Author’s Accepted Manuscript
Still holding after all these years: An actionperception dissociation in patient DF
Tzvi Ganel, Melvyn A. Goodale

www.elsevier.com/locate/neuropsychologia

PII:
DOI:
Reference:

S0028-3932(17)30342-1
http://dx.doi.org/10.1016/j.neuropsychologia.2017.09.016
NSY6498

To appear in: Neuropsychologia
Received date: 17 August 2017
Accepted date: 17 September 2017
Cite this article as: Tzvi Ganel and Melvyn A. Goodale, Still holding after all
these years: An action-perception dissociation in patient DF, Neuropsychologia,
http://dx.doi.org/10.1016/j.neuropsychologia.2017.09.016
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Still holding after all these years: An action-perception dissociation in patient DF

Tzvi Ganel1 and Melvyn A. Goodale2
1

Psychology Department, Ben-Gurion University of the Negev, Beer-Sheva 8410501, Israel

2

The Brain and Mind Institute, The University of Western Ontario, London, Ontario N6A 5B7, Canada

*Corresponding author: Department of Psychology, Ben-Gurion University of the Negev, Beer-Sheva
8410501, Israel. Phone: +972-8-6428515. tganel@bgu.ac.il

Abstract
Patient DF, who has bilateral damage in the ventral visual stream, is perhaps the best known
individual with visual form agnosia in the world, and has been the focus of scores of research
papers over the past twenty-five years. The remarkable dissociation she exhibits between a
profound deficit in perceptual report and a preserved ability to generate relatively normal
visuomotor behaviour was early on a cornerstone in Goodale and Milner’s (1992) two visual
systems hypothesis. In recent years, however, there has been a greater emphasis on the damage
that is evident in the posterior regions of her parietal cortex in both hemispheres. Deficits in
several aspects of visuomotor control in the visual periphery have been demonstrated, leading
some researchers to conclude that the double dissociation between vision-for-perception and
vision-for-action in DF and patients with classic optic ataxia can no longer be assumed to be
strong evidence for the division of labour between the dorsal and ventral streams of visual
processing. In this short review, we argue that this is not the case. Indeed, after evaluating DF’s
performance and the location of her brain lesions, a clear picture of a double dissociation
between DF and patients with optic ataxia is revealed. More than quarter of a century after the
initial presentation of DF’s unique case, she continues to provide compelling evidence for the
idea that the ventral stream is critical for the perception of the shape and orientation of objects
but not the visual control of skilled actions directed at those objects.
Key words: Perception and action, visual-form agnosia, visuomotor control, two-visual streams

2

1. Introduction
In the winter of 1988, a young woman was having a shower in her cottage in rural Italy. She and
her partner had recently purchased the cottage, and were slowly renovating it. The bathroom still
had an old-fashioned gas-fired water heater and unbeknownst to her the heater was improperly
vented. As she continued to shower, the carbon monoxide from the burning gas slowly
accumulated in the bathroom – and her brain, starved of oxygen, began to shut down. She
eventually passed out, but fortunately, her partner found her a few minutes after she had
collapsed and managed to get her to the hospital in the nick of time.
When she re-gained consciousness, it was immediately evident that DF’s vision had been
severely affected by the hypoxic event. Indeed, she seemed essentially blind and was unable to
see anything beyond light and dark. Over the next few days and weeks, her visual deficits
resolved considerably, but she continued to exhibit a profound visual form agnosia: she was
unable to recognize objects on the basis of their shape or form. Her ability to see colour and
visual texture seemed to be largely spared, however. For example, even though she could not
recognize her partner or his facial expression, if he happened to be wearing a cardigan of a
particular colour and material that she remembered, she would immediately be able to identify it
– and thus, him. As we were to discover later, however, her deficit in form vision was so severe
that she could not discriminate between the simplest of geometric shapes – or tell whether a line
was oriented horizontally or vertically.
David Milner, then at the University of St. Andrews, met the young woman some months
later when she returned to her native Scotland for a visit – and another neurological assessment.
Thus began a long and fruitful collaboration between David Milner, his colleague Mel Goodale,
by then in Canada, and the young woman now known as DF. In fact, there is little doubt that DF
is the most studied individual with visual form agnosia in the history of neuropsychology and
cognitive neuroscience.
It soon became evident that even though DF cannot discriminate between objects on the
basis of their form and orientation, she is able to use these same visual features to program and
control actions directed at those objects. For example, even though she could not indicate the
orientation of a slot either verbally or by matching the orientation of a hand-held card with that
of the slot, she had no difficulty ‘posting’ the card into the slot, rotating her hand in the correct
orientation well before making contact (Goodale et al., 1991). Similarly, even though she could

3

not indicate the width of a rectangular block placed in front of her by opening her finger and
thumb a matching amount, she scaled her grasp mid-flight to the width of the block when she
was asked to reach out and pick it up. When presented with pairs of wooden plaques with
different (or identical) outline contours, rather like pebbles one might find on the seashore, she
could not reliably say whether they were the same or different. Yet again, when she reached out
to pick up the plaques using a precision grasp, she immediately placed her finger and thumb on
stable grasp points on the edge of the plaque much like a normally sighted person would
(Goodale et al., 1994b). This sharp dissociation between defective perceptual report and spared
visually guided behaviour has been demonstrated in many different studies and across a broad
range of tasks in DF, from dealing with obstacles in locomotor tasks (Patla et al., 1997) to
grasping tools (Carey et al., 1996) and avoiding potential obstacles during reaching and grasping
(Rice et al., 2006). As Larry Weiskrantz might put it, DF seems to have ‘blindsight’ for visual
form.
The fact that DF has great difficulty perceiving the shape and orientation of objects,
despite being able to use these same visual features to guide her action, became one of the
cornerstones of Goodale and Milner’s (1992) two visual systems account of visual processing in
the primate brain. Although DF has damage in a number of brain areas, consistent with the
effects of carbon monoxide poisoning, she has pronounced bilateral lesions in the lateral
occipital cortex (LOC) a ventral-stream area that has been repeatedly demonstrated in fMRI and
TMS studies to be implicated in visual object recognition, particularly with respect to object
shape (e.g., Grill-Spector 2003). This fact, coupled with evidence that patients with damage to
the dorsal stream show pronounced visuomotor deficits but spared visual perceptual abilities
(e.g. Perenin & Vighetto, 1988), originally led to the proposed division of labour between the
ventral and dorsal visual streams that is central to the two visual systems hypothesis (TVSH).
These neuropsychological observations are supported by a broad range of findings from human
neuroimaging and psychophysics to non-human primate neurophysiology, neuroanatomy, and
lesion studies (for psychophysical evidence and reviews, see Chen et al., 2015; Ganel, et al.,
2008; Ganel et al., 2012; Ganel & Goodale, 2003, 2015; Namdar et al., 2016; for reviews of nonhuman primate work, see Andersen et al., 2014; Kravitz et al., 2011, 2013; for reviews of human
neuroimaging work, see Gallivan & Goodale, 2017; Goodale, 2011, 2014). Moreover, there is
evidence that other patients with developmental visual agnosia or with visual agnosia caused by

4

ventral-stream damage also show spared visuomotor abilities (e.g., Campion, 1987; Karnath et
al., 2009; Freud et al., 2016). It is important to note, in other words, that although DF has
figured prominently in discussions of the TVSH, the dissociations that she exhibits are only a
small part of the evidence for this influential account of the workings of the visual brain.
Nevertheless, some have argued that DF’s spared visuomotor abilities are not always
entirely ‘normal’, or at least not as normal as one would expect to see in a non-neurological
individual. Himmelbach et al., (2012), for example, compared DF’s performance on various
action tasks (recovered from the earlier published papers) to those of a group of present-day agematched control participants. Although DF’s visually guided grasping did not differ from
controls that Himmelbach et al. tested, her performance on the posting task and on grasping the
pebble-shaped objects was found to be significantly worse than their new controls. But even so,
her performance on all the action tasks, including posting and grasping pebble-shaped objects,
was always far superior to her performance on the corresponding perceptual tasks. Moreover,
this striking dissociation was still evident in Himmelbach et al.’s re-analysis of her data (for a
more detailed discussion of this issue, see Whitwell et al., 2014a). As it turns out, part of the
reason for DF’s slight deficit in posting and grasping pebble-shaped objects is likely due to some
difficulty she has in selecting a comfortable hand posture, not in rotating her hand to the correct
orientation. But more of that later.
It has also been suggested that DF’s spared visuomotor abilities reflect her use of haptic
information as an adjunct for her compromised visual abilities (Schenk, 2012). According to
this idea, haptic feedback about the locations of the edges of the target is available to calibrate
DF’s grip aperture in grasping tasks, but it is not available in standard visual perceptual tasks
when she is asked to give either a verbal report or to indicate the width of the target object by
opening her finger and thumb a matching amount. Thus, the argument goes, when haptic
feedback is removed, DF’s performance on a grasping task deteriorates. As a consequence, DF’s
apparently spared visuomotor abilities may not reflect the normal operation of intact visuomotor
modules in her dorsal stream. This claim, however, has been recently undercut by the
demonstration that the removal of tactile contact with the goal object at the end of a reach-tograsp movement leads DF, as well as normally-sighted controls, to produce pantomime
movements rather than authentic grasping movements (Whitwell et al., 2014a, 2014b, 2015). In
addition, providing DF with haptic feedback after she indicates with a matching response what

5

she thinks is the width of the target object does not improve her performance; she is still unable
to perceive the width of the target, even though she shows normal grip scaling when she picks it
up (Whitwell et al., 2014b). In short, there is no reason to believe that the dorsal-stream
mechanisms that are presumed to be mediating DF’s ability to reach out and grasp objects are
affected by haptic/tactile information any differently from those mediating the same behaviour in
the healthy brain (Milner et al., 2012).
But there is some need for caution. Although the bilateral lesions to LOC are generally
assumed to be the source of DF’s profound visual form agnosia, right from the beginning there
was evidence that these are not the only brain areas that are damaged in her brain (Milner et al.,
1991). This is not surprising given that patients suffering from hypoxia from carbon monoxide
poisoning show evidence for brain damage in a number of different brain areas (Prockop &
Chichkova, 2007). In DF’s case, there is indication that, in addition to the damage in the
ventrolateral cortex, there is evidence of bilateral cortical thinning in the posterior extent of the
intraparietal sulcus (pIPS) (Bridge et al., 2013; James et al., 2003). The presence of these
additional lesions, particularly those in the dorsal stream, has led to a recent flurry of papers that
have suggested that DF may have other visual problems and that this weakens the argument that
DF’s pattern of deficits and spared visual abilities provide support for the TVSH. In the
following section, we review this work and show that despite the presence of other visuomotor
anomalies, DF continues to exhibit clear evidence of a sharp dissociation between vision-forperception and vision-for-action.

2. DF' difficulties in selecting comfortable grip posture
As we have already discussed, people are remarkably sensitive to the orientation of an object
when they reach out and grasp it, and DF is no exception, typically rotating their wrist so that
they can place their fingers on the object properly. Because of the biomechanical constraints of
the wrist, however, there is another complication. When people are faced with the task of
picking up an elongated object such as a wooden dowel lying flat on a table, most orientations of
the dowel clearly afford variations of just one of two possible wrist posture patterns: either the
thumb is placed on one side of the dowel or it is placed on the other (see Figs. 1a, 1c). But there
are some orientations that will afford both of these postures; i.e., the two postures will be equally
(un)comfortable (see Figs. 1b, 1d). Notice that there is no perceptual ambiguity here: it is

6

patently clear what the orientation of the dowel is. But there is a motor ambiguity: it is not clear
whether the wrist should be rotated one way or the other when reaching out to grasp the object.

Figure 1: Depiction of typical postures for object orientations where posture selection is either stable or bistable.

Using fMRI, Wood et al. (2016) have recently demonstrated that the posterior part of
superior parietal lobule (pSPL) and the pIPS is differentially activated when individuals select
biomechanically appropriate (vs. inappropriate) postures to grasp objects such as those shown in
Figure 1. Ventral stream areas, however, show no such activation. In the same study, Wood et
al. showed that DF, who as just mentioned, has bilateral lesions in this same region of the dorsal
stream (Bridge et al., 2013), often selects uncomfortable hand postures when picking up
elongated objects – even though her hand is oriented ‘correctly’ when grasping the object.
Indeed, when we went back and looked carefully at the videos of her ‘posting’ in the
study by Goodale et al. (1991), we noticed that even though her hand was always rotated in the
correct orientation when she inserted the card into the slot, in a small fraction of the cases she
sometimes rotated her hand almost to the biomechanical limits of what is possible (see Fig. 2 for
representative images, and see supplementary materials for sample videos of DF’s performance
during posting and perceptual estimation of target rotation). This failure to select
biomechanically appropriate hand postures could explain why the orientation of her hand was
not quite as well-tuned as that of most healthy control participants. But again, it must be

7

emphasized that despite the fact that DF sometimes selected uncomfortable hand postures, she
always managed to correctly post the card into the slot, which stands in sharp contrast to her
complete inability to report the slot’s orientation, either verbally or by using a manual matching
response (see Fig. 2 and supplementary materials).

Posting
Trial 1

Trial 2

Trial 3

Matching
Trial 1

Trial 2

Trial 3

8
Figure 2: Posting and matching performance in patient DF. Photographs of representative trials of DF’s
performance during posting (upper panel) and matching (lower panel) tasks. Note, the while DF was successful in
posting the card into the slot in its correct orientation, her perceptual estimations of the orientation of the slot during
matching were entirely inappropriate. Also notice, that in a small percentage of the trials (illustrated in trial 2, upper
panel), DF did not use a comfortable posture to complete the posting task, even though the orientation of her hand
was correct. The snapshots were taken from 1990 short videos of DF’s performance in the two tasks. The videos are
provided as supplementary materials.

Although these dorsal-stream lesions could account for DF’s impaired selection of
comfortable grasp postures (Wood et al., 2016), what about other aspects of her visuomotor
performance? As reviewed earlier, there is wealth of evidence to support the idea that DF’s
remarkable dissociation between vision-for-perception and vision-for-action extends over a wide
range of visuomotor tasks, such as grasping, posting, and obstacle avoidance. Nevertheless, it is
entirely possible, given the damage to her dorsal stream, that a number of visuomotor
impairments still exist in DF (compared to neurologically-intact controls). In the following
paragraphs, we explore some of these visuomotor impairments and their relevance for the
proposed division of labour between vision-for-action and vision-for-perception.

3. DF' visuomotor difficulties in visual periphery
In two studies, Hesse and her colleagues (Hesse, Ball, & Schenk, 2012; 2014) tested DF’s
visuomotor performance across different visuomotor tasks. Unlike the majority of earlier studies,
however, which examined DF’s performance in central vision under natural, free-viewing
conditions, Hesse and her colleagues focussed on DF’s visuomotor performance in both central
and peripheral vision by constraining head and gaze position. Under these more challenging
conditions, the chances of revealing more subtle visuomotor impairments in DF were almost
certainly larger.
In the first of the two studies, Hesse et al. (2012) compared DF’s performance to that of
neurologically intact individuals in two different visuomotor tasks, target-directed pointing and
grasping. For the pointing task, participants were asked to point to a 2D virtual target that
appeared in different locations either under free viewing or under a condition in which gaze
direction was fixed. In addition, performance in these two conditions was compared in closed

9

and open loop, in which the target was either visible or concealed during the execution of the
meeting. For the grasping task, real 3D objects were attached to a computer screen in front of the
participant during a free viewing condition or in a condition in which fixation was restricted to a
peripheral location. The results showed that during pointing, DF’s performance in the freeviewing condition was within the normal range and did not differ from controls for the closedand open-loop conditions. A different pattern of performance was revealed when gaze was
restricted and pointing movements were made to peripheral targets. Under this condition, DF’s
performance was significantly worse than that of the control participants. A similar pattern of
results was found for the grasping task: in the free-viewing condition, the maximum grip
aperture (MGA) was significantly correlated with the target size for both DF and for the control
participants. This finding is entirely consistent with her performance in earlier studies and stands
in sharp contrast with her inability to perceive the size of objects (for reviews, see Goodale 2011,
2014). As was the case with pointing, however, DF’s performance in the restricted-viewing
condition in which she grasped the targets in her periphery was significantly worse than that of
the control participants. In a more recent study, Hesse et al. (2014) again examined DF’s
pointing ability in central and peripheral vision but in this case with both the right and the left
hand. As was found in the previous study, her pointing abilities were significantly worse in the
periphery than in central vision (compared to controls) – and this was true for both hands.
Overall, these findings show that DF has visuomotor difficulties when it comes to visual
periphery.
At the same time, it should be noted that, unlike DF’s pointing performance, which was
normal in the free-viewing condition, the slope relating the size of the MGA to the size of the
target object was smaller compared to that of the control participants in Hesse et al.’s (2012) free
viewing condition. To the best of our knowledge, this is the only example of difference in
sensitivity to object size during grasping between DF and neurologically-intact control
participants, among the many studies that have examined DF’s grasping behavior. Although such
a result can represent a statistical outlier related to multiple testing repetitions, it is also possible
that the size of the slope is not a good measure of sensitivity to object size during grasping.
Without additional measures of grasping precision, it is difficult to conclude anything about the
sensitivity of grip aperture to object size (Ganel et al., 2008). To illustrate, smaller slopes are
typically found during grasping compared to perceptual estimations of size (Franz et al., 2001).

10

Yet, despite this difference in the slopes, when response precision is taken into account, grasping
is consistently found to be more accurate than perceptual estimations of size (Ganel et al., 2008;
2012). In any case, it seems that DF’s ability to grasp objects and to point to targets is
essentially preserved in central vision, although she clearly has a deficit in both of these
visuomotor abilities in the visual periphery relative to control participants. As Hesse et al.
(2012) themselves conclude “DF's visuomotor deficits [in the visual periphery] do not contradict
claims by the perception-action model, but serve as evidence for a functional deficit in DF's
dorsal stream areas”.
Another, more recent, study has also compared DF’s visuomotor abilities in central and
peripheral vision with those of neurologically-intact control participants (Rossit et al., 2017). In
the first experiment, Rossit et al. again compared DF’s pointing performance with that of control
participants under a natural, free-viewing condition and under a fixated-viewing condition in
which gaze direction was constrained. The results again showed that DF’s pointing performance
in the free-viewing condition was intact, and did not differ than that of controls whereas her
performance was impaired for peripheral targets in the restricted viewing condition. In addition,
unlike the two earlier studies by Hesse et al. (2012, 2014), a perceptual “control” experiment was
included, in which the participants were asked to detect the onset of a target. It is important to
note that the perceptual task did provide an equivalent control condition for the pointing task, it
was used only in the restricted viewing condition task and was not matched with the pointing
task in terms of difficulty or task demands. In any case, DF’s perceptual target detection was
found to be impaired. Her reaction times were much slower compared to those of control
participants – even for targets presented in central vision, a finding which is at odds with her
normal pointing performance for the same targets. In short, DF continued to show a dissociation
between her visuomotor and perceptual abilities in central vision, even when gaze was
restrained.
Rossit et al. (2017) went on to test DF’s performance in a more elaborate task, the
“double step” reaching paradigm, in which the participant is required to respond to an abrupt
change in the location of the target when the reaching movement is initiated. DF showed
difficulties in responding to abrupt location changes during reaching. Although, in both
restrained and free-viewing conditions, she was able to complete the tasks successfully, her
performance was still significantly slower than that of control participants. Rossit et al.

11

suggested that DF’s difficulty on this task was a consequence of the bilateral thinning of her
parieto-occipital cortex (Bridge et al., 2013). This is indeed a possibility, although it is important
to note that her deficit is relatively mild compared to patients with optic ataxia, possibly because
DF’s parieto-occipital lesions are less extensive than those of the typical patient with optic
ataxia.

4. Double dissociations between DF and patients with optic ataxia

One common thread running through the Hesse et al. (2012; 2014) and Rossit et al. (2017)
studies is the finding that DF’s visuomotor performance across different tasks is intact in central
vision but is impaired in the periphery. There is little doubt that these findings deepen our
understanding of DF’s spared abilities and deficits in visuomotor control – and converge nicely
on a number of studies showing deficits in visuomotor control in the periphery following damage
to the parieto-occipital cortex (e.g., Karnath & Perenin, 2005). Yet, the Hesse, Rossit, and their
colleagues took things a step further by suggesting that the presence of dorsal-stream damage in
DF undercuts the double dissociation that Goodale and Milner (1992) originally made between
the pattern of deficits and spared abilities in her vs. patients with optic ataxia. Thus, Hesse et al.
(2012) claim that DF’s visuomotor deficits “invalidate earlier attempts to use the complimentary
[sic] nature of DF's behaviour and that of patients with optic ataxia to support the perceptionaction model” and Rossit et al. argue that “DF can no longer be considered as an appropriate
single-case model for testing the neural basis of perception and action dissociations”. But what is
forgotten in the rush to dismiss DF’s value in illuminating the dissociation between perception
and action is the fact that the comparison between her and patients with dorsal-stream damage
has been largely based on differences in their performance on perception and action tasks in
central vision. In the remaining paragraphs, we review some of these differences – and show
how the double dissociation that emerges provides powerful evidence for the division of labour
between the dorsal and ventral streams.
Goodale et al. (1994) tested DF and the optic ataxic patient RV on a task in which they
and a control participant were required to reach out and pick up irregularly shaped wooden
plaques in free viewing conditions. As described earlier, DF had no problem placing her index
finger and thumb on stable grasp points on the edges of these objects, even though she was

12

completely unable to tell them apart in a same-different task. In contrast, RV, who had dorsalstream damage, often chose quite inappropriate grasp points and fumbled the plaques when she
tried to pick them up, although – unlike DF – she was able tell them apart. A similar double
dissociation was observed between DF and RV’s performance when they were asked either to
estimate the width or to pick up rectangular blocks that differed in width (and length) but not
overall surface area: DF showed normal grip scaling but was extremely poor at estimating the
width of the objects, whereas RV showed no evidence of grip scaling, even though she could
reliably estimate their width (Goodale et al., 1991; Milner & Goodale, 2006). In these tasks, DF
and RV were allowed free vision and looked directly at the target objects using central vision.
Another patient (VK), who was diagnosed with optic ataxia following dorsal-stream damage,
also showed extremely poor grip scaling when she was asked to pick up rectangular blocks with
different widths under free viewing conditions (Jakobson et al., 1991) – again highlighting the
double dissociation between deficits in perception and action in DF and patients with dorsalstream damage. This double dissociation in central vision makes it clear that any visuomotor
deficits that DF might exhibit in the visual periphery do not undercut the pivotal role that she has
played in our understanding of the division of labour between the ventral and dorsal streams.
Another line of evidence for a sharp dissociation between DF’s performance and that of
patients diagnosed with optic ataxia after dorsal-stream damage comes from studies that have
examined the effect of delay on visuomotor control. Paradoxically, unlike normally sighted
individuals who show poorer performance in a grasping task when a delay is imposed between
the presentation of the target and movement initiation, optic ataxic patients actually improve
(Milner et al., 2001). Such delays have an opposite effect on DF’s performance: when she is
asked to grasp objects following a delay, her performance deteriorates precipitously (Goodale et
al., 1994a). It has been argued that the reason individuals with dorsal-stream damage do better in
delay whereas DF does so badly is that responding in delay depends on the retrieval of memories
about features of the target object that are subserved by perception-based processing in ventral
stream areas such as LO, which is intact in the dorsal-stream patients but severely compromised
in DF (Milner et al., 2001; but see Schenk & Hesse, 2017, for a different view). Again, this
highlights a strong double dissociation between DF and individuals who have damage in the
dorsal but not the ventral stream.

13

Finally, it should be noted that dissociations between DF’s visuomotor performance and
the performance of patients with classic optic ataxia have been reported even in peripheral
vision. Specifically, when reaching out to a distal goal, DF’s performance is essentially normal
in that she is able to effectively take into account the location of peripheral obstacles and
program her reaching movement accordingly (Rice et al., 2006). Conversely, patients with optic
ataxia fail to program their movements in order to avoid a potential contact with obstacles
presented in the periphery (Schindler et al., 2004). There should be some caution in interpreting
this difference, however, in that DF was not required to maintain fixation. But even so, when
looking at the goal object to which DF was directing her aiming movement, the obstacles would
be positioned well into the visual periphery.

5. Conclusion
Taken together, the results of the studies discussed above make it patently clear that DF and
individuals with ‘classic’ optic ataxia from large lesions of the dorsal stream show a strong
double dissociation between perceptual report and visuomotor performance in central vision. It
is true that DF shows some evidence of abnormal visuomotor performance in her visual
periphery, perhaps reflecting the bilateral thinning of cortex in the parieto-occipital region. In
fact, one of the deficits DF demonstrates, namely selecting a comfortable hand posture during
grasping, is an important new finding that highlights the possible role of the pIPS and pSPL in
the selection of biomechanically appropriate hand postures (Wood et al., 2016). Nevertheless,
she is clearly not as impaired as individuals with optic ataxia who, unlike DF, show massive
deficits in grasping in central vision and obstacle avoidance in the visual periphery. Indeed, the
fact that DF shares some symptoms with such patients does not mean that she has full-blown
optic ataxia any more than the fact that she shares some symptoms with blind people (being
unable to explicitly report the shape and orientation of objects) does not mean that she is blind.
Instead, the accumulating behavioral and neuroanatomical evidence supports what first caught
the attention of Goodale and Milner in 1992 that DF's most remarkable characteristic is not her
rare difficulties in aspects of visuomotor control, but the strong dissociation between DF’s
massive deficit in her ability to perceive the shape and orientation of objects and her normal, or
close-to-normal, visually control of actions.

14

Video caption
match.m4v: A short video of DF’s performance on the perceptual matching task. DF was
asked to match the orientation of the card to that of the target slot without moving her hand in its
direction. The video was taken in 1990
post.m4v: A short video of DF’s performance on the posting task. DF was asked to post
the card into the target slot. The video was taken in 1990.

References
Andersen, R.A., Andersen, K.N., Hwang, E.J., & Hauschild, M. (2014). Optic ataxia: from
Balint's syndrome to the parietal reach region. Neuron 81(5), 967-83.
Bridge, H., Thomas, O.M., Minini, L., Cavina-Pratesi, C., Milner, A.D., & Parker, A.J.
(2013). Structural and functional changes across the visual cortex of a patient with visual form
agnosia. The Journal of Neuroscience: the Official Journal of the Society for Neuroscience,
33(31), 12779e12791.
Campion, J. (1987). Apperceptive agnosia: The specification and description of constructs.
In Humphreys G.W. & Riddoch M.J. (Eds.), Visual object processing: A cognitive
europsychological approach (pp. 197–232). London: Erlbaum.
Carey, D.P., Harvey, M., & Milner, A.D. (1996). Visuomotor sensitivity for shape and
orientation in a patient with visual form agnosia. Neuropsychologia 34(5), 329-337.
Chen, J., Sperandio, I., & Goodale, M.A. (2015). Differences in the effects of crowding on
size perception and grip scaling in densely cluttered 3-D scenes. Psychological science, 26(1),
58-69.
Franz, V.H., Fahle, M., Bulthoff, H.H. & Gegenfurtner, K.R. (2001). Effects of visual
illusions on grasping. Journal of Experimental Psychology-Human Perception and Performance
27, 1124-1144, doi:10.1037/0096-1523.27.5.1124.
Freud, E., Ganel, T., Avidan, G., & Gilaie-Dotan, S. (2016). Functional dissociation
between action and perception of object shape in developmental visual object agnosia. Cortex,
76, 17-27.

15

Gallivan, J.P., & Goodale, M.A. (2017). The dorsal ‘action’ pathway. In: The Parietal
Lobes. Neurological and Neuropsychological Deficits (eds. G. Vallar, H.B. Coslett). Elsevier.
Ganel, T., Chajut, E., & Algom, D. (2008). Visual coding for action violates fundamental
psychophysical principles. Current Biology 18(14), R599-R601.
Ganel, T., Freud, E., Chajut, E., & Algom, D. (2012). Accurate visuomotor control below
the perceptual threshold of size discrimination. PLoS One 7(4), e36253.
Ganel, T., & Goodale, M.A. (2003). Visual control of action but not perception requires
analytical processing of object shape. Nature 426(6967), 664.
Goodale M.A., & Ganel T. (2015). Different modes of visual organization for perception
and for action. Oxford Handbook of Perceptual Organization 3(1), 1–19.
Goodale, M.A., Jakobson, L.S., & Keillor, J.M. (1994a). Differences in the visual control
of pantomimed and natural grasping movements. Neuropsychologia. 32(10), 1159-1178.
Goodale, M.A., Meenan, J.P., Bülthoff, H.H., Nicolle, D.A., Murphy, K.J., & Racicot, C.I.
(1994b). Separate neural pathways for the visual analysis of object shape in perception and
prehension. Current Biology 4(7), 604-610.
Goodale, M.A., Milner, A.D., Jakobson, L.S., & Carey, D.P. (1991). A neurological
dissociation between perceiving objects and grasping them. Nature, 349(6305), 154-156.
Hesse, C., Ball, K., & Schenk, T. (2012). Visuomotor performance based on peripheral
vision is impaired in the visual form agnostic patient DF. Neuropsychologia, 50(1), 90-97.
Hesse, C., Ball, K., & Schenk, T. (2014). Pointing in visual periphery: is DF's dorsal stream
intact? PloS one, 9(3), e91420.
Himmelbach, M., Boehme, R., & Karnath, H.O. (2012). 20 years later: a second look on
DF's motor behaviour. Neuropsychologia, 50(1), 139-144.
Jakobson, L.S., Archibald, Y.M., Carey, D.P., & Goodale M.A. (1991). A kinematic
analysis of reaching and grasping movements in a patient recovering from optic ataxia.
Neuropsychologia, 29(8), 803-809.
James, T.W., Culham, J., Humphrey, G.K., Milner, A.D., & Goodale, M.A. (2003). Ventral
occipital lesions impair object recognition but not object directed grasping: A fMRI study. Brain,
126, 2463–2475.
Karnath, H.O., & Perenin, M.T. (2005). Cortical control of visually guided reaching:
evidence from patients with optic ataxia. Cerebral cortex, 15(10), 1561-1569.

16

Karnath, H.O., Rüter. J., Mandler, A., & Himmelbach, M. (2009). The anatomy of object
recognition – Visual form agnosia caused by medial occipitotemporal stroke. Journal of
Neuroscience, 29, 5854–5862.
Kravitz, D.J., Saleem, K.S., Baker, C.I., & Mishkin, M. (2011). A new neural framework
for visuospatial processing. Nature Reviews Neuroscience 12(4), 217-230.
Kravitz, D.J., Saleem, K.S., Baker, C.I., Ungerleider, L.G., & Mishkin, M. (2013). The
ventral visual pathway: an expanded neural framework for the processing of object quality.
Trends in Cognitive Sciences,17(1), 26-49.
McIntosh, R.D., Mulroue, A., Blangero, A., Pisella, L., & Rossetti, Y. (2011). Correlated
deficits of perception and action in optic ataxia. Neuropsychologia, 49(1), 131-137.
Milner, A.D., Dijkerman, H. C., Pisella, L., McIntosh, R. D., Tilikete, C., Vighetto, A., &
Rossetti, Y. (2001). Grasping the past: Delay can improve visuomotor performance. Current
Biology, 11(23), 1896-1901.
Milner, A.D., Ganel, T., & Goodale, M.A. (2012). Does grasping in patient D.F. depend on
vision? Trends in Cognitive Sciences, 16, 256-257.
Namdar, G., Algom, D., & Ganel, T. (2016). Dissociable effects of stimulus range on
perception and action. Cortex.
Patla, A., Goodale, M.A. (1997). Visuomotor transformation required for obstacle
avoidance during locomotion is unaffected in a patient with visual form agnosia. NeuroReport
8(1), 165-168.
Perenin, M.T., & Vighetto, A. (1988). Optic ataxia: a specific disruption in visuomotor
mechanisms. I. Different aspects of the deficit in reaching for objects. Brain 111(3), 643-674.
Prockop, L.D., & Chichkova, R.I. (2007). Carbon monoxide intoxication: an updated
review. Journal of the Neurological Sciences 262(1), 122-130.
Rice, N. J., McIntosh, R. D., Schindler, I., Mon-Williams, M., Demonet, J. F., & Milner, A.
D. (2006). Intact automatic avoidance of obstacles in patients with visual form agnosia.
Experimental Brain Research, 174(1), 176.
Rossit, S., Harvey, M., Butler, S. H., Szymanek, L., Morand, S., Monaco, S., & McIntosh,
R. D. (2017). Impaired peripheral reaching and on-line corrections in patient DF: optic ataxia
with visual form agnosia. Cortex.

17

Schenk, T. (2012). No dissociation between perception and action in patient DF when
haptic feedback is withdrawn. The Journal of Neuroscience. 32(6), 2013-2017.
Schenk, T., & Hesse, C. (2017). Do we have distinct systems for immediate and delayed
actions? A selective review on the role of visual memory in action, Cortex.
Schindler, I., Rice, N.J., McIntosh, R.D., Rossetti, Y., Vighetto, A., & Milner, A.D. (2004).
Automatic avoidance of obstacles is a dorsal stream function: evidence from optic ataxia. Nature
neuroscience, 7(7), 779.
Whitwell, R.L., Ganel T., Byrne C.M., & Goodale M.A. (2015). Real-time vision, tactile
cues, and visual form agnosia: removing haptic feedback from a "natural" grasping task induces
pantomime-like grasps. Frontiers in Hum. Neuroscience, 9:216.
Whitwell, R.L, Milner, A.D., & Goodale, M.A. (2014a). The two visual systems
hypothesis: new challenges and insights from visual form agnosic patient DF. Front. Neurol. 5,
255.
Whitwell, R.L., Milner, A.D., Cavina-Pratesi, C., Byrne, C.M., & Goodale, M.A. (2014b).
DF’s visual brain in action: the role of tactile cues. Neuropsychologia 55, 41–50.
Wood, D.K, Chouinard, P.A., Major, A.J., & Goodale, M.A. (2016). Sensitivity to
biomechanical limitations during postural decision-making depends on the integrity of posterior
superior parietal cortex, Cortex