1 Sensory Agnosias - Neurobiology of Sensation and Reward Agnosia - NCBI Bookshelf

14/08/13 Sensory Agnosias - Neurobiology of Sensation and Reward - NCBI Bookshelf
NCBI Book sh elf. A ser v ice of t h e Na t ion a l Libr a r y of Medicin e, Na t ion a l In st it u t es of Hea lt h .
Gottfried JA, editor. Neurobiology of Sensation and Reward. Boca Raton (FL): CRC Press; 201 1 .
Chapter 10 Sensory Agnosias

H. Branch Coslett.
10.1 INTRODUCTION AND HISTORICAL OVERVIEW

Sensory agnosias are relativ ely uncommon clinical sy ndromes characterized by a failure of recognition that cannot be
attributed to the loss of primary sensory function, inattentiv eness, general mental impairment, or lack of familiarity with
the stimulus (Fredericks 1 969; Bauer 2003). In other words, sensory agnosias are disorders that are bracketed by failures of
early sensory processing on the input side and inability to attend to or comprehend the output of high-lev el sensory
processing on the output side. For ex ample, a blind subject who fails to recognize a rose or a deaf subject who does not
recognize the sound of a hammer driv ing a nail are not agnosic; similarly , a demented person who no longer knows what a
hammer is would not be considered agnosic if he failed to recognize the object.
Although the term “agnosia” was coined by Freud (1 891 ) in his discussion of aphasia and related disorders, descriptions of
the disorder predate Freud. Finkelnburg (1 87 0) described the sy ndrome of “asy mbolia” in which a percept failed to contact
the knowledge relev ant to that percept and Hughlings Jackson (1 87 6) described the phenomenon of “imperception” in the
contex t of a patient with a large tumor of the posterior portion of the right hemisphere who was unable to nav igate in
familiar places and did not recognize people or places. He postulated that the posterior portion of the right hemisphere was
crucial for v isual memory and recognition. Munk (1 881 ) prov ided an early and influential description of dogs with parieto-
occipital lesions that were able to nav igate about their surroundings without bumping into objects or getting lost y et didn’t
recognize objects of obv ious relev ance such as food; he termed this disorder “Seelenblindheit” or mindblindness.
Interestingly , similar findings were subsequently reported in macaques after lesions of the occipital lobe and temporal
projections (Horel and Keating 1 97 2).
An important early theoretical contribution was made by Lissauer (1 890), who distinguished between two forms of the
disorder: “apperceptiv e” and “associativ e” agnosias. As this terminology continues to be widely employ ed in the clinical
literature, it will be briefly rev iewed. For Lissauer “apperception” referred to the “stage of conscious awareness of a sensory
impression” (translated into English by Jackson [1 932]). Apperceptiv e agnosias, in Lissauer’s scheme, represented a
disorder in which the early v isual processing of a stimulus is disrupted, with a resulting failure to generate a fully specified
perceptual representation. Associativ e agnosias, in contrast, were characterized by preserv ed ability to compute a
representation of a v isual stimulus but an inability to recognize the object, as indicated by the ability to name the object or
produce v erbal or non-v erbal information that would unambiguously identify the stimulus. Presaging a number of
contemporary accounts of the processes inv olv ed in recognition (e.g., Damasio 1 989; Simmons and Barsalou 2003; Farah
2000; Humphrey s, Riddoch, and Fortt 2006), for Lissauer recognition entailed the simultaneous activ ation of multiple
attributes (e.g., sound, touch, smell, etc.) that were linked to the v isual form. That is, for Lissauer as for a number of
contemporary theorists, “recognition” of a telephone entailed the contemporaneous activ ation of the v isual image, sound,
heft, tex ture, manner of use, and function of a telephone (Allport 1 985).
Although the boundaries between agnosia and primary v isual loss on the input side and disorders of semantics at a higher
lev el were hazy in Lissauer’s account (and remain so today ), he offered a clear distinction between apperceptiv e and
associativ e v isual agnosias: the ability to copy a v isual stimulus. Reflecting their inability to generate an adequate sensory
representation of the stimulus, apperceptiv e agnosics are unable to copy a stimulus whereas associativ e agnosics are able to
copy a figure but remain unable to recognize what they hav e copied (Rubens and Benson 1 97 1 ). As has been emphasized by
Farah (2004) in her authoritativ e account of the v isual agnosias, the reproductions of associativ e agnosics may be
ex tremely detailed but appear to be slav ish copies of the stimulus, uninformed by stored knowledge of the stimulus. As
indicated in Figure 1 0.1 , when asked to copy a drawing that has been deliberately distorted, associativ e agnosics may
include the distortion in their drawing without appearing to be aware of the error.
Unlike other classical neurologic sy ndromes such as aphasia or neglect, the status of the concept of agnosia has v aried
substantially ov er the century since its description. In the early portion of the twentieth century , the discussion of the
disorder was largely framed by the prev ailing Gestalt theory of psy chology ; Poppelreuter (1 923) and Goldstein (Goldstein
1 943; Goldstein and Gelb 1 91 8), for ex ample, interpreted their subjects’ behav iors with respect to such concepts such as
“closure” and “inv ariance.”
The v ery ex istence of the phenomenon of v isual agnosia was questioned by a number of inv estigators. This was ex pressed
by Pav lov (1 927 ), for ex ample, who, in response to Munk’s phrase “the dog sees but does not understand,” countered that
“the dog understands but does not see well enough.” In later y ears Bay (1 953) and Bender and Feldman (1 97 2) argued that
apparent v isual agnosias were attributable to perceptual impairment (e.g., “tunnel v ision”), general cognitiv e impairment,
or some combination thereof.
In recent y ears, there has been a resurgence of interest in the topic for sev eral reasons. One is the dev elopment of more
sophisticated and nuanced models of “recognition” according to which perception is taken to be a multi-stage interactiv e
process; on such accounts, the distinction between perceptual disorders and agnosia is seen not as dichotomous but rather
as a process in which sensory inputs giv e rise to progressiv ely more elaborated representations in which different ty pes of
information (e.g., shape, color, location) may be emphasized (Heinke and Humphrey s 2003; Ellis and Y oung 1 988). Second,
in light of the increasingly complex and interactiv e models of recognition, the heuristic v alue of data from patients with
agnosia has prov en to be substantial. As illustrated by the influential contributions of Farah (2004), Riddoch and
www.ncbi.nlm.nih.gov/books/NBK92800/?report=printable 1/17
Humphrey s (1 987 ), and others, data from agnosic subjects may serv e to indicate the fault lines in the process of recognition
and offer important constraints for accounts based on animal work, modeling, and studies of normal subjects.
10.2 TYPES OF AGNOSIA

As sensory agnosia is defined as a modality -specific disorder of recognition, the sy ndrome may be encountered in any
sensory channel that permits entities to be identified. Reflecting the central role of v ision in humans as well as the fact that
v ision has been studied more ex tensiv ely than other sensory modalities, most work has focused on v isual agnosia. Tactile,
auditory , and ev en olfactory and gustatory agnosias hav e also been described. In this chapter we focus on different ty pes of
v isual agnosia and present a theoretical framework for understanding them before discussing other ty pes of agnosia.
10.2.1 VISUAL AGNOSIA

Most studies of v isual agnosia hav e emphasized the recognition of man-made objects. More recently , howev er,
inv estigations of agnosic subjects hav e demonstrated remarkable specificity with respect to the ty pes of stimuli with which
subjects may be impaired. Agnosias for objects, faces (“prosopagnosia”), words (“pure word blindness”), colors, and the
env ironment (including landmarks) hav e all been described. Finally , the disorder of simultanagnosia—an inability to “see”
more than one object at a time—is often regarded as a ty pe of v isual agnosia. Consistent with these behav ioral dissociations,
functional imaging in normal subjects demonstrates that different parts of the brain may , at least to some degree, be
optimized for the processing of different classes of v isual stimuli. We briefly discuss these ty pes of v isual agnosia in turn.
Before considering the specific disorders of higher-lev el v isual processing, howev er, we present an information-processing
account of v isual processing. A detailed account of v isual processing is bey ond the scope of this chapter but can be found in
Chapter 8; the model described briefly below is intended only to prov ide a framework within which the complex and rich
clinical literature can be understood.
10.2.1.1 Visual Object Recognition: A T heoretical Ov erv iew

The most important insight regarding v isual processing during the last century has been the recognition that different ty pes
of v isual information are segregated at the retina and that different ty pes of information are processed in parallel. Although
there are numerous and strong interactions between the processing streams, the segregation persists until late in v isual
processing, as indicated by the fundamental distinction between the v entral “what” stream and the dorsal “where”
(Ungerleider and Mishkin 1 982) or “how to” (Milner and Goodale 1 995) streams. A cartoon depicting the basic architecture
of the v isual processing sy stem is presented in Figure 1 0.2.
Phy siologic, anatomic and, more recently , imaging studies hav e demonstrated conclusiv ely that different v isual attributes
such as color, angle, motion, depth, orientation, and length are processed in parallel in different brain regions. Low-lev el
v isual routines such as “boundary marking” (e.g., Ullman 1 984; Borenstein and Ullman 2008) serv e to group different
stimuli leading to v isual forms. Although these processes were intensiv ely inv estigated by Gestalt psy chologists, the neural
bases of the routines remain poorly understood. These processes, including grouping of v isual features into candidate
“objects” and marking the boundary of candidate objects, appear to be crucial for object recognition. These processes do
not require attention or effort and are assumed to be “automatic.”
Although there is conv incing ev idence that v isual attributes are processed in parallel, the v isual env ironment consists of
stimuli or regions of space in which the attributes are integrated; that is, in order to generate an interpretable, coherent
picture of the env ironment, information that is processed in parallel in different brain regions must be integrated so that, for
ex ample, the color red and ov al shape are linked to generate the percept of an apple. Abundant ex perimental ev idence
demonstrates that this “binding” of v isual feature information is mediated by a limited-capacity operation ty pically referred
to as “v isual attention” (the neural mechanisms of which are described in detail in Chapter 8). The binding function of v isual
attention is illustrated by the phenomenon of “illusory conjunctions.” Treisman (Treisman and Gelade 1 980; Treisman and
Souther 1 985) and others (Cav e 1 999; Prinzmetal et al. 2002) hav e demonstrated that when v isual attention is
“ov erloaded,” v isual attributes can miscombine. For ex ample, when presented with an array containing red X’s and green
T’s for 200 ms, normal subjects may report seeing a red T despite the fact that no such stimulus was present. These and a
host of similar findings suggest that a limited-capacity , relativ ely fast but not infallible, “glue” links v isual feature
information computed in parallel. This glue is v isual attention. As suggested by the commonly used “spotlight” metaphor
(Eriksen and Hoffmann 1 97 3; Posner 1 980), v isual attention appears to be spatially based under most circumstances.
Ex perimental ev idence suggests that attention can also be allocated to other v isual attributes, including color (Cav e 1 999),
motion, and ev en objects (Duncan 1 984; V ecera and Farah 1 994).
Limited-capacity operations that serv e to select v isual information for additional processing (including integration) hav e
been demonstrated at many lev els of neural processing, ranging from relativ ely high-lev el v ision (Corbetta 1 998),
intermediate v ision (Moran and Desimone 1 985), and ev en the primary v isual cortex (V idy asagar 1 998). Despite the fact
that these operations differ somewhat from the procedure by which v isual features are integrated, this process is also
ty pically described as v isual attention. As will be discussed below, disruption of v isual attention at different lev els of v isual
processing may giv e rise to distinctly different clinical sy ndromes.
The integration of v isual feature information generates a v iewer-centered representation of the orientations and depths of
the surfaces of an object as well as the discontinuities between the surfaces. This representation is similar in most important
respects to the “2 1 /2D model” described by Marr (1 982), which makes ex plicit the form, shape, and v olume of the object as
well as the hierarchical relationships between the parts. This ty pe of representation has been termed a “structural
description” (Riddoch and Humphrey s 1 987 ) and is similar to the 3-D representation of Marr (1 982) in that it is assumed to
be perspectiv e independent; that is, at this lev el of processing, the representation computed by the brain specifies the
relationships between the v isual features in an “object-centered” fashion. The nature of the processes that mediate between
the v iewer-dependent, integrated feature representation and the perspectiv e-independent or object-centered structural
description remain unknown. As rev iewed briefly by Farah (2000), a v ariety of accounts hav e been proposed, from
connectionist architectures to template-matching procedures. Details of these proposals are bey ond the scope of this
chapter.
As will be discussed below, a large number of imaging studies suggest that the fusiform gy ri and lateral occipital region are
crucial anatomical substrates for the integrated feature representation and structural description sy stems (Grill-Spector,
Knouf, and Kanwisher 2004; Hax by , Hoffman, and Gobbini 2000 for rev iews).
Familiar objects are quickly and effortlessly recognized when v iewed across a wide range of angles and perspectiv es. A bowl,
for ex ample, may be v iewed from the side when sitting in a dishwasher, from below when stacked on a high shelf, or from
abov e when clearing a table.
Although the low-lev el v isual information regarding surfaces, color, form, etc. is remarkably different in these
circumstances, under normal circumstances these stimuli are immediately recognized as the same. In the contex t of Figure
1 0.2, the mechanism that supports object constancy is termed the v iew normalization sy stem.
The v isual processing discussed to this point is in the serv ice of object recognition—that is, knowledge of the form, function,
name, and other attributes of entities in the env ironment. Following Lissauer (1 890) as well as recent accounts of semantic
representations (Allport 1 985; Damasio 1 989; Rogers et al. 2004; Saffran and Schwartz 1 994; Warrington and Shallice 1 984;
Simmons and Barsalou 2003), I suggest that recognition entails the simultaneous activ ation of the many aspects of
knowledge-specific knowledge; thus, as argued by Lissauer (1 890), for ex ample, recognition of a v iolin consists of the
activ ation of stored knowledge regarding the sound, heft, feel, and manner of manipulation of the instrument.
With this ov erv iew of the series of processes underly ing v isual recognition in mind, we now turn to a consideration of the
different ty pes of v isual agnosia. Rather than use the traditional “apperceptiv e–associativ e” distinction described abov e, we
discuss the clinical phenomena with reference to the putativ e locus of the processing deficit ex hibited by the patient.
10.2.1.2 Disorders of Low-Lev el Vision

A v ariety of clinical disorders hav e been described in which the primary deficit is a disruption of processing of different
ty pes of v isual feature information. These are rev iewed briefly in this section.
1 0.2.1 .2.1 Achromatopsia

Achromatopsia is an acquired disorder of color perception characterized by a loss of the ability to distinguish color. The
disorder is probably far more common than widely appreciated because it v aries in sev erity from a mild loss of the richness
of color (e.g., “red desaturation”) to a complete loss of the sense of color. Milder forms are often not recognized by the
patient whereas in more sev ere forms the patient may indicate that they see the world in black and white or shades of gray .
The defect is often in one v isual field or part thereof but may inv olv e the entire v isual field if both hemispheres are affected.
Since V errey ’s initial report of the disorder in 1 888, achromatopsia has been consistently associated with lesions inv olv ing
the lingual or fusiform gy ri. Subsequent studies with static brain imaging (Damasio et al. 1 980) and functional brain imaging
hav e y ielded generally similar results. Reflecting this anatomic substrate, achromatopsia may be observ ed in isolation or in
association with conditions such as prosopagnosia, alex ia, or superior v isual field deficits that are associated with lesions to
nearby cortex (e.g., Pearlman, Birch, and Meadows 1 97 9). Interestingly , the loss of color perception is not ty pically
associated with v isual object agnosia.
1 0.2.1 .2.2 Impaired Motion Perception (Akinetopsia)

Reflecting its crucial role in v ision, motion cells that appear to be optimized for motion perception may be identified at the
retina, lateral geniculate, primary v isual cortex , and a number of higher-lev el v isual cortices, including MT, a cortical
region that appears to be specialized for motion processing (see below).
Relativ ely pure disorders of motion perception are rare. The sy ndrome was first reported almost a century ago (Goldstein
and Gelb 1 91 8; Potzl and Redlich 1 91 1 ). In recent y ears, a subject with this disorder, LM, has been ex tensiv ely inv estigated
(e.g., Zihl, Cramon, and Mai 1 983). LM dev eloped a profound impairment in the ability to detect motion after bilateral
infarcts inv olv ing the posterior portions of the middle temporal gy ri ex tending into the occipital lobe as well as adjacent
subcortical white matter. The deficit had profound consequences for her ability to negotiate her env ironment. She did not
perceiv e mov ement as a continuous process but stated that objects seemed to jump from one position to the nex t. When she
poured water into a cup, the liquid appeared to be static, like a piece of ice. Although profoundly impaired in motion
perception, LM performed well on other measures of v isual processing. Her v isual fields were full and she performed
normally on tests of stereopsis, v isual acuity , color perception, and critical flicker fusion. At least under most
circumstances, LM ex hibited no impairment in object recognition.
Sev eral patients hav e been reported whose object recognition is influenced by motion. Botez and Serbanescu (1 967 )
reported two patients with a “static form agnosia” characterized by a failure to recognize stationary stimuli but much
improv ed performance when the same stimuli were mov ed. A similar but perhaps less striking facilitation of recognition
with mov ement was ex hibited by the “v isual form agnosic” reported by Benson and Greenberg (1 969) as well as a patient
reported by Horner and Massey (1 986). We hav e also observ ed this phenomenon in a number of patients with hy pox ic brain
injury or degenerativ e diseases preferentially inv olv ing the posterior portions of the hemispheres (that is, “posterior
cortical atrophy ”). In these patients, it appeared that patients were unable to reliably deploy v isual attention to different
stimuli in the array ; mov ement seemed to permit the subjects to fov eate the stimulus that was then recognized by normal
procedures.
Data from patients with achromatopsia as well as abundant fMRI studies implicate the posterior portion of the middle
temporal gy rus (“V 5”) as a cortical region specialized for motion processing (Kable, Lease-Spellmey er, and Chatterjee 2002;
Watson et al. 1 993).
10.2.1.3 Visual Form Agnosia

One well-described but relativ ely rare sy ndrome, “v isual form agnosia,” may represent a deficit in the segregation of
coherent regions of v isual input. One such subject, a 24-y ear-old man who had suffered carbon monox ide poisoning (Mr. S),
was reported by Benson and Greenberg (1 969) and inv estigated in detail by Efron (1 968). Mr. S. was relativ ely intact with
respect to general cognitiv e function but was substantially impaired in recognizing v isually presented objects, drawings,
letters, or faces. Ex tensiv e testing of sev eral low-lev el v isual attributes rev ealed normal performance on tasks requiring the
detection of differences in luminance, wav elength, area, and motion. Perhaps his most striking v isual deficit was a profound
impairment in the ability to discriminate shape; for ex ample, if presented with a square and a tall, skinny rectangle
(height:width ratio of 4:1 ), matched for area, color, and other v isual qualities, he was unable to indicate if the stimuli were
the same or different shapes.
More recently , Milner and Hey wood (1 989) reported a second v isual form agnosic, DF; like LM, she had suffered carbon
monox ide poisoning. Structural MRI demonstrated bilateral lesions inv olv ing the lateral occipital area. She ex hibited
profound v isual recognition problems and performed poorly on tasks requiring that she discriminate between different
shapes.
DF ex hibited a finding of great theoretical interest that had not been described prev iously . Although unable to distinguish
between v isual forms or to name objects, she performed normally with respect to hand posture and shape when asked to
pick up the objects; thus, when asked to pick up rectangles whose shape she was unable to describe, the distance between
her thumb and index finger and timing of the mov ements of the fingers in the reach trajectory were normal. Thus,
information regarding v isual form that was not av ailable for the purposes of object analy sis was av ailable to the motor
sy stem.
10.2.2 DISORDERS OF VISUAL ATTENTION

As described abov e, v isual attention is a limited-capacity resource that is ty pically accorded a v ariety of roles in v isual
processing. Reflecting the div ersity of functions attributed to v isual attention, it is perhaps not surprising that disorders of
v isual attention hav e been implicated in a number of different clinical sy ndromes.
Perhaps the prototy pical sy ndrome of this ty pe is “simultanagnosia,” an inability to “see” more than one object in an array
(Wolpert 1 924). The first detailed description of this sy ndrome was by Balint (1 909), who described a patient with bilateral
posterior parietal infarcts who was able to identify v isually presented familiar objects when presented in isolation but
ex hibited a striking difficulty in the processing of v isual array s. For ex ample, when shown a letter and a triangle, he reported
seeing only the letter; when told that a second object was present, he reported the triangle but no longer saw the letter. As
Balint’s patient had normal v isual fields and v isual acuity , the disorder could not be attributed to low-lev el v isual processing
deficits. Similar patients hav e been reported by a number of inv estigators (Holmes 1 91 8; Luria 1 959; Coslett and Saffran
1 991 ; Coslett and Lie 2008).
Simultanagnosic subjects are often impaired in recognizing single objects as well as array s. The deficit with single objects
usually consists of a failure to appreciate the entire stimulus. For ex ample, when confronted with a complex object such as a
car, subjects with this disorder may report only a tire; similarly , when shown the word “table,” subjects may not see the
entire word but report constituent letters, often identify ing the word in a letter-by -letter fashion. Object size does not
appreciably influence performance. When confronted with an array , simultanagnosics often report seeing only one item at a
time.
Farah (2004) introduced the distinction between “dorsal” and “v entral” simultanagnosia. The former is associated with
dominant hemisphere posterior lesions and is usually associated with a hemianopia. Dorsal simultanagnosia is associated
with bilateral posterior parietal-occipital lesions (Rizzo and Hurtig 1 987 ). We hav e recently suggested that dorsal
simultanagnosia may be further subdiv ided (Coslett and Chatterjee 2003). On our analy sis, one form of dorsal
simultanagnosia may be attributable to an impairment in the process by which v isual attention is allocated or serv es to
integrate v isual feature information. As might be ex pected giv en the role accorded v isual attention in the integration of
v isual feature information, patients with this disorder generate frequent “illusory conjunctions” characterized by the
incorrect combination of v isual features; these subjects may report, for ex ample, a red T when shown an array of red X’s and
green T’s (Pav ese et al. 2002; Robertson et al. 1 997 ). We suggested that simultanagnosia may also be attributable to an
impairment in linking object location and identity (Coslett and Chatterjee 2003). Consistent with this perspectiv e, one
subject with simultanagnosia was impaired in reporting not only more than one object in an array , but also more than one
attribute of a single object. For ex ample, he was unable to report both the color of the ink in which a word was written as
well as the word. For both this and a prev iously reported subject, performance was significantly influenced by semantic
factors; for ex ample, both subjects were able to report two items from an array on a significantly greater number of trials if
the items were semantically related (e.g., both tools) as compared to trials on which they were unrelated (e.g., one tool, one
animal). Thus, data from these subjects demonstrated that the patients were, in fact, processing v isual information that they
were unable to report.
Disorders of v isual attention are a prominent feature of the “simultanagnosia-like” disturbance encountered in other
conditions such as the sy ndrome of “posterior cortical atrophy ” (Benson, Dav is, and Sny der 1 988). The most common cause
of this sy ndrome is Alzheimer’s disease. Like patients with simultanagnosia from focal parieto-occipital lesions, these
patients often identify single objects relativ ely well but are substantially impaired in the processing of array s. These patients
are often sev erely impaired in ev ery day activ ities such as finding an item in the refrigerator or the butter dish on the table.
They may be unable to find their way in their own home; we hav e encountered a number of patients whose v isual world is so
“confusing” that they close their ey es when walking or searching a complex array . The v isual processing deficit in these
patients ty pically differs in one telling way from that of simultanagnosic patients with focal parieto-occipital lesions: they
ex hibit a striking effect of object size (Coslett et al. 1 995; Saffran, Fitzpatrick-DeSalme, and Coslett 1 990). We hav e suggested
that these and other findings ex hibited by these patients are consistent with the hy pothesis that the patients suffer from a
pathologic restriction in their capacity to integrate v isual feature information.
10.2.3 I MPAIRMENTS IN THE VIEW NORMALIZATION SYSTEM

Impairments in identify ing objects seen from unusual perspectiv es or matching objects across different v iews (see Figure
1 0.2) is a relativ ely common but rarely diagnosed disorder. The first sy stematic inv estigations of the phenomenon were
reported by Warrington and Tay lor (1 97 3, 1 97 8). In one study they asked patients to name two sets of 20 objects, one
depicting the object in a canonical or standard v iew and the other in an unusual v iew. They found that subjects with lesions
inv olv ing the posterior portion of the right (non-dominant) hemisphere were selectiv ely impaired in the recognition of the
unusual v iews of these objects. Based on a series of elegant studies and tasks, Humphrey s and Riddoch (1 984, 2006)
suggested that the deficit may be observ ed for different reasons. Thus, they reported one patient who appeared to rely on
distinctiv e feature information; other patients, in contrast, hav e been found to be particularly sensitiv e to foreshortening
(see also Warrington and James 1 986), suggesting that these patients are unable to generate a representation of the object
relativ e to its principal ax is.
10.3 IMPAIRMENTS OF THE STRUCTURAL DESCRIPTION SYSTEM (“ASSOCIATIVE

AGNOSIAS”)
Failures of object recognition in the contex t of v isual processing that is adequate to support copy ing of a stimulus hav e been
reported on a number of occasions (Dav idoff and Wilson 1 985; Lev ine 1 97 8; Lev ine and Calv anio 1 989; see Farah 2004 for a
comprehensiv e rev iew). Rubens and Benson (1 97 1 ) reported a patient who ex hibited a striking form of associativ e agnosia
after a hy potensiv e episode. Elementary v isual function was normal ex cept for right hemianopia; the subject was unable to
name most objects y et drew accurate and quite detailed depictions of the v ery objects that he was unable to name.
From Lissauer’s initial account to the present, the pattern of deficits display ed by Rubens and Benson’s patient has been
attributed to disruption of the structural description sy stem. Associativ e v isual agnosia is differentiated from loss of
knowledge of the object (semantics) by the fact that, when queried v erbally or by means of auditory or tactile input, the
patients are able to prov ide appropriate information about objects that they are unable to recognize v isually ; thus, when
asked to prov ide a v erbal description of a hammer, for ex ample, an associativ e agnosic would be ex pected to describe its
function, heft, and sound as well as demonstrate its manner of manipulation. Similarly , the patient would be ex pected to be
able to name the hammer after handling it.
Despite the fact that they are able to describe or copy v isual stimuli quite adequately , it seems unlikely that they are
contacting normal structural descriptions for sev eral reasons. First, as emphasized by Farah (2000), copies generated by
v isual agnosics are ty pically produced in a “slav ish” and painstaking fashion, presumably reflecting a reliance on the ex act
phy sical attributes of the stimulus with little or no “top-down” input from stored knowledge of object appearance. This point
is illustrated by the copy depicted in Figure 1 0.2; this subject copied the stimulus—including the ex traneous, sinusoidal line
—without distinguishing between the irrelev ant and object-specific information. Second, it is noteworthy that the quality of
the copies generated by associativ e agnosics is ty pically strongly influenced by the richness of the v isual image; subjects
generally perform best with real objects, less well with pictures, and worst with line drawings. This hierarchy would not be
ex pected if all three ty pes of stimuli contacted the same structural description.
Finally , it should be noted that when asked to match or sort complex non-object figures, all associativ e agnosics of which we
are aware ex hibit impairments (Humphrey s and Riddoch 1 987 ; Ratcliff and Newcombe 1 982; see Farah 2004 for
discussion). This observ ation suggests that these subjects hav e some degree of impairment in the processes by which
structural descriptions are accessed.
It should be emphasized that although the model depicted in Figure 1 0.1 prov ides a useful framework for considering the
general principles inv olv ed in v isual processing and their breakdown, it does not readily accommodate the full range of
agnosic performance. One well-studied “integrativ e agnosic,” HJA (Riddoch and Humphrey s 1 987 ), illustrates this fact.
Although the details of the subject’s performance are bey ond the scope of this chapter, it is noteworthy that HJA appears to
be an associativ e agnosic (Farah 2004), whereas in other regards his performance is more ty pical of apperceptiv e agnosia
(Riddoch and Humphrey s 2003).
10.3.1 PROSOPAGNOSIA
Prosopagnosia, or face blindness, is a condition in which subjects are unable to identify another person by v iewing the
person’s face (Bodamer 1 947 ). Although they are often capable of recognizing that a face is a face and, in many instances,
are able to indicate the gender, age, and likely occupation of the person, they are unable to identify the person from v ision
alone. The deficit may be so profound that subjects are unable to identify their own face or that of immediate family
members (Bauer and V erfaellie 1 986).
Prosopagnosia may be asy mptomatic in some cases. Many subjects rely on information regarding hairsty le, habitus,
clothing, or v oice to circumv ent their v isual deficits. In some instances, the deficit may only be apparent when subjects
encounter a familiar person in an aty pical setting. We hav e seen patients, for ex ample, who were only noted to be
prosopagnosic when they failed to recognize a close family member who was wearing a nursing uniform. In this contex t, the
absence of cues from clothes, hairsty le, and contex t unmasked a deficit for which the patient ty pically compensated quite
successfully . Many prosopagnosic subjects ex hibit additional deficits in other aspects of v isual processing such as
achromatopsia, v isual field deficit, topographical memory loss, or object agnosia.
Although “pure” prosopagnosics hav e been described, many patients with the disorder ex hibit deficits in the recognition of
other ty pes of stimuli at a subordinate lev el. That is, subjects may be able to recognize dogs but not able to discriminate
between their dog and others of the same species (see Borenstein, Sroka, and Munitz 1 969).
A v ariety of hy potheses hav e been adv anced to ex plain the phenomenon of prosopagnosia. On some accounts, the deficit is
assumed to be an inability to discriminate indiv idual ex emplars within a class of v isually similar items (Damasio, Damasio,
and Hoesen 1 982). Citing the double dissociation between v isual object agnosia and prosopagnosia (Moscov itch, Winocur,
and Behrmann 1 997 ), other inv estigators hav e argued that faces are processed by a distinct sy stem; Ellis and Y oung (1 988),
for ex ample, hav e proposed that processing of faces is in some respects akin to that of objects, in that “early ” disruptions of
the face processing sy stem are associated with “apperceptiv e” prosopagnosia whereas disorders later in the information
processing cascade are associated with “associativ e” prosopagnosia. On these accounts, subjects who are unable to deriv e
information about gender and other attributes from faces fall into the former category whereas patients who performed well
on these tasks but who could not identify the subject would fall into the latter category .
A third account has been articulated by Farah (2004). She noted that whereas deficits restricted to words or faces hav e been
described with some regularity , patients ex hibiting deficits with objects in the contex t of normal recognition of faces or
words are quite rare (but see Bux baum, Glosser, and Coslett 1 996). These findings led Farah (2004) to propose that two
distinct but interactiv e modes of object processing subserv e stimulus recognition; by her account, one mechanism is
specialized for objects that are processed as a unit, that is, with relativ ely little decomposition into simpler parts; v arious
lines of ev idence support the v iew that faces are processed in this manner. A second mechanism is specialized for stimuli
that undergo substantial decomposition; for these items, the parts into which they are segregated may be discrete objects
themselv es. As they are composed of a finite number of discrete objects and their identity is entirely determined by the
constituent units, words may represent the prototy pical stimulus for the decomposition procedure. Intense debate
continues regarding the degree to which the processing of faces represents a distinct module or is best conceptualized as
one end of a continuum along the dimension of holistic (configural) as opposed to decompositional (elemental) processing
(for sensory -perceptual perspectiv es on this debate, see Chapter 5).
Prosopagnosia is usually associated with lesions of the temporo-occipital regions bilaterally (Damasio, Damasio, and Hoesen
1 982). Sev eral well-studied cases (e.g., DeRenzi 1 986; Michel, Pernin, and Sieroff 1 986) hav e been described with lesions
inv olv ing only the right hemisphere. Additionally , a substantial functional imaging literature has identified a region of the
fusiform gy rus bilaterally —the “fusiform face area”—that is reliably activ ated by faces (for a recent rev iew see Harris and
Aguirre 2008). Whether this should be considered a cortical module dedicated to the processing of faces or a part of the
broader v isual recognition sy stem that is optimized for stimulus properties that characterize faces has been a topic of
substantial debate (Hax by , Hoffman, and Gobbini 2000; McKone and Kanwisher 2004; Tarr and Gauthier 2000).
10.3.2 AGNOSIA FOR WORDS

This is also known as pure alex ia, alex ia without agraphia, or pure word blindness. Although this condition is usually
discussed in the contex t of language impairments, it is an agnostic sy mptom, as the condition is restricted to a specific ty pe
of v isual stimulus; most subjects with this condition do not ex hibit impairment in language (e.g., speech, writing, auditory
comprehension). As noted initially by Dejerine (1 892) and confirmed by multiple inv estigators since (e.g., Saffran and
Coslett 1 996), the disorder is ty pically associated with lesions inv olv ing the dominant occipital lobe and the splenium of the
corpus callosum that serv e to depriv e the left hemisphere of v isual input while simultaneously disconnecting the right
occipital lobe from the left peri-Sy lv ian cortex .
10.4 CATEGORY-SPECIFIC RECOGNITION DEFICITS

As noted prev iously , for some patients the ability to recognize v isually presented stimuli is significantly influenced by the
semantic category of the item. Thus, a number of patients hav e been reported who are able to name man-made objects but
are sev erely impaired in naming naturally occurring items such as animals, fruits, and v egetables. A number of competing
hy potheses hav e been proposed to ex plain these category -specific deficits. The modality -specific hy pothesis (Warrington
and Shallice 1 984), later named the Sensory -Functional Theory (SFT; Caramazza and Shelton 1 998), postulates that the
animate v ersus inanimate dissociation follows from a selectiv e impairment of the sensory or functional attributes that
subserv e the processing of either of these two categories. By this account, the identification of animate objects relies more
on sensory attributes and would be disproportionately impaired by damage to the processing of sensory features associated
with these objects. In contrast, inanimate objects may be known primarily by v irtue of their function and the manner in
which they are used or manipulated. As a consequence, a deficit in the recognition of inanimate objects would be disrupted
by loss of information regarding the function of an object or sensory -motor knowledge regarding the manner in which the
object is manipulated. A competing hy pothesis is that the semantic knowledge is organized categorically in the brain
(Caramazza and Shelton 1 998). Clearly , teleologic ex planations for this organization can be made. Ev olutionary pressures
would fav or an animal that could easily recognize and distinguish other animals that are potential predator or prey , or
plants that are potential sources of food. Further, dev elopmental data support the idea that infants as y oung as 3 months of
age can differentiate liv ing from non-liv ing things.
We recently studied a 50-y ear-old patient who ex hibited a category -specific v isual agnosia. After suffering a stroke of the
right temporal lobe in the contex t of hy potension, he dev eloped amnesia, prosopagnosia, and an inability to recognize
liv ing things. He was significantly impaired naming animate as compared with inanimate items. Additional testing rev ealed
that his ability to name objects was strongly predicted by the nature of his ex perience with the object. Those objects—
whether animate or inanimate—that he knew by v irtue of using or manipulating (e.g., hammer) were named relativ ely
accurately whereas those items that he knew primarily by sight (e.g., elephant) were named poorly . We argued that he
ex hibited a mild v isual agnosia but that “motor knowledge” could compensate at least in part for the deficits in the object
recognition sy stem.
10.5 FUNCTIONAL IMAGING CORRELATES OF OBJECT RECOGNITION

In the past few y ears, functional imaging in healthy subjects has helped to elucidate the anatomy of object recognition.
Malach et al. (1 995) described the lateral occipital cortex (LOC), which is located at the lateral and v entral aspects of the
occipito-temporal cortex . This area is activ ated preferentially by objects compared with scrambled objects or tex tures,
regardless of the nature of the object (e.g., faces, cars, common objects, and ev en unfamiliar abstract objects) (Malach,
Lev y , and Hasson 2002). Further processing of the v isual stimuli occurs in specific brain areas according to stimulus
category . For ex ample, faces are processed in the fusiform face area (Kanwisher, McDermott, and Chun 1 997 ) and in the
occipital face area (Gauthier, Tarr, and Moy lan 2000); the former is more selectiv e for faces than the latter. Places and/or
spatial lay outs are processed in the parahippocampal place area (Epstein and Kanwisher 1 998), whereas objects like animals
and tools are processed in specific loci in the fusiform and the superior and middle temporal gy ri (Chao, Hax by , and Martin
1 999). Orthographic stimuli are processed in the left inferior occipito-temporal cortex on or near the left fusiform gy rus
(Polk et al. 2002; Allison et al. 1 994), although this localization is highly debated (Price and Dev lin 2003). Thus, the specific
perceptual categories of v isual stimuli dictate v ery different way s of their processing in the brain and might giv e an
anatomic basis for the different agnostic sy ndromes.
10.6 AUDITORY AGNOSIAS

Like the v isual agnosias discussed abov e, auditory agnosias are characterized by an inability to recognize a stimulus—in this
case, a sound—that cannot be ex plained by inadequate elementary sensory processing, generalized loss of knowledge (e.g.,
dementia), or inadequate attention to the task. Perhaps reflecting the fact that in the contex t of normal v ision, audition is
not usually needed for object recognition, auditory agnosias hav e been reported far less commonly than v isual agnosias.
Indeed, in most naturalistic settings, one is rarely required to identify an object on the basis of sound alone. Ev en in those
circumstances in which this is necessary —for ex ample, identify ing the sound of a telephone—the range of potential stimuli is
narrow, limiting the complex ity of the processing required. Perhaps because of these factors, complaints of poor
recognition of objects from sound are uncommon. In our ex perience, when questioned about their auditory recognition
deficit, most subjects will concede that sounds simply “aren’t right” and, when pressed, attribute the disorder to hearing
loss.
There is, howev er, reason to believ e that auditory agnosias are under-reported. On a number of occasions, we hav e
observ ed patients with auditory recognition deficits ev ident on formal testing who were utterly unaware of the deficit.
Additionally , most patients with documented auditory sound agnosia do not complain of an inability to recognize sounds
(Say gin et al. 2003; V ignolo 2003). For ex ample, although aphasic subjects rarely complain of difficulty understanding non-
speech sounds, acquired language disorders are frequently associated with a generalized disorder of auditory recognition
(V ignolo 1 982). Say gin et al. (2003), for ex ample, recently reported an elegant study in which 30 left-hemisphere-damaged
aphasic subjects were asked to match env ironmental sounds (sound of a cow mooing) or linguistic phrases (“cow mooing”)
to pictures. They found that the aphasic subjects were impaired with both word and sound stimuli; furthermore, there was a
high performance correlation between sev erity of aphasia and accuracy on the env ironmental sound-matching task. Lesion
ov erlay analy sis demonstrated that damage to posterior regions in the left middle and superior temporal gy ri and inferior
parietal lobe was a predictor of impaired performance on both tasks.
Although accounts of auditory recognition are, in general, less well-dev eloped than those in the v isual domain, sev eral
inv estigators hav e argued that two forms of auditory sound agnosia may be identified (Kleist 1 928). Echoing Lissauer’s
distinction between apperceptiv e and associativ e v isual agnosias, V ignolo (1 982) distinguished between a perceptual or
discriminativ e agnosia linked to right hemisphere lesions and an associativ e agnosia observ ed with left hemisphere lesions.
Although subjects with both conditions ex hibit the same fundamental deficit, an inability to recognize objects from sound,
support for the claim that they arise at different lev els of processing comes from analy ses of the errors (V ignolo 1 982;
Schnider, Benson, and Scharre 1 994). In the right hemisphere “apperceptiv e” form of auditory agnosia, confusions are
ty pically based on the features of the sound; for ex ample, a car horn may be confused with a musical instrument. In left
hemisphere “associativ e” auditory agnosia, errors appear to be semantically based; for ex ample, subjects may confuse a
police whistle with a siren.
V ignolo (2003) recently reported data that support this lateralization of function. Tasks assessing music and identification
of env ironmental sounds were administered to 40 subjects with unilateral stroke; right hemisphere lesions tended to disrupt
the apperception of env ironmental sounds whereas left hemisphere lesions disrupted the semantic identification of sounds.
Finally , recent ev idence from functional imaging studies in normal subjects is also consistent with this claim. Lewis et al.
(2004) reported data from an fMRI study in which participants listened to a wide range of env ironmental sounds while
undergoing BOLD imaging. The contrast between recognizable and unrecognizable stimuli rev ealed activ ity in a distributed
network of brain regions prev iously associated with semantic processing, most of which were in the left hemisphere.
Auditory information is not only relev ant to the recognition of words and entities in the env ironment but also to spatial
processing; that is, just as v ision may be used in parallel for the identification and localization of objects, sound may also be
employ ed in the serv ice of spatial processing. Although bey ond the purv iew of this chapter, it should be noted that both
functional imaging and lesion studies hav e suggested that distinct processing pathway s may underlie these capacities
(Griffiths et al. 1 998). Clarke et al. (2000), for ex ample, reported detailed inv estigations of four subjects with left
hemisphere lesions who ex hibited substantial dissociations in the ability to recognize and localize auditory stimuli,
suggesting that the capacities to identify and localize auditory stimuli are, at least in part, distinct and dissociable.
In the following sections we briefly rev iew the major sy ndromes of auditory agnosia starting with cortical deafness and
culminating in disorders of auditory word recognition and recognition of auditory affect.
10.7 CORTICAL DEAFNESS AND GENERALIZED AUDITORY AGNOSIA

Cortical deafness is a rare disorder characterized by profound loss of awareness of sound; in its most profound form,
patients appear to be deaf. Patients with this disorder hav e ty pically suffered ex tensiv e bihemispheric lesions that destroy
superior temporal cortex crucial for the analy sis of sound (Leicester 1 980). More recently , sev eral patients with ex tensiv e
subcortical bilateral lesions that undercut and, presumably , disconnect the auditory cortex from input from the medial
geniculate hav e been reported (e.g., Kazui et al. 1 990; Kaga et al. 2005).
As in the v isual domain, the distinction between a primary disorder of auditory processing and “higher-lev el” disorders of
auditory recognition remains controv ersial. Michel et al. (1 980) proposed a number of criteria to differentiate these
conditions, including auditory ev oked potentials as well as mapping of the lesions to koniocortex as opposed to pro- and
parakoniocortex . Neither of these approaches has been demonstrated to prov ide an adequate account of the distinction.
Consistent with the animal literature demonstrating that bilateral ablations of auditory cortex do not consistently produce
complete deafness (e.g., Neff 1 961 ; see also Celesia 1 97 6), most subjects who present with substantial bilateral temporal
strokes also do not ex hibit cortical deafness. Similarly , the rare subjects who present with cortical deafness ty pically regain
awareness of sound but continue to ex hibit profound auditory recognition deficits for all kinds of stimuli. With formal
testing, the errors of subjects with generalized auditory agnosia ty pically represent confusions based on acoustic properties
rather than meaning; thus, in the contex t of the two-stage model described abov e, the disorder may be described as an
“apperceptiv e” disorder.
10.7.1 AUDITORY SOUND AGNOSIA (AUDITORY AGNOSIA FOR NON-SPEECH SOUNDS)

This disorder is characterized by apparently normal awareness of sound but an inability to identify non-v erbal sounds.
Although, as noted abov e, some patients with auditory sound agnosia may be impaired in recognizing speech (e.g., Say gin et
al. 2003), in other subjects the disorder may be relativ ely “pure” in that speech may be largely preserv ed whereas
recognition of auditory sounds is clearly impaired (e.g., Spreen, Benton, and Finchman 1 965; V ignolo 1 982; Fujii et al. 1 990;
Schnider, Benson, and Scharre 1 994; Clarke et al. 2000). Auditory agnosia restricted to non-v erbal sounds is ty pically
associated with bilateral (Spreen, Benton, and Finchman 1 965; Albert et al. 1 97 2; Kazui et al. 1 990) or right hemisphere
lesions (Fujii et al. 1 990).
10.7.2 PURE WORD DEAFNESS (AUDITORY AGNOSIA FOR SPEECH)

Pure word deafness is a rare and often striking disorder in which patients are unable to understand auditory language but
ex hibit no significant hearing loss and maintain the ability to recognize sounds. It is differentiated from aphasia by the
preserv ation of language as manifested by speech production, naming, reading, and writing. Reflecting the central role of
speech in human interaction, patients with pure word deafness are ty pically aware of and distressed by their difficulty .
Two major pathologic substrates of pure word deafness hav e been identified. One group of patients suffers from bitemporal
lesions, usually inv olv ing the primary and secondary auditory cortices of the superior temporal gy rus (e.g., Coslett,
Brashear, and Heilman 1 984); another group suffers from single lesions of the left temporal lobe or temporal isthmus. From
an anatomic perspectiv e, both ty pes of lesions are assumed to giv e rise to the same basic problem: a dissociation between
intact language cortices in the left perisy lv ian region and auditory input.
Sev eral inv estigators hav e proposed that distinct subty pes of pure word deafness may be identified. Some inv estigators
(e.g., Albert and Bear 1 97 4; Auerbach et al. 1 982; Mendez and Geehan 1 988; Coslett, Brashear, and Heilman 1 984) hav e
proposed that the disorder is attributable to a loss of temporal acuity ; on this account, speech is selectiv ely impaired
because it is far more demanding with respect to temporal precision than auditory sounds, most of which do not require fine
auditory distinctions (Albert and Bear 1 97 4). An alternativ e interpretation is that the disorder is caused by a disruption of
phonemic discrimination (Saffran, Schwartz, and Marin 1 97 6; Denes and Semenza 1 97 5). Auerbach et al. (1 982) suggested
that the former subty pe of pure word deafness was associated with bilateral lesions whereas the latter was associated with
left hemisphere lesions.
Although they may perform at chance on auditory word-to-picture matching tasks, patients with pure word deafness often
perform surprisingly well in some naturalistic settings. For ex ample, whereas patients with pure word deafness may be
utterly unable to communicate by telephone, they may communicate relativ ely well in face-to-face conv ersations. Sev eral
factors may contribute to this. One possibility is that these patients may be adept at using a “top-down” process to narrow
the possible range of candidate words from which the subject must choose; thus, in a conv ersation about politics, one may
ex pect certain words to appear, and this knowledge may bias the impaired recognition sy stem by priming likely words.
Alternativ ely , some inv estigators hav e argued that lip-reading prov ides useful information regarding factors such as place
of articulation that supplements the impov erished auditory input (Auerbach et al. 1 982). Additionally , some patients may
be able to use non-v erbal auditory cues to supplement deficient word decoding.
Of note, we demonstrated that a patient with profound pure word deafness from bilateral lesions performed well on a task in
which he was asked to identify the affect (or emotion) ex pressed in a v erbal utterance (Coslett, Brashear, and Heilman 1 984).
These results suggested that patients with pure word deafness perform better in face-to-face communication because they
are able to make use of affectiv e and other non-v erbal cues. The dissociation between object identification and affectiv e
discrimination underscores the possibility that patients with sensory agnosia, in the tex tbook sense, may y et hav e sensory -
specific access to the emotional content of a stimulus. This prov ocativ e idea is taken up in Section 1 0.9.
10.7.3 T ACTILE AGNOSIAS (SOMATOSENSORY AGNOSIA )
Disorders of object recognition from tactile input hav e receiv ed little attention and remain poorly understood. There is
substantial terminologic confusion in this domain with a v ariety of different names being applied, often inconsistently , to
the general class of phenomena. Some inv estigators hav e applied the general term “tactile agnosia” to all disorders
characterized by poor identification of objects from palpation whereas others hav e suggested that this term be restricted to
those cases in which subjects fail to recognize palpated objects y et hav e normal or near normal somatosensory processing
(see Bauer 2003 and Bohlhalter, Fretz, and Weder 2002 for rev iews). Similarly , some inv estigators hav e reserv ed the term
“astereognosis” for disorders arising from a failure in low-lev el sensory processing. Here we refer to the entire range of
phenomena as tactile agnosias with the understanding that this does not reflect a consensus v iew and that the reader must
be attentiv e to terminology when ex ploring this literature.
As with the v isual and auditory agnosias prev iously discussed, tactile agnosias are defined both by what they are and what
they are not. Tactile agnosia is a disorder of object recognition from touch that cannot be ex plained by sev ere sensory -
motor disturbance, inattentiv eness, or general intellectual decline. Although there may be debate regarding the boundaries
between primary sensory -motor disorders at the one end of the spectrum and cognitiv e disorders at the other (e.g., Bay
1 944; Head and Holmes 1 91 1 ), there is considerable support for the claim that tactile agnosia is more than either of these
disorders. One strong argument for this point comes from the double dissociation between elementary somatosensory
processing disorders and tactile agnosia. In an inv estigation of 84 subjects with tactile object recognition impairments,
Caselli (1 991 ) reported substantial deficits in patients with normal or only mild deficits in somatosensory processing; these
and other data (see also Hecaen 1 97 2; Corkin 1 97 8) demonstrate that significant disorders of somatosensory processing do
not necessarily cause tactile agnosia; other inv estigators (e.g., Bohlhalter, Fretz, and Weder 2002) hav e demonstrated that
subjects with tactile agnosia may ex hibit normal or near-normal processing across a range of somatosensory tasks.
Wernicke (1 895) distinguished between tactile agnosias that inv olv ed a loss of a “tactile image” from those that were
attributable to an inability to associate the tactile image to its meaning. This distinction is, of course, reminiscent of
Lissauer’s apperceptiv e/associativ e dichotomy . More recently , Wernicke’s distinction has been adv ocated by a number of
inv estigators (Mauguiere and Isnard 1 995; Platz 1 996; Reed, Caselli, and Farah 1 996; Caselli 1 997 ). On this classification,
those subjects with at least relativ ely normal elementary somatosensory function but an impairment in object identification
through palpation who are impaired in matching items on the basis of tactile input and whose errors appear to reflect
deficits in form, shape, or tex ture discrimination are considered apperceptiv e tactile agnosics; those subjects who perform
well on tasks such as matching items by palpation and whose errors appear to reflect imprecision at the lev el of meaning
(e.g., responding “fork” when presented a spoon) are considered to be associativ e agnosics. Hecaen and Dav id (1 945), for
ex ample, reported a patient who could not name palpated objects but could draw the object with sufficient precision that he
could then name the object (see also Newcombe and Ratcliff 1 97 4).
Other inv estigators hav e attempted to classify the disorders on the basis of the putativ e ty pe of sensory processing deficit.
For ex ample, Delay (1 935), building on the work of v on Frey (1 895), Head (1 91 8), and others, proposed that tactile agnosias
could be div ided into subty pes in which shape and size were predominantly affected, as compared to forms in which other
qualities such as tex ture, weight, and temperature were affected. Consistent with such a v iew, Bohlhalter, Fretz, and Weder
(2002) recently reported data from two subjects who differed with respect to their ability to process “microgeometrical”
(e.g., tex ture) from “macrogeometrical” (e.g., length) information.
There is substantial agreement from the clinical literature regarding the anatomic bases of the tactile agnosias. The great
majority of patients with these disorders for which imaging or autopsy data are av ailable ex hibit lesions inv olv ing the
parietal lobe. Three regions of parietal cortex may be particularly relev ant. First, the post-central gy rus or primary
somatosensory cortex is inv olv ed in many subjects (Platz 1 996). Second, the parietal operculum, often designated SII or the
“secondary sensory cortex ,” has been lesioned in a number of well-studied subjects (e.g., Reed, Caselli, and Farah 1 996;
Caselli 1 991 ; Bohlhalter, Fretz, and Weder 2002). Finally , the superior parietal lobule (BA 5 and 7 ) receiv es direct
projections from the primary somatosensory cortex and has been postulated to be crucial for the integration of
somatosensory information to generate a high-lev el representation of the body and objects in space (Platz 1 996).
There is some data suggesting that the parietal operculum and the superior parietal lobule subserv e different functions
(Ledberg et al. 1 995; Roland, Sulliv an, and Kawashima 1 998). Whereas both areas are implicated in tactile recognition in
normal subjects undergoing fMRI scanning, Binkofski et al. (1 999) suggested that the parietal operculum (SII) may be
important for “microgeometrical” properties such as tex ture, while the superior parietal lobule may be more important for
macrogeometrical properties such as size and shape. Chapter 7 contains further general details of somatosensory function.
Finally , an unresolv ed issue concerns the hemispheric basis of tactile agnosias. As rev iewed by Bauer (2003), there is no
clear ev idence of a hemispheric difference in tactile agnosia. There is, howev er, suggestiv e ev idence that disorders in the
ability to appreciate shape, size, and other “spatial” attributes of a stimulus may be more marked in subjects with right
hemisphere lesions. A second point concerns the observ ation that some patients ex hibit tactile agnosia in both hands after a
unilateral hemispheric lesion. For ex ample, Corkin, Milner, and Rasmussen (1 97 0) demonstrated that 20 of 50 patients with
unilateral hemispheric cortical ex cisions ex hibited bilateral sensory deficits; this was attributed to damage inv olv ing the
parietal operculum, which receiv es prominent projections from both primary somatosensory cortices. Lesions of primary
somatosensory cortex ty pically produce contralesional tactile agnosia.
10.8 OTHER AGNOSIAS

There has been an ex plosion of interest in the human neurobiology of taste and smell in recent y ears (Small 2006; Gottfried
2006). As the brain mechanisms underly ing these disorders hav e been inv estigated, patients who meet traditional criteria
for olfactory or gustatory agnosia hav e been identified. Small et al. (2005) reported a thorough inv estigation of a subject
with longstanding bilateral temporal lobe dy sfunction who dev eloped a gustatory agnosia after surgical resection of the left
anterior medial temporal lobe for control of seizures. Prior to surgery the patient performed normally on tasks assessing her
ability to detect and estimate the intensity of taste; additionally , although recognition thresholds were elev ated, she
successfully named basic tastes (e.g., sweet, sour). After resection of the left anterior medial temporal lobe she retained the
ability to detect, discriminate, and react hedonically to tastants but lost the ability to recognize the tastant. Based on these
and other data (Small et al. 1 997 ), the inv estigators argued that recognition of taste inv olv es interactions between the
primary gustatory centers located in the insula/opercular region and the regions of the anterior medial temporal lobe that
are critical for taste recognition.
Olfactory agnosias hav e also been described. For ex ample, Mendez and Ghajarnia (2001 ) reported a patient who was unable
to recognize faces and odors in the contex t of right temporal lobe dy sfunction. Jones-Gotman, Rouleau, and Sny der (1 997 )
reported data from a smell recognition test from 7 0 subjects who had undergone temporal lobectomy for seizure control.
They found that damage to the anterior temporal lobe of either hemisphere was associated with impairment of odor
identification with impairment of odor identification. Howev er, because basic smell detection was not formally assessed,
one cannot state with certainty that these subjects met the criteria for olfactory agnosia. Eichenbaum et al. (1 983)
demonstrated that patient HM, who had undergone bilateral anterior temporal lobe resections, ex hibited preserv ed early
olfactory processing but substantial impairment in higher-order olfactory processing. Gottfried and Zald (2005) hav e
recently rev iewed the contributions of orbitofrontal cortex to olfactory processing and the implications of lesions in this
region for odor recognition in humans and non-human primates.
10.9 AGNOSIA AND CONSCIOUS AWARENESS

As prev iously noted, agnosia is a disorder of stimulus recognition. Traditionally , ev idence for “recognition” has come from
ex plicit report: subjects name the object or prov ide a v erbal description that is sufficiently detailed so that the object can be
unambiguously identified. As in other domains in neuropsy chology (e.g., Schacter 1 987 ; McGlinchey -Berroth 1 996), there
is now abundant ev idence that some agnosic subjects may deriv e substantial information from stimuli that they fail to
“recognize.” This phenomenon, which has been termed “cov ert recognition” or “processing without awareness,” has been
studied most ex tensiv ely in prosopagnosics (see Bruy er 1 991 for rev iew). Bruy er et al. (1 983) demonstrated that a
prosopagnosic subject ex hibited greater difficulty learning to associate familiar but unrecognized faces with fictitious as
opposed to real names. Similar data from a face-name interference test were reported by DeHaan, Y oung, and Newcombe
(1 987 ) as well as Y oung et al. (1 986). Additional ev idence from psy chophy sical tasks was prov ided by Bauer (1 984), who
demonstrated that a prosopagnosic subject ex hibited the largest galv anic skin (“lie detector”) response to unrecognized
faces when the face was paired with the correct name (see also Bauer and V erfaellie 1 986; Tranel and Damasio 1 985).
Similar findings hav e been reported in other agnosic sy ndromes. For ex ample, Coslett and Saffran (1 989; Coslett et al. 1 993)
reported that patients with written word agnosia (“pure word deafness”) responded significantly better than chance on a
v ariety of semantic tasks to words that they claimed not to hav e “seen.” Sev eral inv estigators hav e demonstrated that
simultanagnosic subjects deriv e information from unreported stimuli (Robertson et al. 1 997 ). Coslett and Lie (2008), for
ex ample, demonstrated that a simultanagnosic patient’s report of two items in an array was significantly influenced by the
semantic relationship between items in an array : like a prev iously reported subject (Coslett and Saffran 1 993), this subject
reported items from the same semantic class (e.g., hammer, pliers) more reliably than items from different categories (e.g.,
skirt, apple).
The interpretation of these and other data demonstrating dissociations between implicit and ex plicit knowledge remains
controv ersial. It is clear from these and multiple other sources of data that patients with brain injury deriv e substantially
more information than they realize. In light of this observ ation, one may question whether this implicit knowledge
influences the behav ior of these subjects. For ex ample, is a v isual agnosic who can’t discriminate between a shotgun and
fire-poker on formal testing likely to pick up a shotgun to stoke a fire? Although one might ex pect that implicit knowledge
would constrain the behav ior of agnosic subjects, we are unaware of any data that speak directly to this point.
Although bey ond the scope of this chapter, there is a substantial (but controv ersial) body of literature suggesting that
unconscious thought is an important and efficient tool for decision making, problem solv ing, and attitude formation (see
Dijksterhuis et al. 2006). In a series of ex periments, Dijksterhuis and colleagues hav e demonstrated that “unconscious”
thought may , under some circumstances, be more rational and efficient than “conscious” thought. In one ex periment
(Dijksterhuis 2004), subjects were asked to rate the desirability of different apartments after being told of 1 2 dimensions
(e.g., cost, size) on which the apartments differed. One group of subjects was asked to respond immediately upon hearing
about the apartments, another group was giv en three minutes to ex plicitly consider the apartments, and a third group
performed a distracting (2-back) task for three minutes before responding. Subjects who were distracted for three minutes
and therefore not able to consciously consider the apartments performed significantly better in discriminating between the
most and least desirable apartments than those subjects who responded immediately or carefully considered the options.
Dijksterhuis and colleagues argue for a fundamental discrepancy between “conscious” and “unconscious” thought and claim
that the latter may be more efficient for certain ty pes of mental operations. The implications of “unconscious” thought for
the processes underly ing sensory agnosia are, at present, not clear, but they represent a fertile domain for future research.
10.10 CONCLUSION
Sensory agnosias are relativ ely uncommon but often striking disorders of recognition that are of considerable interest for
sev eral reasons. Not only are they important from a clinical perspectiv e but also, as emphasized by a number of
inv estigators (e.g., Farah 2004; Bauer 2003), by dissecting perceptual sy stems at the fault lines, the disorders may prov ide
important insights regarding the basic mechanisms of information processing and their anatomic bases. Using v isual agnosia
as the prototy pe, we hav e attempted to demonstrate that the confusion that arises from a consideration of the
phenomenology of the disorders can be mitigated by considering the disorders in the contex t of a theoretically motiv ated
account of basic perceptual mechanisms. Such an approach is, in principle, applicable to agnosias in other modalities.
Finally , there are sev eral areas for future research that appear particularly promising. The study of high-lev el recognition
deficits including the study of “category -specific” agnosias offers great promise in the ex ploration of semantics, including
the role of “grounded cognition” (see Barsalou 2008).
Additionally , in light of recent demonstrations of the role of unconscious operations in perception and thought, the study of
agnosic subjects may contribute to the understanding of the manner in which motiv ations, goals, and driv es are influenced
by factors about which subjects remain unaware. For ex ample, the ex tent to which Pav lov ian conditioning is present in
agnosic patients is not clear. The demonstration that a prosopagnosic patient could form a conditioned response between a
face that they could not identify and an unconditioned stimulus would suggest that reward-based conditioning does not
require recognition but could be based on lower-lev el sensory processing. To this end, Chapter 1 6 ex plicitly considers the
nov el idea that many of the behav ioral deficits observ ed in patients with frontal lobe dy sfunction reflect an agnosia for
object “v alue.”
REFERENCES
1 . Albert M., Bear D. Time to understand: a case study of word deafness with reference to the role of time in auditory
comprehension. Brain. 1 97 4;97 :37 3–84. [PubMed: 44341 84]
2. Albert M., Sparks R., v on Stockert T., Sax D. A case study of auditory agnosia: Linguistic and nonlinguistic
processing. Cortex . 1 97 2;8:427 –33. [PubMed: 467 1 67 4]
3. Allison T., McCarthy G., Nobre A., Puce A., Belger A. Human ex trastriate v isual cortex and the perception of faces,
words, numbers, and colors. Cerebral Cortex . 1 994;4:544–54. [PubMed: 7 833655]
4. Allport D. Newman S.K., Epstein R. Current Perspectiv es in Dy sphasia. Edinburgh: Churchill Liv ingstone; 1 985.
Distributed memory , modular subsy stems and dy sphasia; pp. 32–60.
5. Auerbach S., Allard T., Naeser M., Alex ander M., Albert M. Pure word deafness Analy sis of a case with bilateral
lesions and a defect at the prephonemic lev el. Brain. 1 982;1 05:27 1 –300. [PubMed: 7 082991 ]
6. Balint R. Seelenlahmung des “Schauens”, optische Atax ie, raumliche Storung der Aufmerksamkeit. Monatschrift fur
Psy chiatrie undNeurologie. 1 909;25:51 –81 .
7 . Barsalou L. Grounded cognition. Annual Rev iew of Psy chology . 2008;59:61 7 –45. [PubMed: 1 7 7 05682]
8. Bauer R. Autonomic recognition of names and faces in prosopagnosia: A neuropsy chological application of the
Guilty Knowledge Test. Neuropsy chologia. 1 984;22:457 –69. [PubMed: 64831 7 2]
9. Bauer R. Agnosia. In: V alenstein K.H.a.E., editor. Clinical Neuropsy chology . Ox ford: Ox ford Univ ersity Press;
2003. pp. 236–95.
1 0. Bauer R., V erfaellie M. Electrodermal discrimination of familiar but not unfamiliar faces in prosopagnosia. Brain
Cognition. 1 986;8:240–52. [PubMed: 31 96486]
1 1 . Bay E. 1 944. Zum problem der taktilen Agnosie Deutsche Zeitschrift für Nerv enheilkunde 1 56 1 3; 64. 96 .
1 2. Bay E. Disturbances of v isual perception and their ex amination. Brain. 1 953;7 6:51 5–50. [PubMed: 1 31 1 557 2]
1 3. Bender M., Feldman M. The so called “v isual agnosias” Brain. 1 97 2;95:1 7 3–86. [PubMed: 5023087 ]
1 4. Benson D., Dav is J., Sny der B. Posterior cortical atrophy . Archiv es of Neurology . 1 988;45:7 89–93. [PubMed:
3390033]
1 5. Benson D., Greenberg J. V isual form agnosia. A specific defect in v isual discrimination. Archiv es of Neurology .
1 969;20:82–89. [PubMed: 4303441 ]
1 6. Binkofski F., Buccino G., Posse S., Seitz R., Rizzolatti G., Freund H. A fronto-parietal circuit for object manipulation
in man: Ev idence from an fMRI-study . European Journal of Neuroscience. 1 999;1 1 :327 6–86. [PubMed: 1 051 01 91 ]
1 7 . Bodamer J. Die Prosop-Agnosie. Archiv für Psy chiatric und Zeitschrift Neurologic. 1 947 ;1 7 9:6–54. [PubMed:
1 8901 7 47 ]
1 8. Bohlhalter S., Fretz C., Weder B. Hierarchical v ersus parallel processing in tactile object recognition: A behav ioural-
neuroanatomical study of aperceptiv e tactile agnosia. Brain. 2002;1 25:2537 –48. [PubMed: 1 239097 8]
1 9. Borenstein E., Ullman S. Combined top-down/bottom-up segmentation. IEEE Transactions on Pattern Analy sis &
Machine Intelligence. 2008;30:21 09–25. [PubMed: 1 8988946]
20. Bornstein B., Sroka H., Munitz H. Prosopagnosia with animal face agnosia. Cortex . 1 969;5:1 64–69. [PubMed:
5387 907 ]
21 . Botez M., Serbănescu T. Course and outcome of v isual static agnosia. Journal of the Neurological Sciences.
1 967 ;4:289–97 . [PubMed: 6037 098]
22. Bruy er R. Cov ert face recognition in prosopagnosia: A rev iew. Brain Cognition. 1 991 ;1 5:223–35. [PubMed:
2043365]
23. Bruy er R., Laterre C., Seron X., Fey ereisen P., Stry pstein E., Pierrard E., Rectem D. A case of prosopagnosia with
some preserv ed cov ert remembrance of familiar faces. Brain Cognition. 1 983;2:257 –84. [PubMed: 6546027 ]
24. Bux baum L., Glosser G., Coslett H. Relativ e sparing of object recognition in alex ia-prosopagnosia. Brain and
Cognition. 1 996;32:202–5.
25. Caramazza A., Shelton J. Domain-specific knowledge sy stems in the brain the animate-inanimate distinction.
Journal of Cognitiv e Neuroscience. 1 998;1 0:1 –34. [PubMed: 9526080]
26. Caselli R. Bilateral impairment of somesthetically mediated object recognition in humans. May o Clinic Proceedings.
1 991 ;66:357 –64. [PubMed: 201 3986]
27 . Caselli R. Feinberg T. E., Farah M.J. Behav ioral Neurology andNeuropsy chology . New Y ork: McGraw-Hill; 1 997 .
Tactile agnosia and disorders of tactile perception; pp. 27 7 –88.
28. Cav e K. The Feature Gate model of v isual selection. Psy chological Research. 1 999;62:1 82–94. [PubMed: 1 0490397 ]
29. Celesia G. Organization of auditory cortical areas in man. Brain. 1 97 6;99:403–1 4. [PubMed: 1 00027 9]
30. Chao L., Hax by J., Martin A. Attribute-based neural substrates in temporal cortex for perceiv ing and knowing
about objects. Nature Neuroscience. 1 999;2:91 3–1 9. [PubMed: 1 0491 61 3]
31 . Clarke S., Bellmanm A., Meuli R., Assal G., Steck A. Auditory agnosia and auditory spatial deficits following left
hemispheric lesions: Ev idence for distinct processing pathway s. Neuropsy chologia. 2000;38:7 97 –807 . [PubMed:
1 0689055]
32. Corbetta M. Frontoparietal cortical networks for directing attention and the ey e to v isual locations: Identical,
independent, or ov erlapping neural sy stems. Proceedings of the National Academy of Science. 1 998;95:831 –38.
[PMC free article: PMC33805] [PubMed: 9448248]
33. Corkin S. The role of different cerebral structures in somesthetic perception. In: Handbook of Perception. New
Y ork: Academic Press; 1 97 8. pp. 1 05–55.
34. Corkin S., Milner B., Rasmussen T. Somatosensory thresholds: Contrasting effects of postcentral-gy rus and
posterior parietal-lobe ex cision. Archiv es of Neurology . 1 97 0;23:41 –58. [PubMed: 4987 1 42]
35. Coslett H., Brashear H., Heilman K. Pure word deafness after bilateral primary auditory cortex infarcts. Neurology .
1 984;34:347 –52. [PubMed: 653827 4]
36. Coslett H., Chatterjee A. Balint’s sy ndrome and related disorders. Feinberg T., Farah M. McGraw-Hill; Behav ioral
Neurology and Neuropsy chology . 2003:325–36.
37 . Coslett H., Lie E. Simultanagnosia: Effects of semantic category and repetition blindness. Neuropsy chologia.
2008;46:1 853–63. [PMC free article: PMC24947 04] [PubMed: 1 851 467 7 ]
38. Coslett H., Saffran E. Ev idence for preserv ed reading in “pure alex ia” Brain. 1 989;1 1 2:327 –59. [PubMed: 27 06436]
39. Coslett H., Saffran E. Simultanagnosia. To see but not two see. Brain. 1 993;1 1 4:1 523–45. [PubMed: 1 8841 65]
40. Coslett H., Saffran E., Greenbaum S., Schwartz J. Reading in pure alex ia. The effect of strategy . Brain. 1 993;1 1 6:21 –
37 . [PubMed: 8453458]
41 . Coslett H., Stark M., Rajaram S., Saffran E. Narrowing the spotlight: A v isual attentional disorder in presumed
alzheimer’s disease. Neurocase. 1 995;1 :305–1 8.
42. Damasio A. Time-locked multiregional retroactiv ation: A sy stem lev el proposal for the neural substrates of recall
and recognition. Cognition. 1 989;33:25–62. [PubMed: 2691 1 84]
43. Damasio A., Damasio H., Hoesen G.V . Prosopagnosia: Anatomic basis and behav ioral mechanisms. Neurology .
1 982;32:331 –41 . [PubMed: 7 1 99655]
44. Damasio A., Y amada T., Damasio H., Corbet J., McKee J. Central achromatopsia: Behav ioral, anatomic, and
phy siologic aspects. Neurology . 1 980;30:1 064–7 1 . [PubMed: 696841 9]
45. Dav idoff J., Wilson B. A case of v isual agnosia showing a disorder of pre-semantic v isual classification. Cortex .
1 985;21 :1 21 –34. [PubMed: 3987 306]
46. DeHaan E., Y oung A., Newcombe F. Faces interfere with name classification in a prosopagnosic patient. Cortex .
1 987 ;23:309–1 6. [PubMed: 3608524]
47 . Dejerine J. Contribution a l’etude anatomo-pathologique et clinique des differentes v arietes de cecite v erbale.
Memoires de la Societe de Biologie. 1 892;4:61 –90.
48. Delay J. Clinique, Phy siologie, Topographie. Paris: Masson; 1 935. Les Astereognosies. Pathologie due Toucher.
49. Denes G., Semenza C. Auditory modality -specific anomia: Ev idence from a case of pure word deafness. Cortex .
1 97 5;1 1 :401 –1 1 . [PubMed: 1 222584]
50. DeRenzi E. Prosopagnosia in two patients with CT scan ev idence of damage confined to the right hemisphere.
Neuropsy chologia. 1 986;24:1 7 1 –80. [PubMed: 37 36820]
51 . Dijksterhuis A. Think different: The merits of unconscious thought in preference dev elopment and decision making.
Journal of Personality and Social Psy chology . 2004;87 :586–98. [PubMed: 1 55357 7 3]
52. Dijksterhuis A., Bos M., Nordgren L., Baaren R.v . On making the right choice: The deliberation-without-attention
effect. Science. 2006;31 1 :1 005–7 . [PubMed: 1 6484496]
53. Duncan J. Selectiv e attention and the organization of v isual information. Journal of Ex perimental Psy chology :
General. 1 984;1 1 3:501 –1 7 . [PubMed: 6240521 ]
54. Efron R. What is perception? Boston Studies in Philosophy of Science. 1 968;4:1 37 –7 3.
55. Eichenbaum H., Morton T., Potter H., Corkin S. Selectiv e olfactory deficits in case H.M. Brain. 1 983;1 06:459–7 2.
[PubMed: 685027 8]
56. Ellis A., Y oung A. Human Cognitiv e Neuropsy chology . Hillsdale, NJ: Lawrence Erlbaum Associates; 1 988.
57 . Epstein R., Kanwisher N. A cortical representation of the local v isual env ironment. Nature. 1 998;392:598–601 .
[PubMed: 95601 55]
58. Eriksen C., Hoffman J. The ex tent of processing noise elements during selectiv e encoding from v isual display s.
Perception and Psy chophy sics. 1 97 3;1 2:201 –4.
59. Farah M.J. The Cognitiv e Neuroscience of V ision. Malden: Blackwell Publishers; 2000.
60. Farah M.J. V isual Agnosia. Second. Cambridge, MA: MIT Press; 2004.
61 . Finkelnburg F.C. Niederrheinische Gesellschaft in Bonn. Medicinische Section. Berliner klinische Wochenschrift.
1 87 0:449–50. 460–61 .
62. Fredericks J.A.M. The agnosias. In: Handbook of Clinical Neurology . Amsterdam: Nother-Holland; 1 969.
63. Freud S. Zur Auffasun der Aphasien Eine Kritische Studie. V ienna: Franz Deuticke; 1 891 .
64. Fujii T., Fukatsu R., Watabe S., Ohnuma A., Teramura K., Kimura I., Saso S., Kogure K. Auditory sound agnosia
without aphasia following a right temporal lobe lesion. Cortex . 1 990;26:263–68. [PubMed: 2082956]
65. Gauthier I., Tarr M., Moy lan J. The fusiform “face area” is part of a network that processes faces at the indiv idual
lev el. Journal of Cognitiv e Neuroscience. 2000;1 2:495–504. [PubMed: 1 0931 7 7 4]
66. Goldstein K. Some remarks on Russell Brain’s article concerning v isual object-agnosia. Journal of Nerv ous and
Mental Disease. 1 943;98:1 48–53.
67 . Goldstein K., Gelb A. Psy chologische Analy sen hirnpathologischer Falle auf Grund v on Untersuchungen
Hirnv erletzter. Zeitschrift fur die Neurologie and Psy chiatrie. 1 91 8;41 :1 –42.
68. Gottfried J. Central nerv ous processing. Adv ances in Oto-rhino-lary ngology . 2006;63:44–69. [PubMed: 1 67 33332]
69. Gottfried J., Zald D. On the scent of human olfactory orbitofrontal cortex : Meta-analy sis and comparison to non-
human primates. Brain Research Rev iews. 2005;50:287 –304. [PubMed: 1 621 3593]
7 0. Griffiths T., Elliot C., Coulthard A., Cartlidge N., Green G. A distinct low-lev el mechanism for interaural timing
analy sis in human hearing. Neuroreport. 1 998;9:3383–86. [PubMed: 9855285]
7 1 . Grill-Spector K., Knouf N., Kanwisher N. The fusiform face area subserv es face perception, not generic within-
category identification. Nature Neuroscience. 2004;7 :555–62. [PubMed: 1 507 7 1 1 2]
7 2. Harris A., Aguirre G. The representation of parts and wholes in face-selectiv e cortex . Journal of Cognitiv e
Neuroscience. 2008;20:863–7 8. [PubMed: 1 8201 1 34]
7 3. Hax by J., Hoffman E., Gobbini M. The distributed human neural sy stem for face perception. Trends in Cognitiv e
Sciences. 2000;4:223–33. [PubMed: 1 0827 445]
7 4. Head H. Sensation and the cerebral cortex . Brain. 1 91 8;41 :57 –253.
7 5. Head H., Holmes G. Sensory disturbances from cerebral lesions. Brain. 1 91 1 ;34:1 02–254.
7 6. Hecaen H. Introduction a la Neuropsy chologie. Paris: Larousse; 1 97 2.
7 7 . Hecaen H., Dav id M. Sy ndrome parietale traumatique: Asy mbolie tactile et hemiasomatognosie parox y stique et
douloureuse. Rev ue Neurologique. 1 945;7 7 :1 1 3–23.
7 8. Heinke D., Humphrey s G. Attention, spatial representation, and v isual neglect: Simulating emergent attention and
spatial memory in the selectiv e attention for identification model SAIM. Psy chological Rev iew. 2003;1 1 0:29–87 .
[PubMed: 1 2529057 ]
7 9. Holmes G. Disturbances of v isual orientation. British Journal of Ophthalmology . 1 91 8;2:449–68. [PMC free article:
PMC51 3529] [PubMed: 1 81 67 81 6]
80. Horel J., Keating E. Recov ery from a partial Kluv er-Bucy sy ndrome induced by disconnection. Journal of
Comparativ e and Phy siological Psy chology . 1 97 2;7 9:1 05–1 4. [PubMed: 4623001 ]
81 . Horner J., Massey E. Dy namic spelling alex ia. Journal of Neurology , Neurosurgery & Psy chiatry . 1 986;49:455–57 .
[PMC free article: PMC1 0287 7 8] [PubMed: 37 01 357 ]
82. Humphrey s G., Riddoch M. Routes to object constancy : Implications from neurological impairments of object
constancy . Quarterly Journal of Ex perimental Psy chology . 1 984;36A:385–41 5. [PubMed: 6533691 ]
83. Humphrey s G., Riddoch M. The fractionation of v isual agnosia. In: V isual Object Processing: A Cognitiv e
Neuropsy chological Approach. London: Lawrence Erlbaum Associates; 1 987 .
84. Humphrey s G., Riddoch M., Fortt H. Action relations, semantic relations, and familiarity of spatial position in
Balint’s sy ndrome: Crossov er effects on perceptual report and on localization. Cognitiv e, Affectiv e, & Behav ioral
Neuroscience. 2006;6:236–45. [PubMed: 1 7 243359]
85. Jackson J. Case of large cerebral tumor without optic neuritis and with left hemiplegia and imperception. Roy al
London Ophthalmic Hospital Reports. 1 87 6;8:434–44.
86. Jackson J. Selected Writings of John Hughlings Jackson. London: Hodder and Stoughton; 1 932.
87 . Jones-Gotman M., Rouleau I., Sny der P. Clinical and research contributions of the intracarotid amobarbital
procedure to neuropsy chology . Brain and Cognition. 1 997 ;33:1 –6. [PubMed: 905627 0]
88. Kable J., Lease-Spellmey er J., Chatterjee A. Neural substrates of action ev ent knowledge. Journal of Cognitiv e
Neuroscience. 2002;1 4:7 95–805. [PubMed: 1 21 67 263]
89. Kaga K., Kurauchi T., Nakamura M., Shindo M., Ishii K. Magnetoencephalography and positron emission
tomography studies of a patient with auditory agnosia caused by bilateral lesions confined to the auditory
radiations. Acta Oto-lary ngologica. 2005;1 25:1 351 –55. [PubMed: 1 6303686]
90. Kanwisher N., McDermott J., Chun M. The fusiform face area: A module in human ex trastriate cortex specialized for
face perception. Journal of Neuroscience. 1 997 ;1 7 :4302–1 1 . [PubMed: 91 51 7 47 ]
91 . Kazui S., Naritomi H., Sawada T., Inoue N., Okuda J. Subcortical auditory agnosia. Brain and Language.
1 990;38:47 6–87 . [PubMed: 237 597 7 ]
92. Kleist K. Gehirnpathologische und lokalisatorische Ergebnisse über Hörstörungen, Geräuschtaubheiten und
Amusien. Monatsschrift für Psy chiatrie und Neurologie. 1 928;68:853–60.
93. Ledberg A., O’Sulliv an B., Kinomura S., Roland P. Somatosensory activ ations of the parietal operculum of man. A
PET Study . European Journal of Neuroscience. 1 995;7 :1 934–41 . [PubMed: 8528469]
94. Leicester J. Central deafness and subcortical motor aphasia. Brain and Language. 1 980;1 0:224–42. [PubMed:
7 407 544]
95. Lev ine D. Prosopagnosia and v isual object agnosia: A behav ioral study . Brain and Language. 1 97 8;5:341 –65.
[PubMed: 656903]
96. Lev ine D., Calv anio R. Prosopagnosia: A defect in v isual configural processing. Brain and Cognition. 1 989;1 0:1 49–
7 0. [PubMed: 27 57 824]
97 . Lewis J., Wightman F., Brefczy nski J., Phinney R., Binder J., DeY oe E. Human brain regions inv olv ed in recognizing
env ironmental sounds. Cerebral Cortex . 2004;1 4:1 008–21 . [PubMed: 1 51 66097 ]
98. Lissauer H. Ein Fall v on Seelenblindheit Nebst Einem Beitrage zur Theori derselben. Archiv fur Psy chiatrie und
Nerv enkrankheiten. 1 890;21 :222–7 0.
99. Luria A. Disorders of “simultaneous perception” in a case of bilateral occipito-parietal brain injury . Brain.
1 959;83:437 –49. [PubMed: 1 441 9007 ]
1 00. Malach R., Lev y I., Hasson U. The topography of high-order human object areas. Trends in Cognitiv e Sciences.
2002;6:1 7 6–84. [PubMed: 1 1 91 2041 ]
1 01 . Malach R., Reppas J., Benson R., Kwong K., Jiang H., Kennedy W., Ledden P., Brady T., Rosen B., Tootell R. Object-
related activ ity rev ealed by functional magnetic resonance imaging in human occipital cortex . Proceedings of the
National Academy of Sciences. 1 995;92:81 35–39. [PMC free article: PMC41 1 1 0] [PubMed: 7 667 258]
1 02. Marr D. V ision. A Computational Inv estigation into the Human Representation and Processing of V isual
Information. New Y ork: Freeman W.H.; 1 982.
1 03. Mauguiere F., Isnard J. Tactile agnosia and dy sfunction of the primary somatosensory area Data of the study by
somatosensory ev oked potentials in patients with deficits of tactile object recognition. Rev ue Neurologique.
1 995;1 51 :51 8–27 . [PubMed: 857 807 3]
1 04. McGlinchey -Berroth R., Bullis D.P., Milberg W.P., V erfaellie M., Alex ander M., D’Esposito M. Assessment of neglect
rev eals dissociable behav ioral but not neuroanatomic subty pes. Journal of the International Neurospy chological
Society . 1 996;2:441 –51 . [PubMed: 937 51 69]
1 05. McKone E., Kanwisher N. Does the human brain process objects-of-ex pertise like faces: An update. In: From
Monkey Brain to Human Brain. Cambridge: MA: MIT Press; 2004.
1 06. Mendez M., Geehan G. Cortical auditory disorders: Clinical and psy choacoustic features. Journal of Neurology ,
Neurosurgery & Psy chiatry . 1 988;51 :1 –9. [PMC free article: PMC1 0327 06] [PubMed: 2450968]
1 07 . Mendez M., Ghajarnia M. Agnosia for familiar faces and odors in a patient with right temporal lobe dy sfunction.
Neurology . 2001 ;57 :51 9–21 . [PubMed: 1 1 502924]
1 08. Michel F., Pernin M., Sieroff E. Porospagnosie sans hemianopsie apres lesion unilateralie occipito-temporale driote.
Rev ue Neurologique. 1 986;1 42:545–49. [PubMed: 37 87 058]
1 09. Michel J., Peronnet F., Schott B. A case of cortical deafness: Clinical and electrophy siological data. Brain Lang.
1 980;1 0:367 –7 7 . [PubMed: 7 407 553]
1 1 0. Milner A., Goodale M. The V isual Brain in Action. Ox ford: Ox ford Univ ersity Press; 1 995.
1 1 1 . Milner A., Hey wood C. A disorder of lightness discrimination in a case of v isual form agnosia. Cortex . 1 989;25:489–
94. [PubMed: 28057 34]
1 1 2. Moran J., Desimone R. Selectiv e attention gates v isual processing in ex trastriate cortex . Science. 1 985;229:7 82–
84. [PubMed: 40237 1 3]
1 1 3. Moscov itch M., Winocur G., Behrmann N. What is special about face recognition? Nineteen ex periments on a person
with v isual object agnosia and dy slex ia but normal face recognition. Journal of Cognitiv e Neuroscience.
1 997 ;9:555–604.
1 1 4. Munk H. Ueber die Functionen der Grosshirnrinde. Gesammelte Mittheilungenaus den lahren. Berlin: Hirschwald;
1 881 .
1 1 5. Neff W. Neuronal mechanisms of auditory discrimination. In: Sensory Communication. Cambridge, MA: MIT Press;
1 961 .
1 1 6. Newcombe F., Ratcliff G. Agnosia: A disorder of object recognition. In: Les Sy ndromes de Disconnex ion Calleuse
chez L’homme. Ly on: Colloque International de Ly on; 1 97 4.
1 1 7 . Pav ese A., Coslett H., Saffran E., Bux baum L. Limitations of attentional orienting. Effects of abrupt v isual onsets and
offsets on naming two objects in a patient with simultanagnosia. Neuropsy chologia. 2002;40:1 097 –1 1 03.
[PubMed: 1 1 9007 60]
1 1 8. Pav lov I.P. Conditioned Reflex es; An Inv estigation of the Phy siological Activ ity of the Cerebral Cortex . Ox ford:
Humphrey Milford; 1 927 .
1 1 9. Pearlman A., Birch J., Meadows J. Cerebral color blindness: An acquired defect in hue discrimination. Annals of
Neurology . 1 97 9;5:253–61 . [PubMed: 31 261 9]
1 20. Platz T. Tactile agnosia: Casuistic ev idence and theoretical remarks on modality -specific meaning representations
and sensorimotor integration. Brain. 1 996;1 1 9:1 565–7 4. [PubMed: 8931 580]
1 21 . Polk T., Stallcup M., Aguirre G.K., Alsop D., D’Esposito M., Detre J., Farah M. Neural specialization for letter
recognition. Journal of Cognitiv e Neuroscience. 2002;1 4:1 45–59. [PubMed: 1 1 97 07 82]
1 22. Poppelreuter W. Zur Psy chologie und Pathologie der optischen Wahrehmung. Zur Gesamtgebiete der Neurologie
und Psy chiatrie. 1 923;83:26–1 52.
1 23. Posner M. Orienting of attention. Quarterly Journal of Ex perimental Psy chology . 1 980;32:3–25. [PubMed:
7 367 57 7 ]
1 24. Potzl O., Redlich E. Demonstration eines Falles v on bilateraler Affektion beider Occipitallapen. Weiner Klinsche
Wochenschrift. 1 91 1 ;24:51 7 –1 8.
1 25. Price C., Dev lin J. The my th of the v isual word form area. Neuroimage. 2003;1 9:47 3–81 . [PubMed: 1 28807 81 ]
1 26. Prinzmetal W., Iv ry R., Beck D., Shimizu N. A measurement theory of illusory conjunctions. Journal of
Ex perimental Psy chology : Human Perception and Performance. 2002;28:251 –69. [PubMed: 1 1 999853]
1 27 . Ratcliff G., Newcombe F. Object recognition: Some deductions from the clinical ev idence. In: Normality and
Pathology in Cognitiv e Functions. New Y ork: Academic Press; 1 982.
1 28. Reed C., Caselli R., Farah M. Tactile agnosia. Underly ing impairment and implications for normal tactile object
recognition. Brain. 1 996;1 1 9:87 5–88. [PubMed: 867 3499]
1 29. Riddoch M., Humphrey s G. A case of integrativ e v isual agnosia. Brain. 1 987 ;1 1 0:1 431 –62. [PubMed: 3427 396]
1 30. Riddoch M., Humphrey s G. V isual agnosia. Clinical Neurology . 2003;21 :501 –20. [PubMed: 1 291 6489]
1 31 . Rizzo M., Hurtig R. Looking but not seeing: Attention, perception, and ey e mov ements in simultanagnosia.
Neurology . 1 987 ;37 :1 642–48. [PubMed: 36581 7 2]
1 32. Robertson L., Treisman A., Freidman-Hill S., Grabowecky M. The interaction of spatial and object pathway s:
Ev idence from Balint’s Sy ndrome. Journal of Cognitiv e Neuroscience. 1 997 ;9:295–31 7 .
1 33. Rogers T., Lambon R., Garrard P., Bozeat S., McClelland J., Hodges J., Patterson K. Structure and deterioration of
semantic memory : A neuropsy chological and computational inv estigation. Psy chological Rev iew. 2004;1 1 1 :205–
35. [PubMed: 1 47 56594]
1 34. Roland P., Sulliv an B., Kawashima R. Shape and roughness activ ate different somatosensory areas in the human
brain. Proceedings of the National Academy of Sciences. 1 998;95:3295–3300. [PMC free article: PMC1 97 35]
[PubMed: 9501 256]
1 35. Rubens A., Benson D. Associativ e v isual agnosia. Archiv es of Neurology . 1 97 1 ;24:304–1 6. [PubMed: 5548450]
1 36. Saffran E., Coslett H. Attentional dy slex ia in Alzheimer’s disease: A case study . Cognitiv e Neuropsy chology .
1 996;1 3:205–28.
1 37 . Saffran E., Fitzpatrick-DeSalme E., Coslett H. V isual disturbances in dementia. In: Modular Deficits in Alzheimer-
ty pe Dementia. Cambridge, MA: MIT Press; 1 990. pp. 297 –327 .
1 38. Saffran E., Schwartz M. Impairments of sentence comprehension. Philosophical Transactions of the Roy al Society B:
Biological Sciences. 1 994;346:47 –53. [PubMed: 7 8861 52]
1 39. Saffran E., Schwartz M., Marin O. An analy sis of speech perception in word deafness. Brain and Language.
1 97 6;3:209–28. [PubMed: 938928]
1 40. Say gin A., Dick F., Wilson S., Dronkers N., Bates E. Neural resources for processing language and env ironmental
sounds. Ev idence from aphasia. Brain. 2003;1 26:928–45. [PubMed: 1 261 5649]
1 41 . Schacter D. Memory , amnesia, and frontal lobe dy sfunction. Psy chobiology . 1 987 ;1 5:21 –36.
1 42. Schnider A., Benson D., Scharre D. V isual agnosia and optic aphasia: Are they anatomically distinct? Cortex .
1 994;30:445–57 . [PubMed: 7 805385]
1 43. Simmons W., Barsalou L. The similarity -in-topography principle: Reconciling theories of conceptual deficits.
Cognitiv e Neuropsy chology . 2003;20:451 –86. [PubMed: 20957 580]
1 44. Small D. Gustatory processing in humans. Adv ances in Oto-rhino-lary ngology . 2006;63:1 91 –220. [PubMed:
1 67 33340]
1 45. Small D., Bernasconi N., Bernasconi A., Sziklas V ., Jones-Gotman M. Gustatory agnosia. Neurology . 2005;64:31 1 –
1 7 . [PubMed: 1 5668430]
1 46. Small D., Jones-Gotman M., Zatorre R., Petrides M., Ev ans A. A role for the right anterior temporal lobe in taste
quality recognition. Journal of Neuroscience. 1 997 ;1 7 :51 36–42. [PubMed: 91 85551 ]
1 47 . Spreen O., Benton A., Finchman R. Auditory agnosia without aphasia. Archiv es of Neurology . 1 965;1 3:84–92.
[PubMed: 1 431 427 9]
1 48. Tarr M., Gauthier I. FFA: A flex ible fusiform area for subordinate-lev el v isual processing automatized by ex pertise.
Nature Neuroscience. 2000;3:7 64–69. [PubMed: 1 0903568]
1 49. Tranel D., Damasio A. Knowledge without awareness: An autonomic index of facial recognition by prosopagnosics.
Science. 1 985;228:1 453–54. [PubMed: 401 2303]
1 50. Treisman A., Gelade G. A Feature-Integration Theory of Attention. Cognitiv e Psy chology . 1 980;1 2:97 –1 36.
[PubMed: 7 351 1 25]
1 51 . Treisman A., Souther J. Search asy mmetry : A diagnostic for preattentiv e processing of separable features. Journal
of Ex perimental Psy chology : General. 1 985;1 1 4:285–31 0. [PubMed: 31 61 97 8]
1 52. Ullman S. V isual routines. Cognition. 1 984;1 8:97 –1 59. [PubMed: 65431 65]
1 53. Ungerleider L., Mishkin M. Two cortical v isual sy stems. In: Analy sis of V isual Behav ior. Cambridge: MIT Press;
1 982. pp. 549–86.
1 54. V ecera S., Farah M. Does v isual attention select objects or locations? Journal of Ex perimental Psy chology : General.
1 994;1 23:1 46–60. [PubMed: 801 461 0]
1 55. V errey . He miachromatopsie droite absolue. Arch Ophthalmol. 1 888;1 4:422–34. Paris.
1 56. V idy asagar T. Gating of neuronal responses in macaque primary v isual cortex by an attentional spotlight.
Neuroreport. 1 998;9:1 947 –52. [PubMed: 967 457 2]
1 57 . V ignolo L. Auditory agnosia. Philosophical Transactions of The Roy al Society London B Biological Science.
1 982;298:49–57 . [PubMed: 61 2597 5]
1 58. V ignolo L. Music agnosia and auditory agnosia. Dissociations in stroke patients. Annals of the New Y ork Academy
of Sciences. 2003;999:50–57 . [PubMed: 1 4681 1 1 7 ]
1 59. V on Frey M. Beitrage zur Sinnesphy siologie des Haut. Bericht uber die V erhandlungen der Sdchsischen Akademie
der Wissenschaften zu Leipzig, Math.-Phy s. Kl. 1 895;47 :1 66–84.
1 60. Warrington E., James M. V isual object recognition in patients with right-hemisphere lesions: Ax es or features?
Perception. 1 986;1 5:355–66. [PubMed: 37 97 207 ]
1 61 . Warrington E., Shallice T. Category specific semantic impairments. Brain. 1 984;1 07 :829–54. [PubMed: 620691 0]
1 62. Warrington E., Tay lor A. Immediate memory for faces: Long- or short-term memory ? Quarterly Journal of
Ex perimental Psy chology . 1 97 3;25:31 6–22. [PubMed: 4807 292]
1 63. Warrington E., Tay lor A. Two categorical stages in object recognition. Perception. 1 97 8;7 :695–7 05. [PubMed:
7 4051 0]
1 64. Watson J., My ers R., Frackowiak R., Hanjal J., Woods R., Mazziotta J., Shipp S., Zeki S. Area V 5 of the human brain:
Ev idence from a combined study using positron emission tomography and magnetic resonance imaging. Cerebral
Cortex . 1 993;3:7 9–94. [PubMed: 8490322]
1 65. Wernicke C. 1 895. Zwei Falle v on Rindenlasion Arbeiten Aus die Psy chiatrische Klinic In Breslau 2:35.
1 66. Wolpert I. Die Simultanagnosie: Storung der gesamtauffassung. Zeitschrift fur die gesamte Neurologie und
Psy chiatrie. 1 924;93:397 –41 3.
1 67 . Y oung A., Ellis A., Flude B., McWeeny K., Hay D. Face-name interference. Journal of Ex perimental Psy chology :
Human Perception and Performance. 1 986;1 2:466–7 5. [PubMed: 2946803]
1 68. Zihl J., Cramon D.V ., Mai N. Selectiv e disturbance of mov ement v ision after bilateral brain damage. Brain.
1 983;1 06:31 3–40. [PubMed: 685027 2]
Figures
FIGURE 10.1
A copy of a drawing by a v isual agnosic of a carrot with a line drawn through it. The patient failed to recognize the carrot;
reflecting the lack of top-down processing from stored knowledge of the object, his copy incorporated the out of place line.
FIGURE 10.2
A cartoon of the processes inv olv ed in v isual object recognition.
Copy right © 201 1 by Tay lor and Francis Group, LLC.

Book sh elf ID: NBK9 2 8 0 0 PMID: 2 2 5 9 3 9 1 3

1 Sensory Agnosias - Neurobiology of Sensation and Reward Agnosia - NCBI Bookshelf

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 Sensory Agnosias - Neurobiology of Sensation and Reward Agnosia - NCBI Bookshelf

Uploaded by

Copyright:

Available Formats

14/08/13 Sensory Agnosias - Neurobiology of Sensation and Reward - NCBI Bookshelf

Chapter 10 Sensory Agnosias

10.1 INTRODUCTION AND HISTORICAL OVERVIEW

10.2 TYPES OF AGNOSIA

10.2.1 VISUAL AGNOSIA

10.2.1.1 Visual Object Recognition: A T heoretical Ov erv iew

10.2.1.2 Disorders of Low-Lev el Vision

1 0.2.1 .2.1 Achromatopsia

1 0.2.1 .2.2 Impaired Motion Perception (Akinetopsia)

10.2.1.3 Visual Form Agnosia

10.2.2 DISORDERS OF VISUAL ATTENTION

10.2.3 I MPAIRMENTS IN THE VIEW NORMALIZATION SYSTEM

10.3 IMPAIRMENTS OF THE STRUCTURAL DESCRIPTION SYSTEM (“ASSOCIATIVE

10.3.2 AGNOSIA FOR WORDS

10.4 CATEGORY-SPECIFIC RECOGNITION DEFICITS

10.5 FUNCTIONAL IMAGING CORRELATES OF OBJECT RECOGNITION

10.6 AUDITORY AGNOSIAS

10.7 CORTICAL DEAFNESS AND GENERALIZED AUDITORY AGNOSIA

10.7.1 AUDITORY SOUND AGNOSIA (AUDITORY AGNOSIA FOR NON-SPEECH SOUNDS)

10.7.2 PURE WORD DEAFNESS (AUDITORY AGNOSIA FOR SPEECH)

10.8 OTHER AGNOSIAS

10.9 AGNOSIA AND CONSCIOUS AWARENESS

Copy right © 201 1 by Tay lor and Francis Group, LLC.

You might also like