382

J Neurol Neurosurg Psychiatry 1998;65:382–385

SHORT REPORT

Separate visual pathways for perception of actions
and objects: evidence from a case of apperceptive
agnosia
Carla Teixeira Ferreira, Mathieu Ceccaldi, Bernard Giusiano, Michel Poncet

Abstract
Recognition of diVerent kinds of visual
stimuli was studied in a patient who
acquired apperceptive visual agnosia after
a bilateral occipitotemporal lesion which
partially spared the primary visual cortex. Impairment in recognising static
objects perceived visually sharply contrasts with the relatively well preserved
ability to recognise objects from gestures
illustrating their use, and to recognise
actions shown in line drawings. It is
suggested that the occipitoparieto-frontal
pathway is involved in the recognition of
actions, and in the recognition of objects
when sensorimotor experience is evoked.
(J Neurol Neurosurg Psychiatry 1998;65:382–385)
Keywords: recognition; parietal lobe; frontal lobe

Laboratoire de
Neurophysiologie et
Neuropsychologie,
Upres EA 2200 CJF
Inserm 9706, Faculté
de Médecine de
Marseille, Université
de la Méditerranée,
France
C T Ferreira
M Ceccaldi
B Giusiano
M Poncet
Department of
Neurology,
Universidade Federal
do Rio de Janeiro,
Brasil
C T Ferreira
Correspondence to:
Dr Carla Teixeira Ferreira,
Rua Bartolomeu Portela,
50/505, Botafogo CEP:
22.290.190, Rio de Janeiro,
Brasil. Fax 55.21.542.5071
Received 20 March 1997 and
in final revised form 5
February 1998
Accepted 16 February 1998

Visual agnosia is the inability to recognise visual
stimuli despite adequate elementary visual
function. Visually agnosic patients are unable to
produce gestures of object use or to show any
recognition of the objects they fail to name.
According to Lissauer,1 visual agnosias can be
classified as apperceptive, in which case the person never achieves a complete conscious percept
of the stimulus, or associative, in which such an
adequate percept is stripped of meaning.
Several studies2–5 have shown that the recognition of a visually presented static object is
mediated by a hierarchically organised occipitotemporal pathway that includes distinct
regions for processing the visual features of
objects. By contrast, recent work suggests that
the occipitoparietofrontal stream might be
involved not only in the spatial localisation of
visual stimuli,2 visually guided movements,6
particularly grasping,7 movement perception,8
and visuospatial working memory,9 10 but also
in the recognition of goal oriented actions.11–14
The present study considers a case of apperceptive visual agnosia. It focuses on performance diVerences between recognition of actions
represented by line drawings and recognition of
objects presented visually. In the light of recent
studies,14–17 it is hypothesised that the anatomical structures involved in the acquisition of
knowledge about a concept are essential to its

subsequent recognition. Therefore, the recognition of an object from a static visual presentation of it may essentially involve the occipitoinferotemporal pathway. By contrast,
because actions are learned primarily through
sensorimotor experience, the recognition of an
action represented in a photograph, or the recognition of an object from a visual presentation
of a gesture illustrating its use, may chiefly
involve the occipitoparietofrontal stream.

Case report
The patient was a right handed 65 year old
retired male engineer. He had a left occipitotemporal haemorrhage in 1981 and a right
occipitotemporal haemorrhage in 1991 (fig 1).
His general neurological examination was
normal. Neuropsychological assessment disclosed a verbal intelligence quotient of 150
(WAIS), a memory quotient of 85 (Wechsler
R), and intact speech (BDAE). He was able to
write, but he was forced to use a letter by letter
strategy to read. He had complete achromatopsia (Ishiara test), and impaired visual recognition of objects and faces. He exhibited bilateral
quadrantanopia (right inferior and left superior
quadrants). P100 visual evoked potentials were
spared. Visual acuity was normal in the
preserved visual fields. His pupillary responses
were normal and his voluntary eye movements
were complete.
Despite the fact that he could describe a
visually perceived object in detail, he failed to
relate individual elements to the whole, and
consequently, he was not able to identify it. He
could produce accurate copies of line drawings
of objects by implementing a piecemeal, line by
line copying strategy that was extremely slow
(for example, he took eight minutes to copy a
drawing of a comb). He was unable to match
objects seen from diVerent views (13/23). In an
object decision task in which he was required to
determine whether line drawings corresponded
to real objects (Snodgrass and Vanderwart
set18) or to non-objects (Kroll and Potter set19),
his performance was deficient (31/40), indicating that his ability to access structural object
knowledge through visual presentation was
impaired. The utilisation of silhouettes of these
same stimuli did not improve performance

383

Separate visual pathways for perception of actions and objects

(26/40). His score on Benton’s judgment of
line orientation test was normal (23/30).
EXPERIMENTAL INVESTIGATION

To investigate the patient’s visual object recognition deficit, we compared his object recognition performance in diVerent modes including
visual presentation of a static real object, visual
presentation of a real object being moved in a
non-specific way by the examiner (rotation and
horizontal displacement), tactile presentation,
verbal definition, examiner pantomime of how
the object might be used, and examiner utilisa-

Figure 1

tion of the object. The same set of 30 real
objects was used for all tasks. As the table
shows, visually presented object recognition
was impaired. By contrast, recognising objects
on tactile presentation (McNemar symmetry
÷2=19.0, p<0.0001), verbal definition (McNemar symmetry ÷2=21.0, p<0.0001), pantomimes of object use (McNemar symmetry
÷2=14.2, p<0.001), and utilisation of real
objects (McNemar symmetry ÷2=17.0,
p<0.0001) were significantly better. These
results confirmed that he was particularly

Representative horizontal MRI cross section of the patient.

384

Ferreira, Ceccaldi, Giusiano, et al
Table 1

The patient’s results on naming tests

Type of presentation

Stimuli (n)

Correct responses
(n (%))

Visual, real objects:
Static
Non-specific movement
Tactile: real objects
Verbal
Pantomime of use
Object use
Line drawings of objects
Line drawings of actions

30
30
30
30
30
30
122
42

7 (23)
3 (10)
26 (87)
28 (93)
23 (77)
24 (80)
3 (2.5)
24 (57)

impaired in object recognition on visual
presentation.
Two experiments were conducted to verify
the potential dissociation between object and
action naming on static visual presentation.
The first experiment required recognition of
line drawings of objects (n=122) taken from
the Snodgrass and Vanderwart set.18 The
drawings were presented on a screen (25×19
cm) connected to a microcomputer. Each
drawing was displayed in the middle of the
screen and remained there until a response was
given, after which it disappeared. The screen
was blank for 10 seconds between presentations. The stimuli included tools, fruits,
vegetables, and animals.
The second experiment involved retrieving
an action verb when looking at a line drawing
depicting that action (n=42). The line drawings of the actions were presented under the
same conditions as the line drawings of the
objects. Each drawing was obtained by modifying a black and white photograph using Adobe
Photoshop™ 2.5.1 software. The final drawing
essentially retained the contour of the picture
and a few details (example in fig 2).
As shown in the table, the patient’s ability to
name actions when looking at the corresponding line drawings was markedly diVerent from
his ability to name line drawings of objects
taken from the Snodgrass and Vanderwart set
(÷2 = 67.9, p<0.001). These findings indicate
that the severity of this patient’s visual perception deficit depends on the type of stimulus
being recognised.

Figure 2 Example of an action line drawing correctly
identified by the patient.

The line drawings of objects from the
Snodgrass and Vanderwart set and the black
and white photographs corresponding to an
action were presented to a group of 17 normal
subjects matched by age and educational level
to the patient. The experimental conditions
were the same as those used for the patient.
Vocal response time was directly recorded by
the computer with a microphone linked to it.
For this group, the vocal response time
recorded for verb evocation (t=1447 ms (SD
1008)) was significantly longer than it was for
object name evocation (t=1244 ms (SD 868))
(repeated measurement analysis of variance
(ANOVA) F(1,96) = 8.0, p<0.01). This
suggests that, for the control group, verb
evocation was not easier than object name evocation. Nevertheless, with such a task, we cannot determine which level(s) of visual information processing and/or name evocation is
(are) responsible for the significant diVerences
obtained in the response time of the control
group. Note also that even when the patient’s
ability to recognise a line drawing of an action
was compared with that of a real object, action
recognition was significantly better (÷2=8.2,
p<0.01).
Discussion
The patient had apperceptive visual agnosia
after a bilateral occipitotemporal lesion which
partially spared the primary visual cortex (fig
1). He exhibited a deficit in recognising real
objects presented visually and object line
drawings. This deficit is accompanied by a
visual perception disorder characterised by a
piecemeal, line by line copying strategy, and
impairment in matching objects from diVerent
views and performing an object decision task.
By contrast, he was able to name objects on
tactile presentation and verbal definitions,
showing that the deficit solely concerned visual
input. Moreover, he was able to recognise
actions shown in line drawings, in addition to
objects from pantomimes depicting how they
are used (table).
This case is of particular interest because the
patient was able to recognise actions presented
in line drawings and objects the use of which is
pantomimed by the examiner. His ability to
recognise action drawings should be a consequence of the extracontextual information
contained in this type of stimulus compared
with object drawings. In a case study of a
patient with apperceptive visual agnosia, Riddoch and Humpreys (1987)20 showed that their
patient was better at identifying objects when
paired with an appropriate context than when
shown in isolation. Such a finding suggests that
semantic information can be used to decide
between alternative interpretations of visual
input. Nevertheless, when the scene context
was given along with two visually similar
objects, the results did not diVer from those of
isolated object identification. This suggests
that visual object identification is sensitive to
the eVects of context, but, as those authors
claim, “contextual input does not override an
early deficit in the integration of visual form”.
Therefore, extracontextual input of action line

Separate visual pathways for perception of actions and objects

drawings might facilitate recognition of the
action, but it remains undetermined whether
context is the main cause of the patient’s good
performance on action drawing recognition.
An alternative way of interpreting the
patient’s preserved ability to recognise actions
presented in line drawings and objects the use
of which is pantomimed by the examiner, and
the patient’s impaired ability to recognise visually presented objects, would be that the visual
pathways required to perceive actions are
diVerent from those required to perceive
objects. Over the past 20 years, many diVerent
visual areas have been described. In 1982,
Mishkin and Ungerleider2 distinguished two
visual cortical pathways. The occipitotemporal
stream was shown to be mostly involved in
object identification, whereas the occipitoparietal pathway was shown to be implicated in the
spatial localisation of visual stimuli. Later, the
occipitoparietofrontal pathway was found to be
involved in visually guided movements6 and
especially in grasping,7 movement perception,8
visuospatial working memory,9 10 and recognition of a goal oriented action.11–14 The present
findings suggest the involvement of the occipitoparietofrontal pathway in the visual perception of hand and body position and movement.
It was recently shown in monkeys that a
group of F5 neurons (region considered to be
the homologue of the caudal part of the human
inferior frontal gyrus21) responds selectively to
the sight of actions carried out by others.13 It is
important to note that the main cortical input
of F5 comes from anterior intraparietal area
7b,22 where a neuron coding action has been
reported.23 Frontal neurons selectively responding to the sight of actions were also
described in humans by Fadiga et al11 and Rizzolatti et al.12 These authors suggest the
existence of a system that matches action
observation and execution, and suspect that
the motor system is not devoted solely to the
production of movements but is also involved
in their recognition. Therefore, we speculate
that the patient was able to identify objects
from a gesture illustrating their use because this
type of visual information is processed in the
occipitoparietofrontal stream.
Studies on patients with ideomotor apraxia
after parietal lesion support this hypothesis.24
Such patients, who are unable to make
symbolic gestures or act out the use of an
object on verbal request or imitation, are
sometimes also incapable of recognising a correctly produced gesture given a stationary
(photograph) or moving visual presentation.
Other studies25 have shown that these patients
are also unable to use the mental reconstruction of a gestural image to predict the amount
of time it would take to make finger movements
and visually guided pointing gestures, compared with normal subjects and patients with a
lesion in the primary motor area. These
findings thus suggest that the neural network
involved in gestural programming is also
involved in gesture recognition and mental
reconstruction.
The results obtained in the present study, in
which the patient’s impairment in visually rec-

385

ognising static objects contrasts with his
preserved ability to recognise objects from gestures illustrating their use and to recognise
actions shown in line drawings, suggest that the
occipitoparietofrontal pathway is involved in
conscious visual perception and in the
interpretation of goal oriented actions, even
when shown in a static way.
We are very grateful to the patient and his wife, who have always
made the testing a pleasure. Also we extend our thanks to Yves
Tocquer for his technical assistance. This work was supported
by a grant to CTF from the “Institut de la Maladie
d’Alzheimer” in France. A subset of these data was presented at
the XXVI International Congress of Psychology, Montreal,
August 1996.

1 Lissauer H. Ein fall von seelenblindheit nebst einem
Beitrage zur Theori derselben. (A case of visual agnosia
with a contribution theory.) Archiv für Psychiatrie und Nervenkrankheiten 1890; 21:222–70.
2 Ungerleider LG, Mishkin M. Two cortical visual systems.
In:Ingle DJ, Goodale MA, Mansfield RJW, eds. Analysis of
visual behavior. Cambridge, MA: MIT Press 1982:549–86.
3 Fujita I, Tanaka K, Ito M, et al. Columns for visual features
of objects in monkey inferotemporal cortex. Nature
1992;360:343–6.
4 Gross CG. Representation of visual stimuli in inferior temporal cortex. Philos Trans R Soc London B Biol Sci
1992;335:3–10.
5 Sergent J, Ohta B, McDonald B. Functional neuroanatomy
of face and object processing: a positron emission tomography study. Brain 1992;115:15–36.
6 Goodale MA, Milner AD. Separate visual pathways for perception and action. Trends Neurosci 1992;15:20–5.
7 Jeannerod M, Arbib MA, Rizzolatti G, et al. Grasping
objects: the cortical mechanisms of visuomotor transformation. Trends Neurosci 1995;18:314–20.
8 Zeki S, Watson JDG, Lueck CJ, et al. A direct demonstration
of functional specialization in human visual cortex. J Neurosci 1991;1:641–9.
9 Smith EE, Jonides J, Koeppe RA, et al. Spatial versus object
working memory: PET investigations. Journal of Cognitive
Neuroscience 1995;7:337–56.
10 Teixeira Ferreira C, Vérin M, Pillon B, et al. Spatio-temporal
working memory and frontal lesions in man. Cortex
1998;34:83–98.
11 Fadiga L, Fogassi L, Pavesi G, et al. Motor facilitation during action observation: a magnetic stimulation study. J
Neurophysiol 1995;73:2608–11.
12 Rizzolatti G, Fadiga L, Matelli M, et al. Localization of grasp
representations in humans by PET:1. Observation versus
execution. Exp Brain Res 1996;111:246–52.
13 Gallese V, Fadiga L, Fogassi L, et al. Action recognition in
the premotor cortex. Brain 1996;119:593–609.
14 Teixeira Ferreira C, Giusiano B, Ceccaldi M, et al. Optic
aphasia: evidence of the contribution of diVerent neural
systems to object and action naming. Cortex 1997;33:499–
513.
15 Martin A, Haxby JV, Lalonde FM, et al. Discrete cortical
regions associated with knowledge of color and knowledge
of action. Science 1995;270:102–5.
16 Daniele A, Giustolisi L, Silveri MC, et al. Evidence for a
possible neuroanatomical basis for a lexical processing of
nouns and verbs. Neuropsychologia 1994;32:1325–41.
17 Warburton E, Wise RJS, Price CJ, et al. Noun and verb
retrieval by normal subjects:Studies with PET. Brain 1996;
119:159–79.
18 Snodgrass JG, Vanderwart M. A standardized set of 260
pictures: norms for naming agreement, image agreement,
familiarity and visual complexity. Journal of Experimental
Psychology: Human Learning and Memory 1980;6:174–215.
19 Kroll JF, Potter MC. Recognizing words, pictures and
concepts: a comparision of lexical, object, and reality decisions. Journal of Verbal Learning and Verbal Behavior
1984;23:39–66.
20 Riddoch MJ, Humphreys GW. A case of integrative visual
agnosia. Brain 1987;110:1431–62.
21 Petrides M, Pandya DN. Comparative architectonic analysis
of the human and the macaque frontal cortex. In: Boller F,
Spinnler H, Hendler JA, eds. Handbook of neuropsychology.
Vol 9. Amsterdam: Elsevier 1994:17–58.
22 Cavada C, Goldman-Rakic PS. Posterior parietal cortex in
rhesus monkey: II. Evidence for segregated corticocortical
networks linking sensory and limbic areas with the frontal
lobe. J Comp Neurol 1989;287:422–45.
23 Leinonen L, Hyvärinen J, Nyman G, et al. I. Functional
properties of neurons in lateral part of associative area 7 in
awake monkeys. Exp Brain Res 1979;34:299–320.
24 Heilman KM, Rothi LJ, Valenstein E. Two forms of
ideomotor apraxia. Neurology 1982;32:342–6.
25 Sirigu A, Duhamel JR, Cohen L, et al. The mental
representation of hand movements after parietal cortex
damage. Science 1996;273:1564–8.