendai Japan. Available on CD-ROM in
Neuroimage Vol. 16 No. 2.
Caetano G, Jousmäki V. 2006. Evidence of vibrotactile input to human
auditory cortex. Neuroimage. 29:15--28.
Cattaneo Z, Vecchi T, Monegato M, Pece A, Cornoldi C. 2007. Effects of
late visual impairment on mental representations activated by visual
and tactile stimuli. Brain Res. 1148:170--176.
Choi HJ, Zilles K, Mohlberg H, Schleicher A, Fink GR, Armstrong E,
Amunts K. 2006. Cytoarchitectonic identification and probabilistic
mapping of two distinct areas within the anterior ventral bank of
the human intraparietal sulcus. J Comp Neurol. 495:53--69.
Clark S, Bellmann Thiran A, Maeder P, Adriani M, Vernet O, Regli L, et al.
2002. What and where in human audition: selective deficits
following focal hemispheric lesions. Exp Brain Res. 147:8--15.
Cohen LG, Celnik P, Pascual-Leone A, Corwell B, Faiz L, Dambrosia J,
Honda M, Sadato N, Gerloff C, Catala MD, Hallett M. 1997. Functional
relevance of cross-modal plasticity in blind humans. Nature. 389:
180--183.
Connolly JD, Andersen RA, Goodale MA. 2003. FMRI evidence for
a ‘parietal reach region’ in the human brain. Exp Brain Res. 153:
140--145.
Culham JC, Danckert SL, DeSouza JF, Gati JS, Menon RS, Goodale MA.
2003. Visually guided grasping produces fMRI activation in dorsal
but not in ventral stream brain areas. Exp Brain Res. 153:180--189.
Curtis CE. 2006. Prefrontal and parietal contributions to spatial working
memory. Neuroscience. 139:173--180.
Desmurget M, Grafton S. 2000. Forward modeling allows feedback
control for fast reaching movements. Trends Cogn Sci. 4:423--431.
Dijkerman HC, DeHaan EHF. 2007. Somatosensory processes subserving
perception and action. Behav Brain Sci. 30:224--230.
Druzgal TJ, D’Esposito M. 2001. Activity in fusiform face area modulated
as a function of working memory load. Cog Brain Res. 10:355--364.
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.
Fiehler K, Burke M, Engel A, Bien S, Rösler F. 2008. Kinesthetic working
memory and action control within the dorsal stream. Cereb Cortex.
18:243--253.
Fiehler K, Engel A, Rösler F. 2007. Where are somatosensory
representations stored and reactivated? Behav Brain Sci. 30:206--207.
Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC.
1995. Improved assessment of significant activation in functional
magnetic resonance imaging (fMRI): use of a cluster-size threshold.
Magn Reson Med. 33:636--647.
Foxe JJ, Wylie GR, Martinez A, Schroeder CE, Javitt DC, Guilfolyle D,
Ritter W, Murray MM. 2002. Auditory-somatosensory multisensory
processing in auditory association cortex: an fMRI study. J Neurophysiol. 88:540--543.
Frey SH, Vinton D, Norlund R, Grafton ST. 2005. Cortical topography of
human anterior intraparietal cortex active during visually guided
grasping. Brain Res. 23:397--405.
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.
Fu K-M, Johnston TA, Shah AS, Arnold L, Smily J, Hackett TA,
Garraghty PE, Schroeder CE. 2003. Auditory cortical neurons
respond to somatosensory stimulation. J Neurosci. 23:7510--7515.
Garg A, Schwartz D, Stevens AA. 2007. Orienting auditory spatial
attention engages frontal eye fields and medial occipital cortex in
congenitally blind humans. Neuropsychol. 45:2307--2321.
Gerardin E, Sirigu A, Lehéricy S, Poline J-B, Gaymard B, Marsault C,
Agid Y, Le Bihan D. 2000. Partially overlapping neural networks for
real and imagined hand movements. Cereb Cortex. 10:1093--1104.
Goodale MA, Milner AD. 1992. Separate visual pathways for perception
and action. Trends Neurosci. 15:20--25.
Gougoux F, Zatorre RJ, Lassonde M, Voss P, Lepore F. 2005. A functional
neuroimaging study of sound localization: visual cortex activity
predicts performance in early-blind individuals. PLoS Biol. 3:
324--333.
Harrington DL, Rao SM, Haaland KY, Bobholz JA, Mayer AR, Binder JR,
Cox RW. 2000. Spezialized neural systems underlying representations of sequential movements. J Cognitive Neurosci. 12:56--77.
Imbiriba LA, Rodrigues EC, Magalhaes J, Vargas CD. 2006. Motor
imagery in blind subjects: the influence of the previous visual
experience. Neurosci Lett. 400:181--185.
James TW, Culham J, Humphrey GK, Milner AD, Goodale MA. 2003.
Ventral occipital lesions impair object recognition but not objectdirected grasping: an fMRI study. Brain. 126:2463--2475.
Jones EG, Powell TP. 1969. Connexions of the somatic sensory cortex
of the rhesus monkey. I. Ipsilateral cortical connexions. Brain. 92:
477--502.
Kaas A, van Mier H, Goebel R. 2007. The neural correlates of human
working memory for haptically explored object orientations. Cereb
Cortex. 17:1637--1649.
Karnath HO, Perenin MT. 2005. Cortical control of visually guided
reaching: evidence from patients with optic ataxia. Cereb Cortex.
15:1561--1569.
Malach R, Reppas JB, Benson RR, Kwong KK, Jiang H, Kennedy WA,
Ledden PJ, Brady TJ, Rosen BR, Tootell RBH. 1995. Object-related
activity revealed by functional magnetic resonance imaging in
human occipital cortex. Proc Natl Acad Sci USA. 92:8135--8139.
Malouin F, Richards CL, Jackson PL, Dumas F, Doyon J. 2003. Brain
activations during motor imagery of locomotor-related tasks: a PET
study. Hum Brain Mapp. 19:47--62.
Morosan P, Rademacher J, Schleicher A, Amunts K, Schormann T,
Zilles K. 2001. Human primary auditory cortex: cytoarchitectonic
subdivisions and mapping into a spatial reference system. Neuroimage. 13:684--701.
Oldfield RC. 1971. The assessment and analysis of handedness: The
Edinburgh inventory. Neuropsychologia. 9:97--113.
Picard N, Strick PL. 1996. Motor areas of the medial wall: a review of
their location and functional activation. Cereb Cortex. 6:342--353.
Cerebral Cortex January 2009, V 19 N 1 11
Pascual-Leone A, Amedi A, Fregni F, Merabet LB. 2005. The plastic
human brain cortex. Ann Rev Neurosci. 28:377--401.
Pascual-Leone A, Hamilton R. 2001. The metamodal organization of the
brain. Prog Brain Res. 134:427--445.
Pietrini P, Furey ML, Ricciardi E, Gobbini MI, Wu W-H, Cohen L,
Guazzelli M, Haxby JV. 2004. Beyond sensory images: Object-based
representation in the human ventral pathway. Proc Natl Acad Sci
USA. 101:5658--5663.
Postle BR, Berger JS, D’Esposito M. 1999. Functional neuroanatomical
double dissociation of mnemonic and executive control processes
contributing to working memory performance. Proc Natl Acad Sci
USA. 96:12959--12964.
Postle BR, Zarahn E, D’Esposito M. 2000. Using event-related fMRI to
assess delay-period activity during performance of spatial and
nonspatial working memory tasks. Brain Res Protoc. 5:57--66.
Rauschecker JP, Tian B. 2000. Mechanisms and streams for processing
of ‘‘what’’ and ‘‘where’’ in auditory cortex. Proc Natl Acad Sci USA.
97:11800--11806.
Reed CL, Klatzky RL, Halgren E. 2005. What vs. where in touch: an fMRI
study. Neuroimage. 25:718--726.
Röder B, Kusmierek A, Spence C, Schicke T. 2007. Developmental
vision determines the reference frame for the multisensory control
of action. Proc Natl Acad Sci USA. 104:4753--4758.
Röder B, Stock O, Bien S, Neville H, Rösler F. 2002. Speech processing
activates visual cortex in congenitally blind humans. Eur J Neurosci.
16:930--936.
Röder B, Rösler F. 2004. Compensatory plasticity as a consequence of
sensory loss. In: Calvert G, Spence C, Stein B, editors. The handbook
of multisensory processes. Cambridge: MIT Press. p. 719--747.
Röder B, Rösler F, Spence C. 2004. Early vision impairs tactile
perception in the blind. Curr Biol. 14:121--124.
Röder B, Rösler F, Hennighausen E. 1997. Different cortical activation
patterns in blind and sighted humans during encoding and transformation of haptic images. Psychophysiology. 34:292--307.
Romanski LM, Tian B, Fritz J, Mishkin M, Goldman-Rakic PS,
Rauschecker JP. 1999. Dual streams of auditory afferents target
multiple domains in the primate prefrontal cortex. Nat Neurosci. 2:
1131--1136.
Romo R, Salinas E. 2003. Flutter discrimination: Neural codes,
perception, memory and decision making. Nat Rev Neurosci. 4:
203--218.
Rushworth MF, Behrens TE, Johansen-Berg H. 2006. Connection
patterns distinguish 3 regions of human parietal cortex. Cereb
Cortex. 16:1418--1430.
Sadato N, Okada T, Kubota K, Yonekura Y. 2004. Tactile discrimination
activates the visual cortex of the recently blind naı̈ve to Braille:
12 Movement Processing in the Blind
d
Fiehler et al.
a functional magnetic resonance imaging study in humans. Neurosci
Letters. 359:49--52.
Sadato N, Pascual-Leone A, Grafman J, Ibanez V, Deiber MP, Dold G,
Hallett M. 1996. Activation of the primary visual cortex by Braille
reading in blind subjects. Nature. 380:526--528.
Sakata H, Takaoka Y, Kawarasaki A, Shibutani H. 1973. Somatosensory
properties of neurons in the superior parietal cortex (area 5) of the
rhesus monkey. Brain Res. 64:85--102.
Scheperjans F, Palomero-Gallagher N, Grefkes C, Schleicher A, Zilles K.
2005. Transmitter receptors reveal segregation of cortical areas in
the human superior parietal cortex: relations to visual and
somatosensory regions. Neuroimage. 28:362--379.
Schlaggar BL, O’Leary DD. 1991. Potential of visual cortex to develop an
array of functional units unique to somatosensory cortex. Science.
252:1556--1560.
Schroeder CE, Lindsley RW, Specht C, Marcovici A, Smiley JF, Javitt DC.
2001. Somatosensory input to auditory association cortex in the
macaque monkey. J Neurophysiol. 85:1322--1327.
Schürmann M, Caetano G, Hlushchuk Y, Jousmaki V, Hari R. 2006.
Touch activates human auditory cortex. Neuroimage. 30:
1325--1331.
Selemon LD, Goldman-Rakic PS. 1988. Common cortical and subcortical targets of the dorsolateral prefrontal and posterior parietal
cortices in the rhesus monkey: evidence for a distributed neural
network subserving spatially guided behavior. J Neurosci. 8:
4049--4068.
Stoeckel MC, Weder B, Binkofski F, Buccino G, Shah NJ, Seitz RJ. 2003. A
fronto-parietal circuit for tactile object discrimination: an eventrelated fMRI study. Neuroimage. 19:1103--1114.
Sur M, Leamey CA. 2001. Development and plasticity of cortical areas
and networks. Nat Rev Neurosci. 2:251--262.
Townsend JT, Ashby FG. 1993. The stochastic modeling of elementary
psychological processes. Cambridge: University Press.
Ungerleider L, Mishkin M. 1982. Two cortical visual systems. Ingle DJ,
Goodale MA, Mansfield RJW. Analysis of visual behavior. Cambridge:
MIT Press. p. 549--586.
Vanlierde A, De Volder AG, Wanet-Defalque MC, Veraart C. 2003.
Occipito-parietal cortex activation during visuo-spatial imagery in
early blind humans. Neuroimage. 19:698--709.
Weeks R, Horwitz B, Aziz-Sultan A, Tian B, Wessinger CM, Cohen LG,
Hallett M, Rauschecker JP. 2000. A positron emission tomographic
study of auditory localization in the congenitally blind. J Neurosci.
20:2664--2672.
Wise SP, Boussaoud D, Johnson PB, Caminiti R. 1997. Premotor and
parietal cortex: corticocortical connectivity and combinatorial
computations. Annu Rev Neurosci. 20:25--42.