Update article
Rise and fall of the two visual systems theory
Yves Rossetti a,b,
*,1
, Laure Pisella a
, Robert D. McIntosh c
a
Inserm UMR-S 1028, CNRS UMR 5292, ImpAct, Centre de Recherche en Neurosciences de Lyon, universite´ Lyon-1, 16, avenue Le´pine, 69676 Bron, France
b
Plate-forme « Mouvement et Handicap », Hoˆpital Henry-Gabrielle, Hospices Civils de Lyon, 20, route de Vourles, Saint-Genis-Laval, France
c
Human Cognitive Neuroscience, Psychology, University of Edinburgh, Edinburgh, UK
Toutes les the´ories qui servent de point de de´part au physicien, au
chimiste, et a` plus forte raison au physiologiste, ne sont vraies que
jusqu’a` ce qu’on de´couvre qu’il y a des faits qu’elles ne renferment
pas ou qui les contredisent. Lorsque ces faits contradictoires se
montreront bien solidement e´tablis, loin de se roidir, comme le
scolastique ou le syste´matique, contre l’expe´rience, pour sauvegarder
son point de de´part, l’expe´rimentateur s’empressera, au
contraire, de modifier sa the´orie, parce qu’il sait que c’est la seule
manie`re d’avancer et de faire des progre`s dans les sciences.
(Claude Bernard, Introduction a` l’e´tude de la me´decine expe´rimentale,
1, II, VI)
1. Context of the theory
Several dissociations have been described within mammalian
vision. One can list for example: conscious vs. unconscious vision;
focal vs. ambient vision; spatial vs. object vision; egocentric vs.
allocentric vision (see [1], for a review). Historically, the focus of
these dissociations moved from anatomically-defined distinctions,
such as cortical vs. sub-cortical vision [2], towards functional
dissociations, such as semantic vs. pragmatic vision [3,4]. It is
interesting to consider the development of arguments, stemming
from diverse lines of evidence, compiled by several authors during
the eighties and nineties in order to propose reconciliations
between anatomical, electrophysiological, psychophysical and
neurophysical elements (e.g. [3–13]).
1.1. Anatomical context
Visual experience is unitary, but visual anatomical networks are
much more complex than a serial hierarchy leading to grandAnnals
of Physical and Rehabilitation Medicine 60 (2017) 130–140
A R T I C L E I N F O
Article history:
Received 22 December 2016
Accepted 15 February 2017
Keywords:
Vision
Action
Dorsal
Optic ataxia
Visuomotor
A B S T R A C T
Among the many dissociations describing the visual system, the dual theory of two visual systems,
respectively dedicated to perception and action, has yielded a lot of support. There are psychophysical,
anatomical and neuropsychological arguments in favor of this theory. Several behavioral studies that
used sensory and motor psychophysical parameters observed differences between perceptive and motor
responses. The anatomical network of the visual system in the non-human primate was very readily
organized according to two major pathways, dorsal and ventral. Neuropsychological studies, exploring
optic ataxia and visual agnosia as characteristic deficits of these two pathways, led to the proposal of a
functional double dissociation between visuomotor and visual perceptual functions. After a major wave
of popularity that promoted great advances, particularly in knowledge of visuomotor functions, the
guiding theory is now being reconsidered. Firstly, the idea of a double dissociation between optic ataxia
and visual form agnosia, as cleanly separating visuomotor from visual perceptual functions, is no longer
tenable; optic ataxia does not support a dissociation between perception and action and might be more
accurately viewed as a negative image of action blindsight. Secondly, dissociations between perceptive
and motor responses highlighted in the framework of this theory concern a very elementary level of
action, even automatically guided action routines. Thirdly, the very rich interconnected network of the
visual brain yields few arguments in favor of a strict perception/action dissociation. Overall, the
dissociation between motor function and perceptive function explored by these behavioral and
neuropsychological studies can help define an automatic level of action organization deficient in optic
ataxia and preserved in action blindsight, and underlines the renewed need to consider the perceptionaction
circle as a functional ensemble.

C 2017 Elsevier Masson SAS. All rights reserved.
* Corresponding author. Inserm UMR-S 1028, CNRS UMR 5292, ImpAct, Centre de
Recherche en Neurosciences de Lyon, universite´ Lyon-1, 16, avenue Le´pine,
69676 Bron, France.
E-mail address: yves.rossetti@inserm.fr (Y. Rossetti).
1
www.U864.lyon.inserm.fr.
Available online at
ScienceDirect
www.sciencedirect.com
http://dx.doi.org/10.1016/j.rehab.2017.02.002
1877-0657/
C 2017 Elsevier Masson SAS. All rights reserved.
mother cells of perceptual consciousness. The most apparent
division of the retrochiasmatic visual system is that between
cortical and sub-cortical visual pathways. Schneider [2] suggested
that the geniculo-striate pathway of the rodent was essential for
visual discrimination, and that the retinotectal pathway corresponded
to a system responsible for spatial orientation. He thus
introduced the idea of distinct systems specialized to answer the
questions ‘‘what is it?’’ and ‘‘where is it?’’. This distinction was
picked-up and developed by several authors.
In humans, lesions of the visual cortex were classically
considered to cause complete blindness, though pupillary
responses and rudimentary sensitivity to sudden contrast
changes might be retained (review: [3]). This opinion was
challenged by the observation that monkeys without a striate
cortex showed a paradoxical ability to avoid obstacles or grab
objects located in their blind visual field [14]. These monkeys lost
their residual visual abilities after damage to the areas of the
superior colliculi corresponding to their cortical scotoma [15],
implicating the retinotectal route as the basis of those abilities.
The existence of parallel pathways from the retina through
cortical and sub-cortical routes allowed the emergence of one of
the most fascinating phenomena in human neuropsychology:
‘‘blindsight’ [16–18]. Specifically, the observation of ‘‘blindsight’’
in patients with cortical hemianopia reinforced the arguments in
favor of the intervention of sub-cortical structures in blindsight
[19,20]. The blindsight phenomenon (reviews: [21,22]), especially
its relation to visually-guided action [23], will be developed in the
last part of this article.
Functional neuroanatomy also unveiled the existence of several
visual pathways through which the retina is connected to the
cortex (see Fig. 1). This anatomical approach was largely
completed and detailed by independent electrophysiological
studies of visual areas (review: [24]) and cortical substrates of
the action (review: [25,26]). First of all, several exploratory
experiments on the visual brain showed the extreme parcellation
of the visual cortex into multiple functional areas ([27] and their
famous figure 4; [28]). Famous experiments performed on
monkeys allowed the identification of two principal cortical
pathways of vision [29]. One of the pathways, the occipitotemporal
or ventral stream, links the prestriate areas to the inferior
temporal cortex. The interruption of this pathway inhibits the
visual discrimination of objects without affecting the perception of
spatial relationships between them (relative positions). The other
pathway, the occipito-parietal or dorsal stream, leads to the
posterior parietal cortex (PPC). The interruption of this pathway
causes spatial disorientation, characterized by a deficit in the
perception of relative positions [29] and a deficit of localization
observed during goal-directed actions [30]. These patterns of
deficit, observed through surgical disconnection studies in
monkeys, suggest that these two divergent ventral and dorsal
streams correspond to two different functions: processing the
‘‘what’’ and the ‘‘where’’ (see Fig. 1).
The exploration of cortical substrates of action, performed
with lesions or electrophysiological recording of the parietal
cortex of the monkey, confirmed the importance of the PPC in
guiding goal-directed movements (e.g.: [31,32]). These results
strongly supported the specialization of the dorsal visual system
for goal-directed actions (reviews: [4,13,25,33–36]). These
findings were complemented by electrophysiological studies
on the latency of visual areas, which revealed a rapid processing
pathway (the dorsal stream), and a slower one (the ventral
stream) [37–39], compatible with the idea of a dorsal stream
enabling fast action responses [40]. Furthermore, human clinical
data corroborated the importance of parietal areas in actionguidance
[3,41,42], followed by functional imaging data
published after Faillenot et al. [43].
1.2. Psychophysical context
A second methodological approach was based on experimental
protocols combining perceptual and motor psychophysics
[44]. Variations on the visual ‘‘double-step’’ paradigm proved
especially fertile. This paradigm refers to experimental conditions
in which a visual target is presented to a subject (1st step: between
the initial fixation point and the position of the target), and then
moved during the subject’s response (second step: between the
target’s first position and the second one). An interesting variation
of the task is one in which the second step is made during a saccade
directed towards the first target position. Since small target
displacements are not perceived if they occur immediately before
or during a saccade (saccadic suppression of image displacement:
[45]), a subject’s awareness of the displacement in double-step
tasks can be eliminated by synchronizing the second step with the
increase in eye velocity at saccade onset (e.g. [46]). In the original
experiment conducted to evaluate the consequences of this
deficient perception on arm movements, Bridgeman et al. [47]
asked subjects to make an eye movement and also to move their
Fig. 1. Representation of the evolution of the visual theories for action. The upper
row represents the intuitive concept of vision according to which our actions are
preceded by conscious awareness of perception. This serial conception is illustrated
by Descartes’ drawing. The second row illustrates the duplex notion according to
which the visual system is divided into two parts, dorsal and ventral, which are
respectively responsible of action and perception, and projected onto the human
brain. The third row illustrates the fact that interactions can be described between
the two anatomical pathways and perceptive and motor visual functions (data
issued from study on the primate’s brain) (from [39]). The fourth row illustrates the
observation that anatomical projections reaching the primate primary motor
cortex, are all subjected to prior interconnections (blue) between dorsal (green) and
ventral (red) pathways, suggesting that motor actions could result from several
interacting visual processing pathways, converging before the motor exit (from
[38,39,49]).
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140 131
hand to the target. Surprisingly, saccadic suppression was not
associated with oculomotor or manual performance costs;
pointing movements were adjusted normally to take account of
the target displacement, even though the displacement went
unnoticed.
Subsequently, a now-famous series of experiments was
initiated by Prablanc and colleagues, to explore the motor control
of rapid movements. In a first work, Pe´lisson et al. [46] and Goodale
et al. [48] asked healthy subjects to direct their eyes and hand
towards targets presented in the dark (synchronised with the
disappearance of the fixation point). These targets could be jumped
to a nearer or further location during the subject’s saccade, so that
the arm movement, already initiated at the time of the jump, had to
be updated to reach the target’s second position. Similarly to the
findings of Bridgeman et al. [47], subjects were able adjust the arm
movement even whilst unaware of the target displacement.
Another striking observation was that subjects did not detect
their own trajectory corrections, suggesting that these did not give
rise to any unusual proprioceptive sensations, [46,48]. These
experiments suggest that two visual systems could be differentiated
at a psychophysical level; a system involved in the
sensorimotor response (i.e. manual vision [49]), and a system
participating in the cognitive processes related to conscious
perception.
Bridgeman also made use of the original Roelofs effects and the
induced Roelofs effects [50]: when a frame of reference is moved
during the fixation of a small visual target, a displacement of the
target is perceived in the opposite direction. The phenomenon can
be observed for slow displacements (induced movements) or
discontinuous ones (induced displacements). Bridgeman et al. [5]
compared the amplitude of the perceptual illusion with the
pointing responses executed by subjects after the target was
extinguished: the motor response was much less affected than the
perceptual response by the apparent displacement. Their conclusion
was that the apparent displacement only affected perception,
whereas the real position only affected the motor behavior; the
demonstration was initially proposed as an argument in favor of a
dissociation between sensorimotor and cognitive visual functions
(see [6,8]). Further studies made it clear that this apparent
dissociation could not be taken as an argument for a dissociation
between perception and action, as the absence of effect in the
visuomotor condition can be explained by the addition of two
effects: a perceptual [51] and a remapping component [52]. Further
dissociations have subsequently been reported, whereby different
visual illusions affect primarily the perceptual or the visuomotor
responses as described further below.
1.3. Neuropsychological context
A third, human neuropsychological approach to visual modularity
combines anatomical and functional aspects and is rooted in
clinical traditions [3,13,53]. In addition to the knowledge acquired
in non-human primates, the observation of clinical cases validated
the modularity of the human visual cortex, adding to the
distinction between cortical and sub-cortical vision as explored
by the works on blindsight. The observation of multiple clinical
pictures of visual agnosia implied the existence of specialized
modules for the analysis of shapes, color, recognition of objects,
and even specific categories of objects (review: [54]). Other
patients presented specific impairments of visual motion detection
(akinetopsia) or impairments in directing their gestures towards a
visual object (optic ataxia). The former impairments resulted from
damage to the occipito-temporal junction through to the temporal
pole, and the latter from damages to dorsal occipito-parietal
regions (see Fig. 2).
1.4. Optic ataxia
The effects of neurological lesions involving a limited part of the
posterior parietal cortex were described in groups of patients by
Jeannerod [41] and Perenin and Vighetto [55]. These patients
presented difficulties in directing their actions toward objects
Fig. 2. Functional specialization of the human visual system. Neuropsychology offers the description of specific visual impairments, supporting the existence of specific visual
information processing centers, damage to which could lead to visual impairments in seeing shapes (agnosia), colors (color agnosia or central achromatopsia), more or less
specific objects (e.g. animal or manufactured objects agnosia) or faces (prosopagnosia), localized in the ventral pathway, movement vision (akinetopsia) or directed
movements (optic ataxia), associated to the dorsal pathway. This figure illustrates the information processing of a visual scene (on the left side) in the visual areas processing
information received by the primary visual cortex (V1).
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140132
presented in their peripheral vision, even though they had no
deficits in object recognition. This neurological deficit, optic ataxia,
is observed after a lesion of the posterior parietal cortex (see [56–
58]) (Pisella et al., 2017: http://dx.doi.org/10.1016/j.rehab.2016.
01.003). Movements directed towards peripheral visual targets are
imprecise, and most often deviated towards the side of the
lesioned hemisphere [58–62]. These disruptions become especially
obvious when the vision of the hand is prevented before and during
the movement. In the presence of visual feedback, the precision is
improved but movements remain slower than normal [41]. Distal
components, such as hand shaping to grasp objects, are also
disrupted by the posterior parietal damage. During grasping
gestures, patients opened their hand very widely, presumably in
order not to miss the object, and made no appropriate adjustments
to their grip aperture until there was tactile contact with the object
[41,59,63,64]. Patients who were asked to insert the hand into a slit
with a variable orientation did not direct their hand accurately
towards the location of this visual goal, but also did not rotate their
wrist to the correct orientation [55,59,65]. However, these patients
seemed to retain the ability to indicate the orientation of a
stimulus with a wrist movement not directed towards the stimulus
itself, as seen in matching tasks (see [59,66]). These results suggest
that the posterior parietal cortex plays a crucial role in spatial
vision (position coding) or object vision (size or shape coding) in
the context of goal-directed action (see [3,33,40,49]). The duplex
theory interprets optic ataxia as one side of a major dissociation
between vision-for-perception and vision-for-action, in which
‘‘action’’ is defective but ‘‘perception’’ is preserved [13,67].
1.5. Visual agnosia
The complementary dissociation profile to optic ataxia yielded
important arguments in favor of the notion of a perception-action
dissociation. Following a bilateral lesion of the occipito-temporal
cortex due to carbon monoxide poisoning, a patient (DF) presented
with visual agnosia of shapes [68]: she was unable to recognize the
size, shape, or orientation of objects, whether her method of report
was verbal or manual (e.g. indicating size by spacing out the
fingers, or orientation by rotating the wrist) [69]. However, when
she was asked to reach for the object with a grasping action, or to
‘post’ an object through an oriented slit, her accuracy and precision
were remarkably increased, even apparently normal. These
observations suggested that during goal-directed movements,
DF could access visual properties of the object that she could not
consciously recognize or report on. This remarkable pattern, then,
suggested another dissociation, in which perception was deficient
and action control was retained. The contrast of these results with
those observed in optic ataxia suggested that deficits in visual
recognition of objects and goal-directed action were fully, doubly,
dissociable.
1.6. Formulation of the duplex theory
The scientific works on these different approaches published
between the 1970s and the 1990s allowed the development of rich
knowledge of the anatomical and functional organization of the
human visual system. The experimental effervescence was initially
organized around the notion of a dissociation of anatomic origin
described by Schneider [2] in rodents (tectum/cortex) then by
Ungerleider and Mishkin [29] in primates (dorsal/ventral). Despite
their very different anatomical bases, the terms proposed to
describe these two identified visual sub-function were identical:
‘‘what?’’ vs. ‘‘where?’’ Thus, the visual function is defined by the
visual attribute processed. But this criterion of the visual attribute,
became progressively insufficient to describe the complexity of
clinical observations. Instead, the critical importance of the
behavioural function subserved by the visual processing was
underlined, from the beginning of the nineties [3,6,11], giving rise
to several independent formulations, including the duplex theory
of Milner and Goodale [13].
As reported above, neuropsychological observations conducted
on optic ataxia and visual agnosia suggested that the dorsalventral
distinction could in reality correspond to the dissociation
between the knowledge of the ‘‘what’’ and the organization of the
‘‘how’’ to direct the action [3,4,11,70]. This new interpretation of
the parallelism between the two visual systems, which dominated
a large part of the scientific literature these past 20 years, takes a
lesser consideration of the passive aspect of vision (‘‘what’’ and
‘‘where’’) in favor of the use of the stimulus, dependent on the
response produced by the subject (‘‘what’’ and ‘‘how’’).
In light of the scientific context that attributed a key role to the
posterior parietal cortex in guiding movements, the initial study by
Goodale et al. [69] that described preserved motor abilities in a
patient (DF) presenting with agnosia of visual shapes, supported a
theory based on the opposition between visuomotor grasping and
visual recognition of objects. As we will see, it is essentially the
visuomotor abilities of visual agnosia, used to build this theory,
which were experimentally explored, although the first detailed
study of DF was focused on perceptual functions [71]. For a long
time the other side of the theory, object recognition, or
‘‘perception’’, was not evaluated in controlled conditions identical
to sensorimotor testing ones. The keystone of the duplex theory is
unquestionably the existence of a double dissociation, key concept
of neuropsychology (see [1]), between optic ataxia and visual
agnosia (see [72]).
1.7. The duplex theory
Optic ataxia and visual agnosia have been very often considered
as representing dissociations between perception and action, as
the two components of a double dissociation between ‘‘perception’’
and ‘‘action’’ (see [13,40]). The duplex theory brought
forward by Milner and Goodale was not exclusively based on the
contrast between these two neuropsychological pictures, but it
drew much strength from this contrast. This argument did have the
advantage of identifying a link between some behavioral
dissociations and certain anatomical concepts of the visual system.
Specifically, the formulation of this theory brought some sense to
the increasingly complex anatomical networks described in
primates. It triggered a considerable number of behavioral studies
that explored the visuomotor functions, leading to numerous new
empirical findings, even if some purely behavioural findings were
inappropriately interpreted on an anatomical level in light of the
duplex notion of an anatomical/functional equivalence (Table 1).
Several dissociations were described in DF or in patients with
optic ataxia, shedding light on the functions of the human dorsal
and ventral streams. The majority part of these studies explored
grasping gestures, pointing movements and visual perception in
these patients (review: [4,13,40,49]). Patients with optic ataxia
showed a paradoxically improved performance with the introduction
of a delay between stimulus presentation and the motor
response [64,73,74], whereas DF showed a deterioration of her
visuomotor abilities in a few seconds [75]. Perceptual and motor
versions of an obstacle-avoidance task completed these descriptions.
A great number of studies on the motor and perceptual
effects of visual illusions, even though they brought no anatomical
results, were also interpreted within this theoretical framework.
Goodale and Milner proposed over time several rather stable
versions of their theory [67,76–78], and popularized the idea that
the visual system is divided in two independent parts dedicated to
perception and action respectively, illustrated and validated by a
dual neuropsychological dissociation.
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140 133
2. Deconstruction of the theory
This theoretical description of a double dissociation between
optic ataxia and visual agnosia contrasts sharply with the clinical
picture, if one asks which patient is more disabled in the actions of
daily life. To the extent that patients with agnosia cannot identify
objects, they cannot select appropriate goals for their actions, and
are functionally incapacitated. By contrast, patients with optic
ataxia execute actions fluently in central vision (unless the ataxia is
bilateral), so suffer very little loss of autonomy in daily life
[56,58,79]. As a striking example, one of the patients we followed
became a taxi driver after his vascular stroke. So how did we come
to formulate and disseminate so widely a theory suggesting the
contrary of this clinical reality? Let’s review some of the key
anatomical and functional data, to deconstruct this theory [60].
First of all, a critical data figure reproduced by Milner and
Goodale [13], to illustrate the proposed dissociation between
perception and action, comes from works by Vighetto and Perenin
[55,59]. The figure selected by Milner and Goodale illustrates the
number of motor errors made by 10 patients with right or left optic
ataxia in simple reaching (Fig. 3). We can observe on this figure
that patients make most errors (corrected and uncorrected) in the
contralesional visual hemifield. However, the version of the figure
presented by Milner and Goodale selectively omitted the central
panel, which showed that patients made hardly any errors in the
central part of their visual field, except for one patient (No. 3) who
presented with multiple associated impairments ([55], Figure 4).
The text from the original article also described the absence of
motor errors when patients visualized the target in foveal vision
([55], Table 9.4), and the existence of errors for perceptual
responses (see [60], Figure 10.2). This aspect was not mentioned by
Milner and Goodale [13], but has been validated by several
subsequent publications (e.g. [74,80–83]), and is illustrated in
Fig. 4.
Table 1
Dissociations related to the perception-action dissociation of the duplex theory.
Perception
Sub-cortical vision Cortical vision Cajal [133]
Tectal vision (Where?) Cortical vision (What?) Schneider [2]
Ambient vision Focal vision Threvarthen (1968), Humphrey [134]
Localization Identification Paillard (1983)
Locus-specific information Global-feature or figural information Held [135]
Spatial Object Ungerleider and Mishkin [29]
Response
Action Experiential Goodale [136]
Motor Cognitive Bridgeman [135], Bridgeman [6]
Sensorimotor Cognitive or representational Paillard [9], Paillard [10]
Implicit Explicit Weiskrantz et al. [18]
Action (How?) Perception (What?) Goodale and Milner [11], Milner and Goodale [13]
Pragmatic Semantic Jeannerod and Rossetti [3], Jeannerod [12]
Direct parameter specification Conscious representation Neumann and Klotz [137]
How? Where? Rossetti et al. [138,139]
Automatic Voluntary Popular
Reflex eye movement Voluntary eye movement Gabersek and Jobert [140]
Elicited extension Guided placing Wolgin et al. [141]
Ballistic movement Terminal guidance Paillard [10]
Reaching channel Grasping channel Jeannerod [142]
Automatic Conscious Pe´lisson et al. [46]
Egocentric reference frame Exocentric reference frame Bridgeman [6], (2006)
This non-exhaustive list of distinctions picked-up in the literature concerns the domains of perception and action. If there are some major overlaps between several of these
distinctions, they can nevertheless be reduced to one. Several reviews of these parallel or successive proposals have been published [3,4,13,23,40].
Fig. 3. Performances of ten patients presenting with optic ataxia. Three patients with a right hemisphere lesion (No. 1 to 3) and seven patients with a left hemisphere lesion (No. 4 to
10) were tested in the left (LVF), central (CVF) and right (RVF) areas of their visual field for grasping movement directed towards an object in the CVF condition, objects were presented
in the central area of the visual field, but not specifically on the fixation point. The performances of five control subjects are presented in the dotted line bar graphs. This figure only
presents the movements performed with the contralesional hand. Errors analyzed on a video recording were classed as corrected when the patient reached his or her goal (white bar
graphs) or not corrected (striped bar graphs). In addition, performances of patients when they actually stare at an object in foveal vision are represented in black dots (not illustrated
on the original figure). This figure reveals that movements in foveal vision are normal, and that corrected but essentially non-corrected errors appear in the contralesional peripheral
vision. Notice that the figure reproduced by Milner and Goodale [13] in their book (part 4.2.1.) takes out the central areas (CVF) of the original figures (from [55]).
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140134
If we review the data available in publications concerning optic
ataxia and visual agnosia, it appears that (1) the arguments in favor
of a dissociation between perception and action in optic ataxia are
decidedly weak, and (2) the conditions used to contrast visuomotor
performance in DF (central vision) and in patients with ataxia
(peripheral vision) are not comparable, rendering evidence for a
double dissociation inadequate. Patients with optic ataxia make
substantial errors only in the peripheral visual hemifield, validated
by pointing and grasping tasks [55,59,64,73,74,84,85], whereas the
absence of perceptual impairments has largely been demonstrated
in central vision only. Where it has been specifically assessed,
perceptual impairments have in fact been found in peripheral
vision [60,74,86–88]. Some patients studied specifically in the
context of optic ataxia [73,89], have even have been studied by
others authors chiefly with regard to their prominent perceptual
impairments (e.g. [86,90]). Thus, the existence and nature of the
single dissociation in optic ataxia is uncertain, and the famous
double dissociation with visual agnosia even more so [79].
Regarding visual agnosia, the fact that Goodale and Milner’s
portrayal was based on the study of a single patient showed
theoretical but also statistical weaknesses. Revisited and considered
in a comprehensive manner, the classical dissociation between
impaired perception and preserved action in patient DF may be more
accurately viewed as a pattern of relative preservation of visuomotor
function, without normal performance in either domain (a strong
rather than a classical dissociation; e.g. [91]), thereby reducing the
inferential significance of these results [92,93]. Even more dramatic,
recent results have revealed that DF actually presents a pointing
pattern in peripheral vision similar to the one that characterizes optic
ataxia [94,95]. The updated analysis of this patient’s brain imaging
confirms dorsal stream damage, in both hemispheres, that would be
consistent with a co-existing diagnosis of optic ataxia. This striking
observation, of co-existing visual agnosia and optic ataxia, even if
they have discrete anatomical bases, fundamentally undermines the
prior claim that DF’s behavior can be taken as one half of a double
dissociation between these conditions. Such considerations have led
several authors now to suggest that we reconsider the duplex theory
of Milner and Goodale [4,79,85,93,96,97].
2.1. Rich and interconnected anatomical networks
In spite of this emphasis on the dissociations within the
visual system, an integrated behavior must necessarily involve
projections onto motor structures from both the dorsal and
ventral pathways. The projections from the ventral system onto
the motor areas of the frontal cortex, less direct than those from
the dorsal system, can involve the sub-cortical structures (basal
ganglia); and the TE areas project onto the prefrontal cortex, and
then on the motor areas (see Fig. 1). These two anatomical
projection systems may be articulated with behavioural arguments
in favor of a duality of visuomotor actions [98]. The famous
figure 4 of Felleman and Van Essen [27] listed more than
30 cortical areas that receive retinal projections, organized in a
rich network within which more than 300 reciprocal connections
were described. Given the extreme complexity of interconnection
within this more exhaustive representation of the visual network,
one can understand that academics preferred a version that
simplifies to three [24] or even two very distinct pathways
[13]. This simplification can be a helpful aid to theorizing, but the
reality remains that there are many interconnections within the
visual system ([37], Figures 1 and 2) as well as within the
visuomotor projections [4,99]. It is interesting to underline that
certain sub-cortical visual networks also project on prestriate
visual areas (see [100,101]).
The inferior row of Fig. 1 ([49]) presents a review of the
neuroanatomical projections of dorsal and ventral systems
connecting primary visual entrances (V1) with primary motor
exits (M1). The connection network obtained via a review of the
literature on primates reveals a multitude of ‘through-routes’
between V1 and M1. Beyond gross distinctions between cortical
and sub-cortical vision, between ventral and dorsal visual
pathways, or between the two occipitofrontal pathways, the
obvious conclusion is that these sub-systems are densely
interconnected, providing much scope for visuoperceptive and
visuomotor functions to interact [40,102].
A recent re-evaluation of dorsal cortical visual networks [103]
proposed three major systems differentiated beyond posterior
parietal cortex: a parietal-premotor system for goal-directed
action (eye, arm and hand movements) with the main pathway
linking V6a and MIP to the PMd, but also including a pathway
between the VIP and the PMv; a parietal-mediotemporal system
linking the caudal intraparietal sulcus (IPS) to the mediotemporal
lobe, including the hippocampus, the posterior cingulate cortex
and the retrosplenial cortex, for orientation and spatial navigation;
and a parietal-prefrontal system having for its sources LIP, VIP, MT,
MST and projecting on the FEF (area 8a and area 46) for voluntary
Fig. 4. Visuomotor field test. Pointing tasks performed by a patient with unilateral optic ataxia. The test was performed in peripheral vision during central visual fixation (left
image) or in central vision when the patient can freely move the eyes (right image) Crosses represent the locations of all targets aimed at, and confidence ellipses describe the
confidence interval of the performance of healthy subjects. Red points represent mean errors made by the patient and are associated with an arrow when they are larger than
to those made by healthy subjects. We notice that significant errors (which can exceed 10 cm) are only observed in contralesional peripheral vision of the patient (from [60]).
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140 135
control of eye movement and spatial working memory. This new
interpretation of visuomotor networks highlights the existence of
several anatomo-functional systems involved in visuomotor
organization.
2.2. The anatomical illusion
Functional dissociations observed in the healthy subject are
also potentially informative regarding sensorimotor organization.
Prominent among these dissociations is the very appealing claim
that contextual visual illusions have a greater impact on perceptual
responses than on motor ones, suggesting different underlying
visual representations for these classes of response. However,
some major objections can be applied to using this argument to
defend a perception-action dissociation mapped onto ventral and
dorsal streams.
A first consideration is that the motor responses used in this
type of experiment are quite basic (reaching or grasping gestures)
and the action parameters used to characterize the response
correspond to motor variables, such as grip aperture, controlled in
an automatic manner. In a highly influential experiment, Aglioti
et al. [104] used the Ebbinghaus size-contrast display, in which a
graspable central chip is perceived as smaller when surrounded by
large circles, and as larger when surrounded by small circles. The
authors reported a clear biasing effect of the surrounding circles
when subjects were asked to direct their grasp to the chip that
appears larger (or smaller), but little effect upon the grip aperture
[104,105]. The conclusion drawn was that, contrarily to perception,
action is not affected by the illusion. But this implies a very narrow
definition of ‘action’, specifically the automatic control of some
action subroutines, within a wider context in which the intentional
directing of the grasping response was influenced by the illusion.
Thus, even if the basic result is accepted (see below), the
dissociation indicated would be between the automatic control
of the grasp and a higher level of control of intentional action.
Second, many have contested the basic result (see [106], for a
recent pre-registered re-examination of the immunity of grasping
to the Ebbinghaus illusion), and it is certainly the case that the
motor system is not, in general, immune to optical illusions. Some
illusions can have effects that are identical for perceptual and
motor responses (e.g. [107,108]), or even more pronounced for
motor responses [109]; the pattern of results obtained may depend
critically upon the particular stimulus and response conditions
employed, and do not indicate any general dissociation between
perception and action responses (e.g. [110–112]). Particular
patterns of dissociation that may be observed empirically need
to be explained in terms of the particular combinations of cues
used in forming different responses.
Third, particular patterns of dissociation cannot be mapped on
to the anatomical distinction between dorsal and ventral streams,
in the absence of relevant experiments. Coello et al. [113] initiated
the exploration of anatomical supports for perceptual and motor
effects of visual illusions. According to the prediction stemming
directly from the theory of two visual systems, a patient with a
bilateral lesion of the dorsal pathway should inevitably differentiate
herself from healthy controls and show a similar effect of the
illusions on perceptual and motor responses (Fig. 5). In fact, both
the perceptual and the motor responses of this patient were totally
similar to those of healthy controls free from brain lesions, and this
for the size-contrast illusion as well as the two other illusions
(sagittal and frontal Roelofs effects). The conclusion validated by
this study is that the behavioral dissociation initially observed in
healthy subjects cannot be related to the two dorsal and ventral
visual pathways, and that the organization of action also depends
on other visual pathways. This conclusion completes the following
arguments: on the one hand the duplex model is too narrow to
account for behavioural dissociations and absence of dissociations,
and on the other hand the perception itself does not depend solely
on the ventral pathway processes (e.g. [114–118]).
2.3. Are there only perception and action?
Another approach to the question of perception-action
dissociations as related to the anatomical dorsal/ventral distinction,
consists in comparing precisely the perceptual performances
of subjects with optic ataxia in the experimental conditions where
they show a sensorimotor deficit, i.e. with fixation maintained
throughout and all targets presented extra-foveally. The first
observation of this type was made in a fortuitous manner during an
experiment using delays during which the visual target could be
displaced without the knowledge of patients [74]. With each
displacement of the targets, pointing movements were initially
directed towards the target’s first position, suggesting that these
subjects acted ‘‘with a delay’’, consistent with the duplex theory.
But a side observation during this study was that the two patients
were apparently also very poor at detecting target displacements,
even when asked to make a perceptual report about whether the
target had moved during the delay. Following this observation, we
started more formally studying the perceptual responses of
patients with unilateral optic ataxia for stimuli presented in their
ataxic hemifield. Responses obtained were impaired for the two
dorsal-related visual qualities, position and orientation [60].
To meticulously explore the possibility of a deficit common to
action and perception, McIntosh et al. [88] adapted the double-step
paradigm in a patient presenting with bilateral optic ataxia,
requiring either manual pointing responses, or a simple discriminatory
button-press to report the detection of target jumps.
Visuomotor and perceptual responses were collected for target
jumps between different parts of the visual field. The close
correlation between perceptual and motor deficits, across the
various jump configurations (Fig. 6), strongly suggest that these two
deficits result from the same primary functional disorder [56,57].
In conclusion, it seems that the idea of a dissociation between
optic ataxia and visual agnosia should be rejected, at least on present
Fig. 5. Perceptive (white) and manual (black) responses to three visual illusions.
Visuoperceptive and visuomanual responses of healthy subjects (CSs) were tested
in three mock situations. For the Roelofs effect on the frontal plane (on the left side),
healthy subjects presented a strong perceptual illusion but no motor error. For the
illusion of size-contrast (in the center), the effects of the illusion are greater for the
perceptual responses (white) than motor ones (black). Finally, for the deep Roelofs
illusion, healthy subjects reveal errors of the same order for the two types of
responses. Predictions of the duplex theory for a subject without a dorsal pathway
are represented by broken red lines. In reality, performances of a patient presenting
with a bilateral lesion of the dorsal pathways were similar to those of controls, for
the three illusions tested. None of the situations evaluated yielded an argument in
favor of a key role of the posterior parietal cortex in visuomanual responses
performed like here in central vision (from [113]).
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140136
evidence. We have also argued that the alleged dissociation between
perception and action does not accurately match the experimental
conditions used and is far from reflecting the complexity of the entire
visual system. It consists only in object recognition and goal-directed
action, i.e. only two of the multiple functions of the visual and
visuomotor systems. The duplex theory did strongly promote the
study of the visuomotor tracking function as an integral aspect of
visual processing, but its dichotomy simplified too far the complexity
of the anatomical networks, as well as overstating the divisions
between visual and visuomotor functions, which might be better
described as gradients rather than rifts [4].
3. Rebuilding after the theory
The simplicity of the duplex theory has proven powerfully
seductive of scientific and philosophical minds. Countless works
were conducted within this theoretical framework, or referenced it.
This theory generated few data on the nature of visual perceptual
processing, and we saw that the action-dedicated vocabulary was
usurped to describe the lower level motor function to which this
two visual systems’ theory refers. One main contribution of works
on this theme has been to refine and even move the frontier
between visuoperceptive response and visuomotor response.
Goodale and Milner adopted the matching task used by Vighetto
[59] and Perenin and Vighetto [55] to discriminate between these
two levels. The matching task used a manual response to
communicate the conscious perception of a stimulus, so here the
visual information processing had a descriptive goal, but the
manual response itself was differentiated from the visuomotor task
only because the hand was not directed towards the stimulus. This
seemingly modest difference in protocol can in fact produce
startling differences in performance, revealing that only the action
performed directly towards the stimulus could be distinguished
from the verbal response. A similar approach could also help
differentiate between two very close responses such as locating a
tactile stimulus and reaching it [119]. The goal of this distinction
was to highlight the specificity of the immediate target-directed
motor response vs. all responses involving an additional mediation
between the stimulus and the response, such as a delay, spatial
separation, or verbalization [4,120,121].
The comparison between perception, a cognitive function
related to consciousness, and visuomotor tracking, a sub-function
or module of the sensorimotor organization of action, is not
balanced in terms of theoretical ‘weight’ or complexity. Visuomotor
tracking is by nature a smaller functional element, easier to
define and isolate for study, which may help explain why most
publications dedicated to the two visual systems have focused on
this side of the distinction. We want to propose here that this
sensorimotor module corresponds in reality to the parietal
automatic pilot evidenced in healthy subjects, and in patients
with ‘‘action blindsight’’.
3.1. Blindsight in action
The ‘‘blindsight’’ oxymoron represents another interesting
neurological deficit to consider in regards to the theoretical
framework of dissociations between implicit and explicit sensory
processing. Within this neuropsychological condition, three major
sub-categories have been proposed: action blindsight, attentionblindsight
and agnosopsia (object processing without awareness)
[23]. The first studies conducted on patients presenting with
cortical blindness following a lesion of the visual cortex, showed
that in spite of the amputation of a visual hemifield, some of them
retained the ability to orientate their eye [16] or hand [18] towards
a visual stimulus presented briefly in that hemifield (see [22]).
After it was initially interpreted as a residual sub-cortical visual
function, this motor ability was subsequently observed in more
complex tasks. For example, we showed that patients with
blindsight were able to orientate their hand and adapt their grip
aperture to visual objects they could not consciously see
[40,122]. Similarly to the patient who presented with visual
agnosia, this manual ability was only observed in the conditions of
a goal-directed action and not in conditions of visual matching
tasks. However, we observed a far greater variability with cortical
blindness than in visual agnosia (see [4]), which can be easily
explained since the occipital lesions responsible for hemianopia
are associated with an alteration of occipito-parietal projections.
The neuroanatomical substrate brought up today to explain this
implicit sensorimotor function is a system of subcorticocortical
projections connecting the superior colliculus to the posterior
parietal cortex via the pulnivar (see [23,100]). The blindsight
phenomenon and particularly its expression in the context of
action, was considered as a strong argument for the dissociation
between the dorsal (occipito-parietal) and ventral (occipitotemporal)
pathways of the visual system (e.g. [3,13,40,49,123]).
3.2. Parietal automatic pilot
We described earlier the existence of visuomotor tracking
independent of conscious perception in the experiment conducted
by Prablanc et al. (1986). This tracking allows the hand to reach the
position of a target that was moved without the subject’s
knowledge. Afterwards, a series of experiments studied the degree
of autonomy of this visuomotor hand tracking, by specifically
analyzing the effects of movement duration on this type of control
[53,124]. A green target was presented at the beginning of each test
and subjects were requested to point to it under different speed
constraints. This target remained still in 80% of the tests, or could
jump in an unpredictable manner to the left or right side at the
beginning of the movement. In the so-called ‘‘position-stop’’
experimental condition, subjects had to systematically interrupt
their movement in response to the target jumps. Thus, the
direction of the jump was not relevant for this task. A strict respect
Fig. 6. Correlation between perceptual and motor deficits in optic ataxia. This figure
represents visuoperceptive performances (measured by the increased reaction time
vs. controls) of patient IG (bilateral optic ataxia) according to her visuomotor
performances (expressed in percentage of correction vs. controls). The correlation
between these two measures is excellent (from [88]).
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140 137
of the instructions to stop would mean that subjects reached the
first position of the target when their movement was too quick to
give them time to process the disruption, or they succeeded in
interrupting their movements. Conversely to these predictions, a
very significant number of movements were re-directed to the
target’s second position. These corrections were made in the
direction of the target, in spite of the fact that the subject had not
been instructed to follow the target. After having touched the
second target, subjects most often expressed a strong feeling of
frustration, corresponding to the impression that their hand was
not responding properly.
A movement time effect was described on the type of response
produced by subjects. In healthy subjects, most corrected
movements have the same duration as non-disrupted movements.
Because they appear in a limited temporal window, escaping the
slower processes of voluntary control, the corrections observed can
be presumed to result from an inability to inhibit an automatic
process of visuomotor tracking of the arm. This ‘‘automatic pilot’,
activated by default during the execution of the movement,
induces subjects to make corrections during a narrow temporal
window (about 200 to 300 ms), even if they have not been asked to
do so. During a portion of this temporal window we observed that
the correction rate produced by subjects in the ‘‘position-stop’’
condition could not be differentiated from the rate produced by
subjects in the ‘‘position-correction’’ condition, where they had to
correct their movement to reach the position of the second target.
These fast movements are in fact corrected by the sole action of the
‘‘automatic pilot’’ of the hand movement independent from
attention. Only movements with a duration > 300 ms seem to
come under the control of voluntary processes.
The study of movements in one patient (IG) presenting with a
bilateral lesion of the dorsal pathway (and thus optic ataxia)
evidenced the loss of this automatic pilot, whereas the intentional
processes of action control were retained [124]. This observation led
to the conclusion that fast movements are controlled by a Parietal
Automatic Pilot for which the dorsal pathway constitutes a necessary
substrate [125]. Slower movements, lasting more than 300 ms can
additionally come under the influence of intentional motor
processes that may be largely independent from the posterior
parietal cortex. Patients with frontal lesions, tested in the same task,
showed a loss of ability to voluntarily inhibit the automatic pilot and
thus by extension performed automatic corrections even for their
slower, voluntarily controlled movements [121].
These studies complete the results obtained with blindsight
[40], numb-sense [120], and optic ataxia (e.g. [64,74]) to extend the
need to take into account a major, invisible, functional property of
anatomo-functional networks, the temporal dimension.
4. Conclusions
Scientific theories are differentiated from experimental facts by
the clarification they bring to the interpretation of empirical data. It
is unavoidable for a theory to escape from the mass of data in order
to bring some sense to it, but by leading to more studies it may also
generate data incompatible with the theory itself. As incompatible
observations accumulate, they may outweigh the theory, and
impose a more or less extensive reappraisal and reformulation. At
the end of this review, it is important to note that the duplex theory
has played this role to the fullest: it generated a large number of
studies and results, made popular an unknown scientific field of
investigation, and produced predictions that were tested by
experiment. Like any scientific theory, it needs to evolve in line
with the data. The main challenge of this evolution is to rebuild a
more nuanced interpretation, based on the rich, multidisciplinary
data (during the proof correction of this paper a relevant paper to
this debate was published by David Milner: Milner [126]).
We underlined how the anatomical data were less and less
supporting the idea of a simple dichotomy of visual brain
organization. Several anatomical data were differentiated within
the occipito-parieto-frontal and occipito-temporo-frontal networks
(e.g. [49,103,127]). Among those an occipito-parietal
pathway involving the V6 and V6a areas, lesional location of optic
ataxia [56] has been associated to an automatic processing of the
‘‘where’’ and the ‘‘how’’ [103]. This pathway seems to correspond
to the one spared in action blindsight [23], enabling us to propose
that optic ataxia and action blindsight represent the negative and
positive neuropsychological image of this parietal automatic pilot.
Specifically, this proposal suggests that automatic correction of
reaching behaviour should be demonstrable in blindsight.
Although it is interesting to be able to identify this facet of the
visual brain on an anatomical and functional level, this visuomotor
function remains only a small element of the human visual and
motor organization. In the framework of the duplex theory, all
other visual pathways were essentially considered in opposition to
this automatic visuomotor pathway, as if they represented in a
monolithic manner ‘‘the perception’’. The exploration of the visual
brain needs to continue to refine the respective contributions of
these other processing pathways for retina information, and
identify the networks involved in action programming
[74,113,121,128], and perception [96,129]. Further studies should
also try to understand the role of attention (or attentions) in these
sensory and motor processes [56,130] (Pisella et al. [131]: this
issue). But above all we must explore the coordination between
perception and action in order to account for the most elaborated
functions of the perception-action cycle allowing to elaborate, test
and update our internal model of the outside world (e.g. [132]).
Disclosure of interest
The authors have not supplied their declaration of competing
interest.
Acknowledgements
Authors would like to sincerely thank Alain Vighetto, Fadila
Hadj-Bouziane, Hisaaki Ota and Denis Pe´lisson for their stimulating
interactions for this chapter. This work was supported by
Hospices Civils de Lyon, Inserm, CNRS, the university of Edimburgh
and Labex Cortex (Labex/Idex ANR-11-LABX-0042).
References
[1] Rossetti Y, Revonsuo A. Beyond dissociations: reassembling the mind-brain
after all? In: Rossetti Y, Revonsuo A, editors. Beyond dissociation: interaction
between dissociated implicit and explicit processing. Amsterdam: Benjamins;
2000. p. 1–16.
[2] Schneider GE. Two visual systems. Science 1969;163:895–902.
[3] Jeannerod M, Rossetti Y. Visuomotor coordination as a dissociable function:
experimental and clinical evidence. In: Kennard C, editor. Visual perceptual
defects Baille`re’s Clinical Neurology International Practise and Research, Vol.
2, No. 2. London: Ballie`re Tindall; 1993. p. 439–60.
[4] Rossetti Y, Pisella L. Several ‘vision for action’ systems: a guide to dissociating
and integrating dorsal and ventral functions. In: Prinz W, Hommel B, editors.
Attention and Performance XIX; Common mechanisms in perception and
action. Oxford: Oxford Univ. Press; 2002. p. 62–119.
[5] Bridgeman B, Kirch M, Sperling A. Segregation of cognitive and motor aspects
of visual function using induced motion. Percept Psychophys 1981;29:336–42.
[6] Bridgeman B. Complementary cognitive and visuo-motor image processing.
In: Obrecht G, Stark LW, editors. Presbyopia research: from molecular biology
to visual adaptation. New York: Plenum Press; 1991. p. 189–98.
[7] Bridgeman B. Conscious vs unconscious processes: the case of vision. Theory
Psychol 1992;2(1):73–88.
[8] Bridgeman B. Interactions between vision for perception and vision for
behaviour. In: Rossetti Y, Revonsuo A, editors. Beyond dissociation: interaction
between dissociated implicit and explicit processing. Amsterdam: Benjamins;
2000. p. 17–39.
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140138
[9] Paillard J. Cognitive versus sensorimotor encoding of spatial information. In:
Ellen P, Thinus-Blanc, editors. Cognitive processes and spatial orientation in
animal and man. Dordrecht: Nijhoff Publ; 1987. p. 43–77.
[10] Paillard J. Motor and representational framing of space. In: Paillard J, editor.
Brain and space. Oxford: Oxford University Press; 1991. p. 163–81.
[11] Goodale MA, Milner AD. Separate visual pathways for perception and action.
Trends Neurosci 1992;15(1):20–5.
[12] Jeannerod M. The hand and the object: the role of posterior parietal cortex in
forming motor representations. Can J Physiol Pharmacol 1994;72(5):535–41.
[13] Milner AD, Goodale MA. The Visual Brain in Action. Oxford: Oxford University
Press; 1995.
[14] Humphreys NK, Weiskrantz L. Vision in monkey after removal of striate
cortex. Nature 1967;215:595–7.
[15] Mohler CW, Wurtz RH. Role of striate cortex and superior colliculus in visual
guidance of saccadic eye movements in monkeys. J Neurophysiol
1977;40:74–94.
[16] Po¨ppel E, Held R, Frost D. Residual visual function after brain wounds
involving the central visual pathways in man. Nature 1973;243:295–6.
[17] Perenin MT, Jeannerod M. Residual vision in cortically blind hemiphields.
Neuropsychologia 1975;13(1):1–7.
[18] Weiskrantz L, Warrington EK, Sanders MD, Marshall J. Visual capacity in hemianopic
field following a restricted occipital ablation. Brain 1974;97:709–28.
[19] Perenin M-T, Jeannerod M. Visual function within the hemianopic field
following early cerebral hemidecortication in man. I. Spatial localisation.
Neuropsychologia 1978;16:1–13.
[20] Ptito A, Lepore F, Ptito M, Lassonde M. Target detection and movement
discrimination in the blind field of hemispherectomized patients. Brain
1991;114(Pt 1B):497–512.
[21] Jackson SR. Perception, awareness and action: insights from blindsight. In:
Rossetti Y, Revonsuo A, editors. Beyond dissociation: interaction between
dissociated implicit and explicit processing.. Amsterdam: Benjamins; 2000.
p. 73–98.
[22] Weiskrantz L. Blindsight. A case study and implications. Oxford: Oxford
University Press; 1986.
[23] Danckert J, Rossetti Y. Blindsight in action: what can the different sub-types
of blindsight tell us about the control of visually guided actions? Neurosci
Biobehav Rev 2005;29(7):1035–46.
[24] Boussaoud D, Ungerleider LG, Desimone R. Pathways for motion analysis:
cortical connections of the medial superior temporal and fundus of the
superior temporal visual areas in the macaque. J Comp Neurol
1990;296(3):462–95.
[25] Jeannerod M, Arbib MA, Rizzolatti G, Sakata H. Grasping objects. The cortical
mechanisms of visuomotor transformation. Trends Neurosci 1995;18:314–20.
[26] Jeannerod M. The cognitive neuroscience of action. Oxford: Blackwell; 1997.
[27] Felleman DJ, Van Essen DC. Distributed hierarchical processing in the primate
cerebral cortex. Cereb Cortex 1991;1(1):1–47.
[28] Zeki S. A vision of the brain. Oxford: Blackwell; 1993.
[29] Ungerleider L, Mishkin M. Two cortical visual systems. In: Ingle DJ, Goodale
MA, Mansfield RJW, editors. Analysis of motor behavior. Cambridge: MIT
Press; 1982. p. 549–86.
[30] Ungerleider LG. Functional brain imaging studies of cortical mechanisms for
memory. Science 1995;270(5237):769–75.
[31] Hyvarinen J, Poranen A. Function of the parietal associative area 7 as revealed
from cellular discharges in the alert monkey. Brain 1974;97:673–92.
[32] Taira M, Mine S, Georgopoulos AP, Murata A, Sakata H. Parietal cortex
neurons of the monkey related to the visual guidance of hand movements.
Exp Brain Res 1990;83:29–36.
[33] Jeannerod M. The neural and behavioural organization of goal-directed
movements. Oxford psychology series, No. 15. New York, NY, US: Clarendon
Press/Oxford University Press; 1988, xii, 283 pp.
[34] Sakata H, Taira M. Parietal control of hand action. Curr Opin Neurobiol
1994;4:847–56.
[35] Sakata H, Taira M, Kusunoki M, Murata A, Tanaka Y. The parietal association
cortex in depth perception and visual control of hand action. Trends Neurosci
1997;20(8):350–6.
[36] Jeannerod M. The 25th Bartlett Lecture. To act or not to act: perspectives on
the representation of actions. Q J Exp Psychol A 1999;52(1):1–29.
[37] Bullier J. Integrated model of visual processing. Brain Res Brain Res Rev
2001;36(2–3):96–107 [Review. PubMed PMID: 11690606].
[38] Bullier J, Schall JD, Morel A. Functional streams in occipito-frontal connections
in the monkey. Behav Brain Res 1996;76(1–2):89–97.
[39] Morel A, Bullier J. Anatomical segregation of two cortical visual pathways in
the macaque monkey. Vis Neurosci 1990;4(6):555–78.
[40] Rossetti Y. Implicit short-lived motor representations of space in brain
damaged and healthy subjects. Conscious Cogn 1998;7(3):520–58.
[41] Jeannerod M. Mechanisms of visuo-motor coordination: a study in normals
and brain-damaged subjects. Neuropsychologia 1986;24:41–78.
[42] Jeannerod M. The formation of finger grip during prehension. A cortically
mediated visuomotor pattern. Behav Brain Res 1986;19:99–116.
[43] Faillenot I, Toni I, Decety J, Gre´goire MC, Jeannerod M. Visual pathways for
object-oriented action and object recognition: functional anatomy with PET.
Cerebral Cortex 1997;7(1):77–85.
[44] Rossetti Y, Pisella L. L’ego Lego : de´construire ou reconstruire le cerveauesprit
? Intellectica 2000;2(31):137.
[45] Bridgeman B, Hendry D, Stark L. Failure to detect displacement of the visual
world during saccadic eye movements. Vision Res 1975;15:719–22.
[46] Pe´lisson D, Prablanc C, Goodale MA, Jeannerod M. Visual control of reaching
movements without vision of the limb. II. Evidence of fast unconscious
processes correcting the trajectory of the hand to the final position of a
double-step stimulus. Exp Brain Res 1986;62(2):303–11.
[47] Bridgeman B, Lewis S, Heit F, Nagle M. Relation between cognitive and motororiented
systems of visual perception. J Exp Psychol Hum Percept Perform
1979;5:692–700.
[48] Goodale MA, Pe´lisson D, Prablanc C. Large adjustments in visually guided
reaching do not depend on vision of the hand or perception of target
displacement. Nature 1986;320(6064):748–50.
[49] Rossetti Y, Pisella L, Pe´lisson D. Eye blindness and hand sight: temporal
constraints and visuo-motor network. Vis Cogn 2000;7(6):785–809.
[50] Roelofs C. Optische localisation. Archives fu¨r Augenheilkunde 1935;109:395–
415.
[51] Dassonville P, Bala J. Perception, action, and Roelofs effect: a mere illusion of
dissociation. PLoS Biol 2004;2(11):e364.
[52] Pisella L, Mattingley JB. The contribution of spatial remapping impairments
to unilateral visual neglect. Neurosci Biobehav Rev 2004;28(2):181–200.
[53] Pisella L, Rossetti Y. Neuropsychologie des relations perception/action. In:
Coello Y, Honore H, editors. Percevoir, s’orienter, agir dans l’espace : approche
pluridisciplinaire des relations perception-action. 2002.
[54] Devinsky O, Farah M, Barr WB. Visual agnosia. In: Goldenberg G, Miller B,
editors. Handbook of clinical Neurology 3rd Series, Vol. 88. Neuropsychology
and Behavioral Neurology. UK: Elsevier; 2008. p. 417–27 [Chapter 21].
[55] Perenin MT, Vighetto A. Optic ataxia: a specific disruption in visuomotor
mechanisms. I. Different aspects of the deficit in reaching for objects. Brain
1988;111:643–74.
[56] Pisella L, Ota H, Vighetto A, Rossetti Y. Optic ataxia and Balint’s syndrome:
neuropsychological and neurophysiological prospects. Handb Clin Neurol
2008;88:393–415.
[57] Biotti D, Pisella L, Vighetto A. Le syndrome de Balint et les fonctions spatiales
du lobe parie´tal. Rev Neurol (Paris) 2012;168(10):741–53.
[58] Rossetti Y, Pisella L. Optic ataxia: beyond the dorsal stream cliche´. In: The
Parietal lobes. Neurological and neuropsychological deficits A Volume in
Handbook of Clinical Neurology; 2017. Series Editors: Michael Aminoff, Dick
Swaab and Franc¸ois Boller; Volume Editors: Giuseppe Vallar & H. Branch
Coslett (in press).
[59] Vighetto A. E´tude neuropsychologique et psychophysique de l’ataxie optique.
The`se de Doctorat de l’Universite´ Claude Bernard Lyon I; 1980.
[60] Rossetti Y, Ota H, Blangero A, Vighetto A, Pisella L. Why does the perceptionaction
functional dichotomy not match the ventral-dorsal streams anatomical
segregation: optic ataxia and the function of the dorsal stream. In: Gangopadhyay
M, Madary M, Spicer F, editors. Perception, action, and consciousness:
sensorimotor dynamics and two visual systems. 2010. p. 163–81.
[61] Blangero A, Delporte L, Vindras P, Ota H, Revol P, Boisson D, et al. Optic ataxia
is not only ‘optic’: impaired spatial integration of proprioceptive information.
NeuroImage 2007;36:61–8.
[62] Vindras P, Blangero A, Ota H, Reilly KT, Rossetti Y, Pisella L. The pointing errors
in optic ataxia reveal the role of ‘‘Peripheral Magnification’’ of the PPC. Front
Integr Neurosci 2016;10:27.
[63] Jakobson LS, Archibald YM, Carey DP, Goodale MA. A kinematic analysis of
reaching and grasping movements in a patient recovering from optic ataxia.
Neuropsychologia 1991;29:803–9.
[64] Milner AD, Dijkerman HC, McIntosh RD, Rossetti Y, Pisella L. Delayed reaching
and grasping in patients with optic ataxia. Prog Brain Res 2003;142:225–42.
[65] Vighetto A, Perenin MT. Optic ataxia: analysis of eye and hand responses in
pointing at visual targets. Rev Neurol (Paris) 1981;137:357–72.
[66] Jeannerod M, Decety J, Michel F. Impairment of grasping movements
following bilateral posterior parietal lesion. Neuropsychologia 1994;32:
369–80.
[67] Milner AD, Goodale MA. Two visual systems re-viewed. Neuropsychologia
2008;46:774–85.
[68] Milner AD, Perrett DI, Johnston RS, Benson PJ, Jordan TR, Heeley DW, et al.
Perception and action in ‘visual form agnosia’. Brain 1991;114:405–28.
[69] Goodale MA, Milner AD, Jakobson LS, Carey DP. A neurological dissociation
between perceiving objects and grasping them. Nature 1991;349(6305):154–6.
[70] Goodale MA, Milner AD. The visual brain in action. OUP; 2005.
[71] Milner AD, Heywood CA. A disorder of lightness discrimination in a case of
visual form agnosia. Cortex 1989;25(3):489–94.
[72] Rossetti Y, Pisella L. Une vision a` deux vitesses. Sci Vie 2001;216:68.
[73] Milner AD, Paulignan Y, Dijkerman HC, Michel F, Jeannerod M. A paradoxical
improvement of misreaching in optic ataxia: new evidence for two separate
neural systems for visual localization. Proc Biol Sci 1999;266:2225–9.
[74] Rossetti Y, Revol P, McIntosh R, Pisella L, Rode G, Danckert J, et al. Visually
guided reaching: bilateral posterior parietal lesions cause a switch from
fast visuomotor to slow cognitive control. Neuropsychologia 2005;43(2):
162–77.
[75] Goodale MA, Jakobson LS, Keillor JM. Differences in the visual control of
pantomimed and natural grasping movements. Neuropsychologia 1994;32:
1159–78.
[76] Goodale MA, Milner AD. Sight Unseen, An Exploration of Conscious and
Subconscious Vision, seconde e´dition, OUP; 2013.
[77] Milner AD, Goodale MA. Cortical visual systems for perception and action. In:
Gangopadhyay M, Madary M, Spicer F, editors. Perception, action, and
consciousness: Sensorimotor Dynamics and Two Visual Systems. 2010. p.
71–94.
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140 139
[78] Goodale MA, Meenan JP, Bu¨ lthoff HH, Nicolle DA, Murphy KJ, Racicot CI.
Separate neural pathways for the visual analysis of object shape in perception
and prehension. Curr Biol 1994;4(7):604–10.
[79] Pisella L, Binkofski F, Lasek K, Toni I, Rossetti Y. No double-dissociation
between optic ataxia and visual agnosia: multiple sub-streams for multiple
visuo-manual integrations. Neuropsychologia 2006;44:2734–48.
[80] Dijkerman HC, McIntosh RD, Anema HA, de Haan EHF, Kapelle LJ, Milner AD.
Reaching errors in optic ataxia are linked to eye position rather than head or
body position. Neuropsychologia 2005;44:2766–73.
[81] Jackson SR, Newport R, Mort D, Husain M. Where the eye looks, the hand
follows; limb-dependent magnetic misreaching in optic ataxia. Curr Biol
2005;15:42–6.
[82] Khan AZ, Pisella L, Vighetto A, Cotton F, Luaute´ J, Boisson D, et al. Optic ataxia
errors depend on remapped, not viewed, target location. Nat Neurosci
2005;8:418–20.
[83] Blangero A, Ota H, Rossetti Y, Fujii T, Ohtake H, Tabuchi M, et al. Systematic
retinotopic reaching error vectors in unilateral optic ataxia. Cortex
2010;46(1):93–7.
[84] Revol P, Rossetti Y, Vighetto A, Pisella L. Pointing errors in immediate and
delayed conditions in unilateral optic ataxia. Spat Vis 2003;16:347–64.
[85] Rossetti Y, Vighetto A, Pisella L. Optic ataxia revisited: immediate motor
control versus visually guided action. Exp Brain Res 2003;153:171–9.
[86] Michel F, He´naff MA. Seeing without the occipito-parietal cortex: simultagnosia
as a shrinkage of the attentional visual field. Behav Neurol 2004;15:3–13.
[87] Striemer C, Blangero A, Rossetti Y, Boisson D, Rode G, Vighetto A, et al. Deficits
in peripheral visual attention in patients with optic ataxia. Neuroreport
2007;18(11):1171–5.
[88] McIntosh RD, Mulroue A, Blangero A, Pisella L, Rossetti Y. Correlated deficits of
perception and action in optic ataxia. Neuropsychologia 2011;49(1):131–7.
[89] Cavina-Pratesi C, Connolly JD, Milner AD. Optic ataxia as a model to investigate
the role of the posterior parietal cortex in visually guided action:
evidence from studies of patient M.H. Front Hum Neurosci 2013;7:336.
[90] Snow JC, Miranda RR, Humphreys GW. Impaired visual sensitivity within the
ipsilesional hemifield following parietal lobe damage. Cortex 2013;49(1):
158–71.
[91] Crawford JR, Garthwaite PH, Gray CD. Wanted: fully operational definitions of
dissociations in single-case studies. Cortex 2003;39(2):357–70.
[92] Himmelbach M, Boehme R, Karnath HO. 20 years later: a second look on DF’s
motor behaviour. Neuropsychologia 2012;50(1):139–44.
[93] Schenk T, McIntosh RD. Do we have independent visual streams for perception
and action? Cogn Neurosci 2010;1(1):52–62.
[94] Hesse C, Ball K, Schenk T. Pointing in visual periphery: is DF’s dorsal stream
intact? PLoS One 2014;9(3):e91420.
[95] Rossit S, Harvey M, Butler SH, Szymanek L, Morand S, Monaco S, et al.
Impaired peripheral reaching and on-line corrections in patient DF: optic
ataxia with visual form agnosia. Cortex 2017. http://dx.doi.org/10.1016/
j.cortex.2017.04.004.
[96] Pisella L, Sergio L, Blangero A, Torchin H, Vighetto A, Rossetti Y. Optic ataxia
and the function of the dorsal stream: contributions to perception and action.
Neuropsychologia 2009;47(14):3033–44.
[97] McIntosh RD, Schenk T. Two visual streams for perception and action: current
trends. Neuropsychologia 2009;47(6):1391–6.
[98] Milner AD, Dijkerman HC, Pisella L, McIntosh RD, Tilikete C, Vighetto A, et al.
Grasping the past. Delay can improve visuomotor performance. Curr Biol
2001;11:1896–901.
[99] Rossetti Y, Revonsuo A. Beyond dissociation: interaction between dissociated
implicit and explicit processing. Amsterdam: Benjamins; 2000.
[100] Bullier J, Girard P, Salin P-A. The role of area 17 in the transfer of information
to extrastriate visual cortex. In: Peters A, Rockland KS, editors. Cerebral
Cortex, 10. New York: Plenum Press; 1994. p. 301–30.
[101] Girard P. Anatomic and physiologic basis of residual vision after damage to
the primary visual area. Rev Neurol (Paris) 1995;151:8–9 [457-465].
[102] Bridgeman B, Peery S, Anand S. Interaction of cognitive and sensorimotor
maps of visual space. Percept Psychophys 1997;59(3):456–69.
[103] Kravitz DJ, Saleem KS, Baker CI, Mishkin M. A new neural framework for
visuospatial processing. Nat Rev Neurosci 2011;12(4):217–30.
[104] Aglioti S, DeSouza JFX, Goodale MA. Size-contrast illusions deceive the eye
but not the hand. Curr Biol 1995;5(6):679–85.
[105] Haffenden A, Goodale MA. The effect of pictorial illusion on prehension and
perception. J Cog Neurosci 1998;10:122–36.
[106] Kopiske KK, Bruno N, Hesse C, Schenk T, Franz VH. The functional subdivision
of the visual brain: Is there a real illusion effect on action? A multi-lab
replication study. Cortex 2016;79:130–52.
[107] Coello Y, Richaud S, Magne P, Rossetti Y. Vision for spatial perception and
vision for action: a dissociation between the left-right and near-far dimensions.
Neuropsychologia 2003;41(5):622–33.
[108] Dyde RT, Milner AD. Two illusions of perceived orientation: one fools all of
the people some of the time; the other fools all of the people all of the time.
Exp Brain Res 2002;144(4):518–27.
[109] Yamagishi N, Anderson SJ, Ashida H. Evidence for dissociation between the
perceptual and visuomotor systems in humans. Proc Biol Sci
2001;268(1470):973–7.
[110] Bruno N, Franz VH. When is grasping affected by the Mu¨ ller-Lyer illusion?: a
quantitative review. Neuropsychologia 2009;47(6):1421–33.
[111] Franz VH, Fahle M, Bu¨lthoff HH, Gegenfurtner KR. Effects of visual illusions on
grasping. J Exp Psychol Hum Percept Perform 2001;27(5):1124–44.
[112] Kerzel D, Gegenfurtner KR. Motion-induced illusory displacement reexamined:
differences between perception and action? Exp Brain Res
2005;162(2):191–201.
[113] Coello Y, Danckert J, Blangero A, Rossetti Y. Do visual illusions probe the visual
brain? Illusions in action without a dorsal visual stream. Neuropsychologia
2007;45(8):1849–58 [Erratum in: Neuropsychologia 2008;46(1):390–1].
[114] Gallese V, Craighero L, Fadiga L, Fogassi L. Perception through action. PSYCHE
1999;5(21).
[115] Fourtassi M, Rode G, Pisella L. Using eye movements to explore mental representations
of space. Ann Phys Rehabil Med 2016 [pii: S1877-0657(16)00044-0].
[116] Fourtassi M, Hajjioui A, Urquizar C, Rossetti Y, Rode G, Pisella L. Iterative
fragmentation of cognitive maps in a visual imagery task. PLoS One
2013;8(7):e68560.
[117] Pisella L. Visual perception is dependent on visuospatial working memory
and thus on the posterior parietal cortex. Ann Phys Rehabil Med 2016.
[118] Pisella L, Biotti D, Vighetto A. Combination of attentional and spatial working
memory deficits in Ba´lint–Holmes syndrome. Ann N Y Acad Sci
2015;1339(1):165–75.
[119] Rossetti Y, Re´gnier C. Representations in action: pointing to a target with
various representations. In: Bardy BG, Bootsma RJ, Guiard Y, editors. Studies
in Perception and Action III. Mahwah, NJ: Lawrence Erlbaum Associates, Inc;
1995. p. 233–6.
[120] Rossetti Y. In search of immaculate perception: evidence from motor perception
of space. In: Hameroff S, Kaszniak A, Chalmers D, editors. Towards a
science of consciousness. MIT Press; 1999. p. 141–8.
[121] Rossetti Y, Pisella L. Mediate responses as direct evidence for intention:
neuropsychology of Not to-, Not now- and Not there- tasks. In: Johnson-Frey
SH, editor. Taking action: cognitive neuroscience perspectives on the problem
of intentional acts. MIT Press; 2003. p. 67–105.
[122] Perenin MT, Rossetti Y. Grasping without form discrimination in a hemianopic
field. Neuroreport 1996;7(3):793–7.
[123] Milner AD. Neuropsychological studies of perception and visuomotor control.
Philos Trans R Soc Lond B Biol Sci 1998;353(1373):1375–84.
[124] Pisella L, Gre´a H, Tilikete C, Vighetto A, Desmurget M, Rode G, et al. An
automatic pilot for the hand in the human posterior parietal cortex toward a
reinterpretation of optic ataxia. Nat Neurosci 2000;3:729–36.
[125] Gaveau V, Pisella L, Priot AE, Fukui T, Rossetti Y, Pe´lisson D, et al. Automatic
online control of motor adjustments in reaching and grasping. Neuropsychologia
2014;55:25–40.
[126] Milner AD. How do the two visual streams interact with each other? Exp
Brain Res 2017. http://dx.doi.org/10.1007/s00221-017-4917-4.
[127] Rizzolatti G, Matelli M. Two different streams form the dorsal visual system:
anatomy and functions. Exp Brain Res 2003;153(2):146–57.
[128] Mattingley JB, Husain M, Rorden C, Kennard C, Driver J. Motor role of human
inferior parietal lobe revealed in unilateral neglect patients. Nature
1998;392(6672):179–82.
[129] Pisella L, Striemer C, Blangero A, Gaveau V, Revol P, Salemme R, et al.
Perceptual deficits in optic ataxia? In: Haggard P, Rossetti Y, Kawato M,
editors. Attention and performance XXI: sensorimotor foundations of higher
cognition. Oxford: Oxford Univ Press; 2007. p. 47–71 [Chapter 3].
[130] Nachev P, Husain M. Disorders of visual attention and the posterior parietal
cortex. Cortex 2006;42(5):766–73.
[131] Pisella L, Rossetti Y, Rode G. Optic ataxia in Ba´lint-Holmes syndrome. Ann
Phys Rehabil Med 2016 [pii: S1877-0657(16)00016-6].
[132] Land MF. Do we have an internal model of the outside world? Philos Trans R
Soc Lond B Biol Sci 2014;369(1636):20130045.
[133] Cajal SR. Histologie du syste`me nerveux de l’homme et des verte´bre´s. Paris:
Maloine; 1909.
[134] Humphrey NK. Vision in a monkey without striate cortex: a case study.
Perception 1974;3:241–55.
[135] Held R, Bauer Jr JA. Visually guided reaching in infant monkeys after restricted
rearing. Science 1967;155(3763):718–20.
[136] Goodale MA. Neural mechanisms of visual orientation in rodents: target vs
places. In: Hein A, Jeannerod M, editors. Spatially oriented behavior. New
York: Springer Verlag; 1983. p. 35–62.
[137] Neumann O, Klotz W. Motor responses to nonreportable, masked stimuli:
where is the limit of direct parameter specification? In: Umilta C, Moscovitch
M, editors. Attention and performance XV conscious and nonconscious
information processing. Cambridge: MIT Press; 1994. p. 123–50.
[138] Rossetti Y, Rode G, Boisson D. Implicit processing of somaesthetic information:
a dissociation between where and how? Neuroreport 1995;6(3):506–10.
[139] Rossetti Y, Rode G, Boisson D. Numbsense: A case study and implications. In:
De Gelder B, De Haan E, Heywood C, editors. Out of Mind: Varieties of
unconscious processes. Oxford University Press; 2001. p. 265–92.
[140] Gabersek V, Jobert F. [Influence of voluntary or reflex eye movements on
nystagmographic vestibular reactions]. Ann Otolaryngol Chir Cervicofac
1965;82(9):685–709.
[141] Wolgin DL, Hein A, Teitelbaum P. Recovery of forelimb placing after lateral
hypothalamic lesions in the cat: parallels and contrasts with development. J
Comp Physiol Psychol 1980;94(5):795–807.
[142] Jeannerod M. Specialized channels for cognitive responses. Cognition
1981;10(1–3):135–7.
Y. Rossetti et al. / Annals of Physical and Rehabilitation Medicine 60 (2017) 130–140140