16
Research Paper
The functional neuroanatomy of implicit-motion perception or
‘representational momentum’
C. Senior*, J. Barnes*†, V. Giampietro‡, A. Simmons§, E.T. Bullmore‡#,
M. Brammer‡ and A.S. David*
Background: When we view static scenes that imply motion — such as an
object dropping off a shelf – recognition memory for the position of the object is
extrapolated forward. It is as if the object in our mind’s eye comes alive and
continues on its course. This phenomenon is known as representational
momentum and results in a distortion of recognition memory in the implied
direction of motion. Representational momentum is modifiable; simply labelling
a drawing of a pointed object as ‘rocket’ will facilitate the effect, whereas the
label ‘steeple’ will impede it. We used functional magnetic resonance imaging
(fMRI) to explore the neural substrate for representational momentum.
Results: Subjects participated in two experiments. In the first, they were
presented with video excerpts of objects in motion (versus the same objects in
a resting position). This identified brain areas responsible for motion perception.
In the second experiment, they were presented with still photographs of the
same target items, only some of which implied motion (representational
momentum stimuli). When viewing still photographs of scenes implying motion,
activity was revealed in secondary visual cortical regions that overlap with areas
responsible for the perception of actual motion. Additional bilateral activity was
revealed within a posterior satellite of V5 for the representational momentum
stimuli. Activation was also engendered in the anterior cingulate cortex.
Addresses: *Department of Psychological Medicine,
Institute of Psychiatry and GKT School of Medicine,
103 Denmark Hill, London SE5 8AZ, UK. †Centre for
Brain and Cognitive Development, Birbeck College,
University of London WC1E 7JL, UK. ‡Brain Image
Analysis Unit and §Neuroimaging Research Unit,
Institute of Psychiatry, 103 Denmark Hill, London
SE5 8AZ, UK. #Department of Psychiatry,
Addenbrooke’s Hospital, Cambridge CB2 2QQ, UK.
Correspondence: C. Senior
E-mail: [email protected]
Received: 5 August 1999
Revised: 17 September 1999
Accepted: 15 November 1999
Published: 15 December 1999
Current Biology 2000, 10:16–22
0960-9822/00/$ – see front matter
© 2000 Elsevier Science Ltd. All rights reserved.
Conclusions: Considering the implicit nature of representational momentum and
its modifiability, the findings suggest that higher-order semantic information can
act on secondary visual cortex to alter perception without explicit awareness.
Background
The sensitivity of the human perceptual system for detecting movement in the outside world is so acute that it has
led investigators to explore the possibility that something
akin to motion perception may occur even when motion is
not present, only implied [1]. Evidence for this comes from
studies exploring the phenomenon known as ‘representational momentum’, best defined as a forward distortion in
recognition memory, in the direction of implied motion [2].
This memory distortion occurs because, when we see an
object, we form a mental representation of it which seems
to include information on the forces acting upon it (such as
gravity). This serves to maintain the mental image in equilibrium [3]. When this state is unbalanced (as by motion)
the position of the represented object appears to be mentally extrapolated, according to how it might behave in
nature, and this maintains the equilibrium, or what has
been termed ‘spatiotemporal coherence’ [4]. Representational momentum is demonstrated convincingly in experiments using the ‘freeze-frame’ paradigm: a still (inducer)
photograph of an object in motion is presented briefly, followed by a second (probe) picture, which is either forward
or behind the direction of movement. The viewers’
increased reaction-time delay when asked to indicate
whether the forward picture is in fact ‘different’ from the
inducer is a measure of the representational momentum
effect (see Figure 1) [1,5].
Studies have shown that representational momentum
can be modified by semantic or non-perceptual information; for example, applying the label ‘rocket’ or ‘fast’ to a
pointed polygon increases the extrapolated motion.
Likewise, the effect is impeded if the label conveys lack
of movement or inertia (for example, ‘steeple’ or
‘weight’) [6]. Thus semantic effects can modulate representational momentum. This is unlike the relatively
‘low-level’ psychophysiologically based illusions such as
apparent motion and motion after-effects [7], which are
impervious (or impenetrable) to knowledge and beliefs.
Representational momentum can therefore be taken as
evidence for the cognitive representation of motion [8].
Functional neuroimaging has implicated the ventral occipito-temporal junction, or V5 system, as the neural substrate
responsible for a range of motion processing [9]. ‘Pure’ visual
motion [10] as well as optic flow and schematic biological
Research Paper The cognitive representation of motion Senior et al.
700
600
500
400
Direction
Forward
Backwards
Stimulus label
Drop
Kettle
Hammer
Ball
Jump
Throw
Walk
Cup
Clock
300
Bus
We hypothesised that both a visual experience of implied
motion and perception of actual motion would share a
common neural substrate. This would support the contention that top-down influences can shape certain basic
perceptual processes, and would account for the ‘cognitive
penetrability’ of representational momentum (that is, it
does not occur reflexively, like apparent motion, but is
modulated by prior beliefs and expectations) ([18–20];
reviewed in [21]). The question would then arise as to
whether the effects are mediated via additional cerebral
regions outside the visual system, for example, the prefrontal cortex [15,22] or spatial attentional system, for
example the superior parietal lobule [23], or whether they
act to modulate activity within the visual cortex itself. We
made a further specific prediction: additional activity coincident with stimuli inducing representational momentum
would be seen in regions affiliated with object identification — the ventral ‘what’ stream — and not with the dorsal
‘where’ pathway [13], as the effect depends on semantic
encoding of the picture content. To test these hypotheses
functional magnetic resonance imaging (fMRI) was used to
identify the putative cerebral regions involved in the perception of actual and implied motion.
Figure 1
Mean reaction time (msec)
motion [11], the motion after-effect [12] and apparent
motion [13], all engender activity within V5. A continuum
between perception and cognition within the same neuroanatomical substrate can be posited as more complex motion
perception such as illusory motion [14] and mental transformations using visual imagery [15,16] also produce activity
within these structures. Recently, activity within a dorsal, or
‘where’, network (see for example [17]) outside V5 has been
postulated in the experience of motion imagery [13].
17
Current Biology
Bar chart showing mean reaction times for the ‘forward’ and
‘backward’ direction pairs for all stimuli used in the current study
(forward > backward = representational momentum). Data taken from
[33] (n = 19).
Figure 2).The range of the mean coordinates of activity
induced by perceived motion (left: X = –43 ± 6,
Y = –70 ± 6, Z = 12 ± 16 and right: X = 46 ± 17, Y = –63 ± 6,
Z = 14 ± 16) was well within the coordinate space previously described as the V5 system [9]. Additional areas of
supra-threshold activity (≥ 5 voxels) included the superior
temporal gyrus (STG). However, this was recorded during
the control, or ‘off’ phase of the experiment. This area has
been implicated in spatial attentional processing [24].
Results
Perception of actual motion
Perception of implied motion
Bilateral activation was revealed within V5 when subjects
performed an actual-motion perception task (Table 1;
Activity was observed bilaterally in the medial occipital and
temporal gyri in phase with the ‘on’ stimuli — stimuli
Table 1
Summary of the areas of activity revealed when subjects were presented with the actual motion task.
Gyrus
Left
Medial occipital
Middle temporal
Middle temporal
Middle temporal
Superior temporal
Medial occipital
Right
Middle temporal
Middle temporal
Middle temporal
Middle temporal
Middle temporal
Brodmann area
X
Y
Z
Cluster size (voxels)
FPQ
Phase
39
39
39
22
42
37
–43
–46
–40
–58
–55
–43
–72
–67
–67
–28
–25
–72
8
16
20
4
8
4
20
16
11
10
10
7
1.9
1.5
2.1
1.6
1.3
1.5
On
On
On
Off
Off
On
39
23
39
37
39
46
49
35
52
46
–61
–61
–67
–61
–64
16
8
20
4
20
25
23
15
6
6
1.9
2.2
1.9
1.8
1.6
On
On
On
On
On
All p values ≤ 0.0002. FPQ, fundamental power quotient.
18
Current Biology Vol 10 No 1
Figure 2
Median generic brain activation maps of
activated voxels registered in the space of a
high resolution spoiled GRASS (SPGR)
template. Four axial slices at Talairach z
coordinates +4, +8, +16 and +20 are shown.
These contain > 95% of the voxels activated
with a probability of a type I error < 0.002.
Activity revealed for the actual-motion
perception task is coloured red, for the
representational momentum task is coloured
yellow, whereas coincident voxels are blue.
Stereotactic (X, Y, Z) coordinates for the
centre of coincident clusters, with respective
voxel numbers in brackets, are 46, –69, 8
inducing representational momentum (Figure 2). Coincident activity within the previously defined reference area
(V5) was clearly evident (Table 2; Figure 3). Activity was
also revealed within the anterior cingulate cortex during the
‘off’ phase of the representational momentum experiment.
Of particular interest is the activity revealed within bilateral
posterior satellites of V5 (mean coordinates left: X = 32,
Y = –72, Z = 8; right: X = –23, Y = –83, Z = 8). These foci are
posterior and inferior (left > right) to the main cluster of
activated voxels seen in the actual-motion experiment.
Significant activation in the V5 complex was detected in
all six subjects in the actual-motion experiment and five
out of six in the implied-motion task (Table 3). The time
course of the activation in the implied-motion task is
shown in Figure 4. There was no detectable response in
the satellite region during the actual-motion task in any
of the subjects.
Discussion
We have shown that viewing still photographs of common
objects and scenes of people can produce robust activation
in cortical areas previously implicated in the perception of
objects and bodies in motion (V5), thus supporting the first
(17); –40, –75, 8 (15); 49, –64, 16 (11) and
–43, –78, 4 (10).
experimental hypothesis. Additionally, bilateral activity
was revealed within a ventro-posterior satellite of the V5
system, thought to be involved in denoting ‘what’ an
object is, which supports the second experimental hypothesis. Such activation was revealed when we fully controlled
for the content and form of the photographs except for the
variable of interest, namely implied motion. Hence, the
pictures were essentially identical, with the exception of,
for example, a slight angular displacement of an object or a
separation between a human figure and a surface. Furthermore, we were able to functionally define the relevant
motion-sensitive visual cortex independently using video
clips of the very same items undergoing real-time motion.
Neuroimaging studies have implicated eye movements in
brain activation within a network involving V5 and the
frontal eye fields (FEF) [25]. It is very unlikely that this
could account for visual cortical activation in the implied
motion experiment because, as argued above, there is no
reason to suspect that eye scanning would be different
between the pictures in the on and off phases. To rule out
the possibility that eye movements might have contributed to the differential activity in this experiment,
scan paths were recorded in four volunteers while they
Table 2
Summary of the areas of activity revealed when subjects were presented with the representational momentum task.
Gyrus
Left
Medial occipital
Middle temporal
Medial occipital
Middle temporal
Right
Middle temporal
Medial occipital
Middle temporal
Cingulate
Medial occipital
Middle temporal
All p values ≤ 0.0002.
Brodmann area
X
Y
Z
Cluster size (voxels)
FPQ
Phase
37
37
19
37
–43
–40
–23
–40
–64
–58
–83
–56
4
8
8
4
20
19
7
5
1.5
1.2
1.2
1.3
On
On
On
On
37
37
39
32
19
39
49
46
43
0
32
43
–56
–64
–56
39
–72
–56
8
4
16
20
8
20
20
17
13
6
5
5
1.3
1.5
1.2
1.6
1.5
1.4
On
On
On
Off
On
On
Research Paper The cognitive representation of motion Senior et al.
19
Figure 3
Orthogonal projection image illustrating
activity overlap in the implied- and actualmotion experiments mapped to the standard
space of Talaraich and Tournoux. Transverse,
sagittal and coronal slices are shown to
illustrate activity within the V5 system and
bilateral posterior satellites (yellow), revealed
during the representational momentum task
only. The mean coordinates of the cross-hairs
are (X = –50, Y = –70, Z = +4).
viewed 18 of the motion-implying and 18 of the nonmotion-implying stimuli under the same conditions as in
the fMRI experiment. Eye movements were recorded by
infrared refractometry and saccade length was calculated
from automatic sampling of the signal every 10 milliseconds. Scan paths, including saccades, were not statistically
different between the two picture types (mean saccade
length in arbitrary units for representational momentum
stimuli 91.6 (SD 49.6) versus mean non-representational
momentum stimuli 77.6 (SD 21.5); DF = 3, t = –0.39,
p = 0.7). Further, a fixation cross was included in the
250 millisecond inter-stimulus interval (ISI) to minimise
eye movement artefacts. Eye movements might, however,
have contributed to the V5 activity found during the
actual-motion perception experiment [26,27], the design
of which necessitated the inclusion of actual video
excerpts. This again seems unlikely, as activity in the
FEF was not detected. The absence of supra-threshold
activity within the FEF does not completely rule out the
possibility that sub-threshold activity might have contributed to the activity revealed within V5. The recording
of eye movements during scanning in future studies
would shed light on this issue.
As the ISI was constant at 250 milliseconds and isoluminant with the stimuli, we controlled for other perceptual
artefacts such as apparent motion and contrast detection,
respectively, which have also been shown to activate the
V5 system [28,29]. Debriefing of subjects confirmed that
none experienced apparent motion or were aware of the
experimental hypothesis.
In addition to the coincident activation within V5, a
further ‘satellite’ of activity was revealed by the representational momentum task only. The suggestion that
middle temporal regions might be involved in the representational momentum task only is consistent with the
cognitive nature of the effect and could represent the
retrieval of stored information on how the objects behave
in the real world. This left temporal involvement has
been shown in previous studies of visual semantic processing [30]. This ventral occipito-temporal network is
also believed to be part of a ‘form-recognition system’
[17] and is further implicated in a model of motion encoding [21]. Furthermore, frontal systems did not appear to
be involved. However, our experimental design cannot
exclude the possibility that, for example, superior parietal
or dorso-lateral prefrontal cortex is involved in this topdown mediation, if these regions showed tonic activity
during both the ‘on’ and ‘off’ phases of the experiment.
This is plausible as all of the stimuli used were of real
objects that would have engaged some semantic processing [22,31]. Also, the need to restrict the type of stimuli
employed, in order to preserve matching in both on and
off phases, meant that a more ‘pure’ motion activation
devoid of form and hue (for example, optic flow) could
not be used. Modest activation in the ‘off’ phase of the
representational momentum experiment was seen in the
Table 3
Summary of the sizes of activation clusters (voxels) within V5 complex across both experiments for each subject.
Actual motion
Representational momentum
Subject
Voxels
FPQ
p value
Voxels
FPQ
p value
1
2
3
4
5
6
8
6
11
20
5
17
3.5
3.5
3.3
3.4
3.0
5.4
0.000001
0.00007
0.00004
0.000005
0.00003
0.00003
<5
9
6
15
10
8
–
3.2
3.8
3.8
3.3
3.3
–
0.00001
0.00001
0.00006
0.00001
0.00002
Mean p values and FPQ, which is essentially a measure of the magnitude of the effect, are also shown.
20
Current Biology Vol 10 No 1
Figure 4
Figure 5
6
4
2
0
On
On
On
On
On
2
4
Off
0
20
40
60
80
Off
Off
Off
100
Off
Current Biology
Current Biology
Periodic time course of the averaged BOLD fMRI signal within the V5
system for five subjects. The black line is the raw time series and the
red line the model fitted to the data. The x-axis denotes magnetic
resonance signal in arbitary units; the y-axis denotes the sequence of
acquired images over time (1 image = 3 sec).
anterior cingulate cortex. This region has previously
been implicated in forms of selective attention [32]
including the direction of attention to the next stage in a
sequence (see for example [21]) which clearly might be
relevant to the experience of representational momentum. The biophysical mechanism for this pattern of
response is currently obscure, but one possible interpretation is that there is relatively greater activity in anterior
cingulate cortex with the anticipation, rather than perception, of motion.
Conclusions
We suggest that the V5 system is capable of perception of
both implied and actual motion; representational momentum is subserved by a cortical network that overlaps with
areas responsible for perception of actual motion. We
further suggest that the higher-order information that
interacts with representational momentum is processed
within the ‘object identification’ ventral pathway without
the need for ‘executive’ involvement. Semantic and conceptual factors can modulate dynamic mental representations (C.S., J.B., T. O’Dowd, C. Loveday, M. Baldwin and
A.S.D.; unpublished observations). This, and the functional neuroimaging findings, imply that higher-order
information can act on specialised motion-specific regions
of the visual cortex to alter perceptual experience in the
absence of awareness.
Schematic representation of the actual-motion fMRI experiment. Each
of the experimental epochs (on and off) are 30 sec long, making up the
total test time of 5 min. The off, or control, phase consisted of stills that
implied no motion. This was contrasted with a 4 sec video excerpt of
the same target item in motion.
indicated that they were free from neurological and perceptual disorder.
Informed consent was obtained and all procedures were approved by
the local research ethics committee.
Stimuli
Dynamic scenes depicting a single aspect of unambiguous, irreversible
motion were filmed (for example a man jumping from a ledge or a kettle
pouring water into a cup). All video sequences were filmed in a studio over
one session to control background luminance. They were recorded on a
PAL Betacam SP machine, Sony type UVW1800. Also, stills were taken
from the same items when they did not imply any motion — that is, a kettle
standing next to a mug and a man standing on a ledge. The stills were
frame grabbed into an Apple Mac 8100-80AV using the Apple Video
Player which is included in the Mac operating system. Stimuli were presented to subjects as back-projected video images. The projection screen
was placed across the bore of the scanning magnet, approximately 2 m
from the subjects’ eyes, subtending 10° horizontally and 8° vertically.
Procedure
The design of each experiment followed a 30 sec periodic ‘on–off’
design with presentation of the experimental material in the on phase
and the control material (items that implied no motion) in the off phase.
Stimuli were recorded onto two video cassettes, each containing a total
of five 30 sec experimental and five 30 sec control phases in alternating
presentation. Subjects were asked to watch the stimuli (or fixation
cross) at all times and anticipated being asked questions later about
their experience during scanning. The order of experiments, either actual
or implied motion, was counterbalanced between subjects.
Actual motion experiment
For this reference task, each experimental, or ‘on’, phase consisted of
six, 4 sec video excerpts of objects or people in motion with a 1 sec
interstimulus interval (ISI) during which a fixation cross was presented.
The control, or ‘off’, stimuli (static scenes with no implied motion) were
presented at the same rate (Figure 5).
Materials and methods
Subjects
Representational momentum experiment
Six volunteers (4 males and 2 females), mean age 34.3 years (range
22–55) were recruited. All were right handed, had normal vision and
For the experimental, or ‘on’, phase, 60 motion-implying stills were
presented at a rate of one every 250 msec separated by a 250 msec
Research Paper The cognitive representation of motion Senior et al.
isoluminant ISI (Figure 6). The same target items when not implying
any motion were presented at the same rate during the control, or off,
phase. The stimuli and ISI have previously been shown to reliably
produce representational momentum in behavioural experiments
(Figure 1) [5]; they were therefore adapted for fMRI. The mean representational momentum effect obtained on a separate off-line behavioural task using the ‘freeze-frame’ paradigm in six subjects was
37 msec (SE, 19 msec). The size of the reaction-time increase is well
within the range expected from previous studies but failed to reach
significance owing to the small number of subjects (see [5] for a procedural description).
Figure 6
fMRI image acquisition and analysis
Gradient-echo echoplanar magnetic resonance images were acquired
using a 1.5 T GE Signa System (General Electric, Milwaukee, WI, USA)
fitted with advanced NMR hardware and software (ANMR, Woburn, MA,
USA) at the Maudsley Hospital, London. Daily quality assurance was
carried out to ensure high signal-to-ghost ratio, high signal-to-noise ratio
and excellent temporal stability using an automated quality control procedure [33]. A quadrature birdcage head coil was used for radio frequency transmission and reception. In each of 14 non-contiguous
planes parallel to the inter-commissural (AC–PC) plane, 100 T2*weighted magnetic resonance images depicting BOLD contrast [34]
were acquired with excitation time (TE) = 40 msec, repeat time
(TR) = 3000 msec, flip = 90º, in-plane resolution = 3.1 mm, slice thickness = 7.7 mm, slice skip = 0.3 mm. Head movement was limited by
foam padding within the head coil and a restraining band across the
forehead. At the same session, a 43 slice, high-resolution inversion
recovery echoplanar image of the whole brain was acquired in the
AC–PC plane with TE = 73 msec, TI = 180 msec, TR = 16,000 msec,
in-plane resolution = 1.5 mm, slice thickness = 3 mm.
Effects of the subject’s head movement in fMRI data were corrected by
realignment and readjustment in each individual’s dataset, as described
previously [35]. The power of periodic signal change at the (fundamental) on–off frequency of stimulation was estimated by iterated leastsquares fitting a sinusoidal regression model to the motion corrected
time series at each voxel of all images. The model included a pair of
sine and cosine waves at the frequency of alternation between on and
off conditions, parameterised by coefficients γ and δ respectively. The
fundamental power quotient (FPQ, fundamental power divided by its
standard error) was estimated at each voxel and represented in a parametric map. The standardised power at the frequency of alternation
between experimental conditions, or FPQ, is derived from the sinusoidal regression parameters γ and δ and their standard errors, SE(γ)
and SE(δ), respectively (see [34] for details):
FPQ =
γ 2 +δ2
2( SE (γ )4 + SE (δ ) 4 )
Each observed fMRI time series was then randomly permuted 10 times,
and FPQ re-estimated after each permutation. This resulted in 10 maps
(for each subject at each plane) of FPQ estimated under the null
hypothesis that FPQ is not determined by experimental design [36].
All maps of FPQ were registered in the standard space of Talairach
and Tournoux [37]. After spatial normalisation, the observed and randomised FPQ maps from each subject were identically smoothed with
a two-dimensional Gaussian filter (full-width half maximum = 7 mm) to
accommodate variability in gyral anatomy and error of voxel displacement during normalisation. Generic activation was then robustly
decided by computing the median value of FPQ at each voxel of the
observed parametric maps, and comparing it to a null distribution of
median FPQ values computed from the permuted maps. If the
observed median FPQ exceeded the critical value of randomised
median FPQ, for a one-tailed test of size α = 0.0002, then that voxel
was considered generically activated. Generically activated voxels
were coloured (red, actual motion; yellow, implicit motion; and blue,
21
On
Off
On
Off
On
Off
On
Off
On
Off
Current Biology
Schematic representation of the implied-motion fMRI experiment. The
experimental epochs are the same size as the actual-motion experiment
(30 sec) and the control, or off phase again consisted of stills that
implied no motion. However, these were contrasted with stills inducing
representational momentum. Representational momentum and control
stills were presented each for 250 msec with a 250 msec ISI.
coincident activity) and superimposed on the grey-scale Talairach
template image to create generic brain activation maps [38]. Thus
activation was decided in these maps by a permutation test requiring
few distributional assumptions about a test statistic (median FPQ) that
is robust to the possible effects of outliers in small samples, but difficult, if not intractable, to test theoretically (see [38] for details).
Acknowledgements
This study was supported by The Royal Society. C.S. is supported by the
McDonnell-Pew programme in Cognitive Neuroscience, E.T.B. by the Wellcome Trust and J.B. by a UK MRC studentship. We thank Terry O’Dowd for
assistance with construction of the material and Dr Carlo Miniussi of the
Brain and Cognition Laboratory, Oxford for his expertise in eye movement
analysis. This study was presented in part at the Fifth International Conference on Functional Mapping of the Human Brain held in Dusseldorf,
Germany from June 22–26, 1999.
References
1. Freyd J: The mental representation of movement when static
stimuli are viewed. Percep Psychophys 1983, 33:575-581.
2. Freyd J, Finke R: Representational momentum. J Exp Psychol Leam
Mem Cogn 1984, 10:126-132.
3. Freyd J, Pantzer T, Cheng J: Representing statics as forces in
equilibrium. J Exp Psychol Gen 1988, 117:395-407.
4. Freyd J: Dynamic mental representations. Psychol Rev 1987,
94:427-438.
5. Senior C, Barnes J, Loveday C, Baldwin M, David A: Factors that affect
the implicit processing of motion. Perception 1998, S27:165c.
6. Reed C, Vinson N: Conceptual effects on representational
momentum. J Exp Psychol Hum Percept Perform 1996,
22:839-850.
7. Moore C, Engel S: Visual perception: Mind and brain see eye to
eye. Curr Biol 1999, 9:R74-R76.
8. Hubbard T: Cognitive representation of motion: Evidence for
friction and gravity analogues. J Exp Psychol Leam Mem Cogn
1995, 21:241-254.
9. Watson J, Myers R, Frackowiak R, Hajnal J, Woods R, Mazziota J, et al.:
Area V5 of the human brain: evidence from a combined study using
positron emission tomography and magnetic resonance imaging.
Cereb Cortex 1993, 3:79-94.
10. Smith A, Greenlee M, Singh K, Kraemer F, Henning J: The
processing of first- and second-order motion in human visual
cortex assessed by functional magnetic resonance imaging
(fMRI). J Neurosci 1998, 18:3816-3830.
22
Current Biology Vol 10 No 1
11. Howard R, Brammer M, Wright I, Woodruff P, Bullmore E, Zeki S: A
direct demonstration of functional specialisation within motionrelated visual and auditory cortex of the human brain. Curr Biol
1996, 6:1015-1019.
12. Tootell R, Reppas J, Dale A, Look R, Serano M, Malach R, et al.:
Visual motion after-effect in human cortical area MT revealed
by functional magnetic resonance imaging. Nature 1995,
375:139-141.
13. Goebel R, Khorram-Sefat D, Muckli L, Hacker H, Singer W: The
constructive nature of vision: direct evidence from functional
magnetic resonance imaging studies of apparent motion and
motion imagery. Eur J Neurosci 1998, 10:1563-1573.
14. Zeki S, Watson J, Frackowiack R: Going beyond the information
given: the relation of illusory visual motion to brain activity. Proc
Roy Soc Lond 1993, 252:215-222.
15. Cohen M, Kosslyn S, Breiter H, DiGoralamo G, Thompson W,
Anderson K, et al.: Changes in cortical activity during mental
rotation: a mapping study using fMRI. Brain 1995, 119:89-100.
16. Barnes J, Howard R, Senior C, Brammer M, Bullmore E, Simmons A,
et al.: Cortical activity during rotational and linear transformation.
Neuropsychologia 1999, in press.
17. Martin A, Haxby J, Lalonde F, Wiggs C, Ungerleider L: Discrete
cortical regions associated with knowledge of color and
knowledge of action. Science 1995, 270:102-105.
18. Hubbard T: Judged displacement: a modular process? Am
J Psychol 1994, 107:359-373.
19. Ranney M: Internally represented forces may be cognitively
penetrable: comment on Freyd, Pantzer, and Cheng. J. Exp Psychol
Gen 1989, 118:399-402.
20. Finke R, Freyd J: Mental extrapolation and cognitive penetrability:
reply to Ranney and proposals for evaluative criteria. J Exp
Psychol Gen 1989, 118:403-408.
21. Kosslyn S: Image and Brain. Cambridge MA: MIT Press; 1996.
22. Kosslyn S, Alpert N, Thompson W, Chabris C, Rauch S, Anderson A:
Identifying objects seen from different viewpoints: a PET
investigation. Brain 1994, 117:1055-1071.
23. Corbetta M, Shulman G, Miezin F, Petersen S: Superior parietal
cortex activation during spatial attention shifts and visual feature
conjunction. J Neurosci 1993, 13:1202-1226.
24. Frances F, Egly R, Rafal D, Beck D: Spatial attention deficits in
humans: A comparison of superior parietal and temporal parietal
junction lesions. Neuropsychology 1998, 12:193-207.
25. Corbetta M, Akbudak E, Conturo T, Snyder A, Ollinger J, Drury H,
et al.: a common network of functional areas for attention and eye
movements. Neuron 1998, 21:761-773.
26. Bodis-Wollner I, Bucher, SF, Seelos KC, Paulus W, Reiser M, Oertel W:
Functional MRI mapping of occipital and frontal cortical activity
during voluntary and imagined saccades. Neurology 1997,
49:416-420.
27. Law I, Svarer C, Rostrup E, Paulson OB: Parieto-occipital cortex
activation during self-generated eye movements in the dark. Brain
1998, 121:2189-2200.
28. Tootell R, Reppas J, Kwong K, Malach R, Born R, Brady T, et al.:
Functional analysis of human MT and related visual cortical
areas using magnetic resonance imaging. J Neurosci 1995,
15:3215-3230.
29 Kaneoke Y, Bundou M, Koyama S, Suzuki H, Kakigi R: Human cortical
areas responding to stimuli in apparent motion. Neuroreport 1997,
8:677-682.
30. Vandenberghe R, Price C, Wise R, Josephs O, Frackowiak RSJ:
Functional anatomy of a common semantic system for words and
pictures. Nature 1996, 383:254-256.
31. Demb J, Desmond J, Wagner A, Vaidya C, Glover G, Gabrieli J:
Semantic encoding and retrieval in the left inferior prefrontal
cortex: a functional MRI study of task difficulty and process
specificity. J Neurosci 1995, 15:5870-5878.
32. Janer K, Pardo J: Deficits in selective attention following bilateral
anterior cingulotomy. J Cogn Neurosci 1991, 3:231-241.
33. Simmons A, Moore E, Williams S: Automated quality control for
functional MRI studies. Neuroimage 1997, 5:s466.
34. Ogawa S, Lee T, Kay A, Tank D: Intrinsic signal changes
accompanying sensory stimulation: functional brain mapping
resonance imaging. Proc Natl Acad Sci USA 1990, 3:9868-9872.
35. Bullmore E, Brammer M, Rabe-Hesketh S, Curtis V, Morris R, Williams
S, et al.: Methods for diagnosis and treatment of stimuluscorrelated motion in generic brain activation studies using fMRI.
Hum Brain Map 1999, 7:38-48.
36. Bullmore E, Brammer M, Williams SCR, Rabe-Hesketh S, Janot N,
David AS, et al.: Statistical methods of estimation and inference for
functional MR image analysis. Magn Res Med 1996, 35:261-277.
37. Talairach J, Tournoux P: Co-Planar Stereotaxic Atlas of the Human
Brain. New York: Thieme Medical Publishers Ltd; 1988.
38. Brammer M, Bullmore E, Simmons A, Williams S, Grasby P, Howard R,
et al.: Generic brain activation mapping in functional magnetic
resonance imaging: a non-parametric approach. Magn Res Imag
1997, 15:763-770.
Because Current Biology operates a ‘Continuous Publication
System’ for Research Papers, this paper has been published
on the internet before being printed. The paper can be
accessed from http://biomednet.com/cbiology/cub — for
further information, see the explanation on the contents page.