doi:10.1093/brain/awq233
Brain 2010: 133; 3564–3577
| 3564
BRAIN
A JOURNAL OF NEUROLOGY
Implicit awareness in anosognosia for
hemiplegia: unconscious interference without
conscious re-representation
Aikaterini Fotopoulou,1,2 Simone Pernigo,3 Rino Maeda,1 Anthony Rudd4 and
Michael A. Kopelman1
1
2
3
4
Institute of Psychiatry, King’s College London, London SE5 8AF, UK
Institute of Cognitive Neuroscience, University College London, London WC1 N 3AR, UK
Department of Philosophy, Education and Psychology, University of Verona, Verona, Italy
Stroke Unit, Guy’s and St. Thomas’s NHS Foundation Trust, London SE1 7EH, UK
Correspondence to: Aikaterini (Katerina) Fotopoulou, PhD,
Psychology Department at Guy’s,
King’s College London,
Institute of Psychiatry,
5th Floor,
Bermondsey Wing,
Guy’s Campus,
London SE1 9RT, UK
E-mail: [email protected]
Some patients with anosognosia for hemiplegia, i.e. apparent unawareness of hemiplegia, have been clinically observed to
show ‘tacit’ or ‘implicit’ awareness of their deficits. Here we have experimentally examined whether implicit and explicit
responses to the same deficit-related material can dissociate. Fourteen stroke patients with right hemisphere lesions and
contralesional paralysis were tested for implicit and explicit responses to brief sentences with deficit-related themes. These
responses were elicited using: (i) a verbal inhibition test in which patients had to inhibit completing each sentence with an
automatic response (implicit task) and (ii) a rating procedure in which patients rated the self-relevance of the same sentences
(explicit task). A group of anosognosic hemiplegic patients was significantly slower than a control group of aware hemiplegic
patients in performing the inhibition task with deficit-related sentences than with other emotionally negative themes (relative
to neutral themes). This occurred despite their explicit denial of the self-relevance of the former sentences. Individual patient
analysis showed that six of the seven anosognosic patients significantly differed from the control group in this dissociation.
Using lesion mapping procedures, we found that the lesions of the anosognosic patients differed from those of the ‘aware’
controls mainly by involving the anterior parts of the insula, inferior motor areas, basal ganglia structures, limbic structures and
deep white matter. In contrast, the anosognosic patient without implicit awareness had more cortical lesions, mostly in frontal
areas, including lateral premotor regions, and also in the parietal and occipital lobes. These results provide strong experimental
support for a specific dissociation between implicit and explicit awareness of deficits. More generally, the combination of our
behavioural and neural findings suggests that an explicit, affectively personalized sensorimotor awareness requires
the re-representation of sensorimotor information in the insular cortex, with possible involvement of limbic areas and
basal ganglia circuits. The delusional features of anosognosia for hemiplegia can be explained as a failure of this
re-representation.
Received May 9, 2010. Revised June 30, 2010. Accepted July 1, 2010. Advance Access publication September 7, 2010
ß The Author (2010). Published by Oxford University Press on behalf of the Guarantors of Brain. All rights reserved.
For Permissions, please email: [email protected]
Implicit awareness in anosognosia for hemiplegia
Brain 2010: 133; 3564–3577
| 3565
Keywords: anosognosia; implicit awareness; insula; basal ganglia; motor awareness
Abbreviation: AHP = anosognosia for hemiplegia
Introduction
Some patients with central neurological damage are apparently
unable to acknowledge their contralesional motor deficits. This
rare neuropsychological symptom is termed anosognosia for hemiplegia (AHP)—a term implying lack of ‘knowledge’ of the deficit
(Babinski, 1914). In extreme cases, patients appear to believe that
their limbs are completely functional. Some may even claim that
they do move on request, while in reality their paralysed limbs
remain motionless (Feinberg et al., 2000; Fotopoulou et al.,
2008). Other patients may provide excuses (confabulations) to
account for the fact that the limbs are not moving (for review,
see Fotopoulou, 2010); still others may admit they cannot move
their limbs, but they show complete indifference towards their
deficit (anosodiaphoria) (Critchley, 1953). AHP can be accompanied by other delusional beliefs. For example, patients may
claim that their paralysed limb belongs to someone else (somatoparaphrenia) (Gerstmann, 1942). AHP occurs more frequently
following right than left hemisphere brain damage (but see
Cocchini et al., 2009). No available treatment exists for AHP,
although temporary remission has been reported following vestibular stimulation (Rubens, 1985); and permanent reinstatement
of awareness was recently reported in a patient with severe
AHP, following self-observation during video replay (Fotopoulou
et al., 2009).
AHP has been viewed by some as a secondary consequence of
sensory feedback deficits, such as spatial or personal neglect
(Cutting, 1978; Levine, 1990). Others have proposed that AHP
might result from the combination of sensory deficits and
higher-order cognitive deficits such as confusion, confabulation
and memory impairments (Levine, 1991; Berti et al., 1996).
However, double dissociations have been demonstrated between
AHP and each of the aforementioned deficits (Bisiach et al., 1986;
Starkstein et al., 1992; Heilman et al., 1998; Berti et al., 2005;
Coslett, 2005; see also Heilman and Harciarek, 2010 for review).
Moreover, Marcel et al. (2004) found that neither sensory loss,
nor intellectual impairment, nor their combination, provides a
complete explanation of AHP.
Alternatively, AHP has been explained as specific deficit of
motor planning (Heilman et al., 1998; Frith et al., 2000; Berti
et al., 2005). A recent study showed that altered awareness of
action in AHP reflects an abnormal dominance of representations
of intended movements over sensory feedback about the actual
effects of movement (Fotopoulou et al., 2008). However, such a
finding cannot explain why some patients deny their motor deficits
even when not attempting to move, or why other patients may
selectively forget instances of detected motor failure and even
related injuries. Some authors suggest that the delusional features
of AHP are the result of a psychological defence against excessive
depression or anxiety (Weinstein and Kahn, 1955, but see Bisiach
and Geminiani, 1991; Heilman et al., 1998).
Others have argued that at least some features of AHP cannot
be explained by disruption of sensorimotor mechanisms and that
neuromotivational factors need to be taken into account (Feinberg
et al., 1994; Ramachandran, 1995; Solms, 1995; Frith et al., 2000;
Marcel et al., 2004; Vuilleumier, 2004; Turnbull et al., 2005;
Fotopoulou, 2010). Thus, AHP can be seen as a multifaceted
phenomenon, with each facet perhaps requiring a separate
explanation (Marcel et al., 2004; Vuileumier, 2004; Heilman and
Harciarek, 2010). In addition, AHP has been observed to take
different forms in different patients (e.g. Marcel et al., 2004;
Orfei et al., 2007), suggesting that different types of patients
with AHP exist, and this heterogeneity must be taken into account
both when classifying patients and when attempting to explain the
mechanisms underlying the various clinical manifestations of the
phenomenon (for discussions, see Karnath and Baier, 2010; Vocat
and Vuileumier, 2010).
One important but poorly investigated distinction is between the
explicit and implicit features of AHP (for discussion, see also Vocat
and Vuileumier, 2010). Several investigators have noted that some
patients with AHP show indications of ‘tacit’ or ‘implicit’ awareness
of their deficits, i.e. ‘knowledge that is expressed in task performance unintentionally and with little or no phenomenal awareness’
(Schacter, 1990, p. 157). Thus, while patients may explicitly deny
their paralysis, they may be unconsciously processing some components of their deficits, including the emotional aspects. For example, a severely anosognosic patient who persistently denied her
paralysis and all associated disabilities spontaneously told junior
doctors during a ward round: ‘Perhaps it would be really useful
to you to come and see me at a time when I’ll be really ill and
unable to move’ (Fotopoulou et al., 2009). Another patient with
AHP who was explicitly denying her paralysis unceasingly complained about everyday difficulties with an emotional intensity
that better fitted her devastating disability than these minor everyday disappointments (Fotopoulou and Conway, 2004). However,
beyond the level of single cases or case series (e.g. Ramachandran,
1995; Berti et al., 1998; Nardone et al., 2007), this issue had not
been systematically explored in AHP until recently. In contrast, the
unconscious processing of cognitive and emotional information has
been experimentally investigated and confirmed in other neuropsychological syndromes, including hemispatial neglect (Marshall and
Halligan, 1988), blindsight (Weiskrantz et al., 1974), prosopagnosia
(Tranel and Damasio, 1985) and amnesia (Johnson et al., 1985).
In a recent group study, Cocchini and colleagues (2010) showed
that patients’ behaviour did not always reflect their explicit estimations of their own abilities in a self-report questionnaire.
Specifically, two patients approached a series of bi-manual tasks
as if they could use both hands but did not show unawareness in a
self-report measure, while another eight patients showed the opposite result. The authors interpreted their findings as a double
dissociation between explicit and implicit awareness for motor deficits. However, this study used two different and unrelated tests to
3566
| Brain 2010: 133; 3564–3577
measure the two forms of unawareness. This raises the possibility
that the two tasks were distinct in some components or items,
posing differing challenges for reasons other than variability in
response explicitness.
To conclude reliably that implicit and explicit awareness dissociate within a single patient or across groups, implicit versus explicit
modes of responding need to be assessed using identical, or
well-controlled and balanced, material presented in the same modality. In addition, the material should be compared with other
material of similar emotional value, to control for the potential role
of negative emotions associated with one’s deficits. Finally, independent confirmation of ‘unawareness’ by tests used in previous
studies would provide additional validity to the measures used to
establish unawareness for hemiplegia. To our knowledge, such
comparisons have not yet been undertaken in the study of AHP.
Taking all of this into consideration, the current study aimed to
compare directly the explicit and implicit processing of the same,
emotionally ‘controlled’, experimental material. To this end, we
employed a verbal inhibition test (based on a standardized
verbal equivalent of the Stroop procedure) and examined whether
patients with AHP were slower in performing the inhibition task
when the material of the test included deficit-related information
than when it included other negative or neutral themes. Typically,
selective slowing of responses in emotional Stroop-like tests is
thought to reflect ‘unconscious’ response competition between
the processing of emotional self-threatening information and the
requirement to complete the task (e.g. Pratto and John, 1991;
Wentura et al., 2000). In our study, such selective slowing
would suggest that although patients with AHP show poor explicit
awareness of their hemiplegia, they may be implicitly aware
of it and hence behave accordingly when confronted with
paralysis-related material. Importantly, our task allows the direct
comparison of explicit and implicit processing of the same verbal
experimental material by each patient, and controls for the role of
negative emotions.
Furthermore, using lesion mapping procedures, we aimed to
investigate the neural correlates of unawareness. As in previous
studies of lesion localization in AHP (Ellis and Small, 1997; Berti
et al., 2005; Karnath et al., 2005; Baier and Karnath, 2008; see
also Pia et al., 2004 for an extensive meta-analysis), we performed
lesion comparisons between hemiplegic patients with and without
AHP in order to distinguish between the neural correlates of AHP
and the parietotemporal network commonly associated with
spatial neglect and other common symptoms of right hemisphere
damage. Importantly, previous lesion localization studies have not
taken into account the behavioural dissociation between implicit
and explicit awareness, and thus our study aimed to specify previous findings based on such investigations.
Materials and methods
Subjects
Fourteen adult neurological patients were consecutively recruited from
an acute rehabilitation stroke unit. Berti et al. (1996) have suggested
that given the clinical heterogeneity of AHP, studying cases with denial
A. Fotopoulou et al.
of complete contralateral hemiplegia can generate more reliable findings than investigating patients with mild or moderate hemiparesis.
Accordingly, inclusion criteria were: (i) complete left upper limb hemiplegia (completely paralysed left arm as reported by the responsible
neurologist and confirmed by direct motor impairment examination);
(ii) unilateral right hemisphere lesions (as detected by CT or MRI neuroimaging investigations); and (iii) 54 months from onset. Exclusion
criteria were: (i) previous neurological or psychiatric history; (ii) 57
years of education or an estimated premorbid Full Scale Intelligence
Quotient based on the Wechsler Test of Adult Reading (Wechsler,
2001) 570; (iii) medication with severe cognitive, mood or side-effects; and (iv) severe language impairment (i.e. insufficient communicative ability).
Patients were classified as having AHP on the basis of an open
interview that explored their awareness of motor deficits (Berti et al.
1996). This included (i) a first set of general questions: e.g. ‘Why are
you in the hospital?’; (ii) specific questions about motor ability: e.g.
‘How is your left arm? Can you move it?’; and (iii) confrontation
questions: e.g. ‘Please, touch my hand with your left hand. Have
you done it?’ The total score indicating awareness of upper limb
motor impairment ranged from 0 to 2. A score of 0 was given if
patients acknowledged their motor impairments; a score of 1 was
given if patients did not acknowledge their motor impairments but
recognized not having reached for the examiner’s hand; and a score
of 2 was given if patients denied both motor impairments and the
failure in reaching for the examiner’s hand. Patients were considered
to be anosognosic when they did not acknowledge the contralesional
hemiplegia after repeated and specific questioning by the examiner
(scores 1 and 2). The scale developed by Feinberg et al. (2000) was
used as a second measure of unawareness. This consists of 10 questions, including general open questions, e.g. ‘Do you have weakness
anywhere?’ or ‘Is your arm causing you any problems?’ and confrontation questions, e.g. when left arm is lifted and dropped in right
hemispace: ‘It seems there is some weakness. Do you agree?’ A
score of 0 was given if the patient showed awareness of deficit, 0.5
for partial awareness and 1.0 for complete unawareness or denial.
The study included seven patients with right hemisphere damage
with complete left arm hemiplegia and AHP (the AHP group) and
seven patients with right hemisphere damage with complete left arm
hemiplegia but without AHP (the HP group). Only one patient (AHP5)
showed somatoparaphrenia (she believed her left arm was her
husband’s). There were no indications of disturbed sense of limb
ownership (Bair and Karnath, 2008) in any of the other patients. All
patients showed complete or very severe left leg hemiplegia at the
time of the assessment. Another five patients met the above inclusion
and exclusion criteria but were not recruited due to time restrictions.
All patients underwent neurological and neuropsychological assessment. All participants gave written informed consent according to
the Declaration of Helsinki and the study was approved by the local
trust’s ethical committee.
Neuropsychological assessment
In addition to the anosognosia tests mentioned above, all patients
were assessed using standardized tests. Premorbid intelligence was
assessed using the Wechsler Test of Adult Reading (Wechsler,
2001). Current intelligence was assessed using five Wechsler Adult
Intelligence Scale-III (Wechsler, 1998) subtests: (i) vocabulary; (ii) similarities; (iii) digit span; (iv) arithmetic; and (v) matrix reasoning. Visual
fields were tested with the customary ‘confrontation’ technique
(Bisiach et al., 1986). Unilateral visuospatial neglect was assessed
using the conventional subtests of the Behavioural Inattention Test,
Implicit awareness in anosognosia for hemiplegia
including line crossing, letter cancellation, star cancellation, figure and
shape copying, line-bisection and representational drawing. The ‘one
item test’ (Bisiach et al., 1986) and ‘comb/razor test’ (McIntosh et al.,
2000) were used for the assessment of personal neglect. Three subtests of the Rivermead Assessment of Somatosensory Performance
(Winward et al., 2002) were used for the measurement of sensory
functions ‘surface touch’, ‘tactile extinction’ and ‘proprioception’.
Reasoning abilities were assessed using the Cognitive Estimates Test
(Shallice and Evans, 1978) and the Proverbs subtest of the Delis–
Kaplan Executive Function System (Delis et al., 2001). Inhibition of
automatic responses was assessed with the Hayling Test (Burgess
and Shallice, 1997; see below for detailed description). The Hospital
Anxiety and Depression Scale (Zigmond and Snaith, 1983) was used to
assess depression and anxiety (scores 0–7 = normal range; 8–10 = borderline; 410 = severe levels of anxiety or depression). The Rosenberg
Self-Esteem Scale (Rosenberg, 1965) was used to measure self-esteem
(range 0–30; 15–25 = normal range; 515 = low self-esteem). Patients’
demographic characteristics and their performance on the aforementioned neuropsychological tests are summarized in Table 1.
Independent sample t-tests revealed that the groups did not differ in
age, education or time since onset at assessment. As expected, the
awareness scores of the two groups differed significantly in both interviews used. The two groups did not differ in their general premorbid
or current intelligence scores and there were no significant differences
in their reasoning and inhibition abilities as measured by three executive tasks. However, both groups performed worse than would be
expected on the basis of their premorbid IQ on the Wechsler Adult
Intelligence Scale-III Matrix reasoning subtest and on the Cognitive
Estimates Test, suggesting difficulties in both visual and verbal abstract
thought and problem-solving. Both groups showed basic left visual
and sensory deficits. In addition, both groups showed left visuospatial,
sensory and personal neglect. Interestingly, the latter appeared significantly more severe in the AHP group. The scores of both groups were
considered to be within normal limits for a patient population in a
general hospital on the Hospital Anxiety and Depression Scale, but
patients with AHP showed lower scores (not statistically significant)
for depression and anxiety compared with patients with hemiplegia.
Finally, both groups scored within the normal limits of the general
population in the self-esteem measure and there were no differences
between them.
Experimental investigations
Design and analysis
We used an inhibition task to study implicit processing of
deficit-related material and a rating task to study explicit processing
of the same material. The experiment manipulated the effect of group
(the AHP group versus the HP group) and the effects of emotional
content (‘negative’, ‘neutral’ and ‘deficit-related’ sentences) on explicit
ratings of self-relevance and separately on implicit measures of response latency (reaction times) and the number of suppression errors
on a modified inhibition (Stroop-like) task. As in several studies using
the Stroop procedure with emotional material (e.g. Charash and
McKay, 2002; Kampman et al., 2002), we computed composite measures of interest by subtracting each participant’s mean ratings, mean
response latencies and mean number of errors for ‘neutral’ sentences
from their mean ratings and mean response latencies, and mean
number of errors for ‘negative’ and ‘deficit-related’ sentences, respectively. This design allowed for 2 (AHP group versus HP group) 2
(‘negative’ minus ‘neutral’ versus ‘deficit-related’ minus ‘neutral’ sentences) comparisons.
Brain 2010: 133; 3564–3577
| 3567
In order to explore which and how many of the patients with AHP
differed significantly from controls in their explicit ratings and implicit
responses to deficit-related sentences, we used the Revised
Standardized Difference Test (RSDT; Crawford et al., 1998; revised
in Crawford and Garthwaite, 2005). This is a specialized statistical
test developed for comparing the difference between a patient’s performances on two tasks with the distribution of differences in controls.
It was developed specifically for use in neuropsychological studies with
small control samples and can address the question of whether there is
evidence of dissociation between two different tasks in an individual
patient.
This study employed a modified version of the Hayling Test
(Burgess and Shallice, 1997), a standardized test of ‘automatic
response suppression’. The original test has two sections, both consisting of 15 sentences that are missing the last word. The critical
Section 2 requires the subject to complete the sentence with a word
that is completely ‘unconnected’ to the sentence in every way. In
other words, this section requires the subject to inhibit an automatic
verbal response and to generate a different response that must be
completely unrelated to the theme of the sentence. It yields two measures of the ability of the participant to suppress a response (an error
score and the time taken to respond). Thus, in summary, this test
provides a measure of response suppression, which has been shown
to be impaired in some patients with frontal lobe lesions (Burgess and
Shallice, 1997).
The Emotional Hayling Task
Section 2 of the original Hayling Test was modified to manipulate the
emotional theme of the sentences. The Emotional Hayling Task used in
the present investigation consisted of 30 sentences that were missing
the last word. They varied in theme as follows: (i) 10 sentences related
to cars and motoring issues (emotionally neutral sentences hereafter
referred to as ‘neutral’); (ii) 10 sentences related to violence and physical assault (emotionally negative sentences hereafter referred to as
‘negative’); and (iii) 10 sentences related to stroke and motor deficit
(emotionally negative sentences specifically related to patients’ deficits
hereafter referred to as ‘deficit-related’). Other than this emotional
theme manipulation, the three sets of sentences were matched for
syntactic structure, word count and semantic content as closely as
possible. Example 1: (i) neutral sentence: ‘A tow truck is often used
to pull broken-down cars off the. . .’; (ii) negative sentence: ‘An ambulance is often used to take assaulted people to the. . .’; and (iii)
deficit-related sentence: ‘A hoist is often used to lift paralysed patients
off the. . .’. Example 2: (i) neutral sentence: ‘Some cars can be repaired
following motor accidents but others might be left with some permanent mechanical. . .’; (ii) negative sentence: ‘Some people recover completely following physical attacks but others might be left with some
permanent emotional. . .’; and (iii) deficit-related sentence: ‘Some
people recover completely following brain damage but others might
be left with some permanent motor. . .’.
A pilot study with 23 adults (11 males and 12 females, graduate
students with mean age 22.5 years and SD 3.6 years, range
20–33 years) was conducted to validate the Emotional Hayling Task
in relation to the original standardized Hayling Test. All subjects completed the original and modified Hayling Test on separate sessions, in
counterbalanced order. There were no gender differences in the performance of either test. Furthermore, the reaction times and errors in
both males and females were never greater than 2 SD from the mean
of each subject for each category of content (‘neutral’, ‘negative’ or
‘deficit-related’). Performance on the original Hayling Test (response
inhibition speed and errors) was positively correlated with performance
on the Emotional Hayling Task (r = 0.78, P50.01 and r = 0.81,
3568
| Brain 2010: 133; 3564–3577
A. Fotopoulou et al.
Table 1 Groups’ demographic characteristics and neuropsychological profile
AHP (n = 7)
Age (years)
Education (years)
Days from onset
Premorbid IQ—WTAR
Berti awareness left upper limb
Berti awareness left lower limb
AHP questionnaire
WAIS—III vocabulary
WAIS—III similarities
WAIS—III digit span
WAIS—III matrix reasoning
WAIS—III arithmetic
Visual fields right
Visual fields left
Visual fields both
RASP surface touch left—max 30
RASP surface touch/spam max 10
RASP sensory extinction—max 12
RASP proprioception—max 30
movement detection left
RASP proprioception—max 30
direction detection left
Comb/razor test right
Comb/razor test left
Comb/razor test ambiguous
Comb/razor test bias
Bisiach test
BIT total score
Line crossing right
Line crossing left
Letter cancellation right
Letter cancellation left
Star cancellation right
Star cancellation left
Copy
Representational drawing
Line bisection right
Line bisection centre
Line bisection left
Hayling Test reaction times
Hayling Test errors
Proverbs
Cognitive estimates
HADS depression
HADS anxiety
Self-esteem
HP controls (n = 7)
t-Test
SD
Mean
SD
t
df
P
64.00
10.57
18.71
102.60
1.43
1.86
5.64
8.14
8.17
8.86
6.00a
7.00
9.29
2.71a
1.14a
6.57a
4.57
4.00
6.57a
6.06
3.10
19.63
16.59
0.53
0.38
1.38
3.89
3.19
3.18
2.12
2.55
0.49
2.93
1.68
4.80
2.51
5.22
2.64
56.86
12.29
21.29
105.50
0.00
0.14
0.93
9.00
9.17
10.43
6.75a
9.00
9.86
5.71a
2.43a
6.23a
3.86
7.5
6.57a
18.52
2.50
8.40
6.60
0.00
0.38
0.84
2.92
3.19
3.55
1.50
2.00
0.38
3.30
2.15
3.73
1.68
3.32
4.43
0.97
0.96
0.32
0.37
7.07
7.07
7.74
0.41
0.54
0.87
0.60
1.28
2.45
1.80
1.25
0.13
0.63
1.11
0.00
7.27
12
8.13
5.05
6
6
12
10
10
12
7
7
12
12
12
12
12
7
12
0.36
0.36
0.76
0.73
0.01*
0.01*
0.01*
0.69
0.60
0.40
0.57
0.24
0.03*
0.10
0.24
0.90
0.54
0.30
1.00
4.00a
2.31
4.57a
4.24
0.31
9.28
0.76
27.9
9.90
7.14
0.49a
0.70
87.83a
18.00
6.50a
16.17
7.67a
22.17
9.67a
1.33a
1.25
2.43
1.57a
0.86a
4.14
3.86
7.80
9.29a
4.57
8.00
21.57
16.0
13.60
9.35
0.39
0.50
30.51
0.00
6.12
2.40
5.32
5.98
11.34
1.75
0.96
0.98
1.52
1.21
1.57
1.57
2.59
2.43
5.29
4.43
5.47
23.14
16.00
8.00
0.17a
0.14
84.00a
16.17
13.50a
14.00
10.00a
22.33
17.83a
2.71a
2.17
2.43
2.14
1.57a
4.29
3.00
12.00
8.33a
7.43
7.86
20.71
3.23
6.73
5.89
0.14
0.38
44.78
3.13
7.15
3.74
6.73
5.05
9.39
1.25
1.17
0.98
1.46
1.51
1.89
2.52
5.10
2.25
4.31
5.34
4.61
0.77
1.07
0.21
2.11
2.45
0.18
1.44
1.82
1.22
0.68
0.05
1.36
1.65
1.30
0.00
0.72
0.97
0.15
0.76
1.64
0.73
1.11
0.05
0.32
6.49
12
12
7.6
12
11
5
10
11
11
10
10
11
8
12
12
12
12
10.07
8
11
12
12
12
Mean
0.47
0.31
0.84
0.07**
0.03*
0.86
0.21
0.10
0.25
0.51
0.96
0.20
0.13
0.23
1.00
0.49
0.35
0.88
0.46
0.14
0.48
0.29
0.96
0.76
AHP questionnaire = Feinberg et al. (2000) Awareness Questionnaire; Berti awareness = Berti et al. (1996) Awareness Interview; BIT total score = sum of scores of
the conventional sub-tests of the Behavioural Inattention Test; Comb/razor test = tests of personal neglect. Bias on the latter is calculated according to McIntosh et al.
(2000); HADS = Hospital Anxiety and Depression Scale (Zigmond and Snaith, 1983); Proverb test = Delis Kaplan—Executive Functions System—Proverbs Subtest (Delis et al.
2001); RASP = The Rivermead Assessment of Somatosensory Performance (Winward et al., 2000); Self-esteem = Rosenberg Self-Esteem Scale (1965); Visual fields = the
customary ‘confrontation’ technique (Bisiach et al., 1986); WAIS-III = Wechsler Adult Intelligence Scale—3 rd Edition (Wechsler, 1998); WTAR = Wechsler Test of
Adult Reading (Wechsler, 2001).
a Scores below tests’ cut-off point, or more than 1 SD below the average mean.
*Significant differences between the groups, P50.05.
**Trends towards significance, P50.10.
Implicit awareness in anosognosia for hemiplegia
P50.001, respectively), indicating that the Emotional Hayling Task is a
valid test of verbal inhibition.
Procedure and scoring
The original Hayling Test and the Emotional Hayling Task were administered in separate sessions scheduled at least 2 days apart. The original Hayling Test was administered in the first session in all patients to
increase their familiarity with the test’s procedure and demands. In the
second session, the instructions from the original Hayling Test (Part 2)
were again read out to participants. They were informed that a new
series of sentences would be read to them. Sentences were read out
clearly in a neutral manner and at a steady pace. Participants were
informed that they would be timed. Two practice examples were given
to each participant, as in the original test. The sentences of the
Emotional Hayling Task were read out to participants in a random
order, and timing began as soon as the last word of the incomplete
sentence was read. Patients’ answers were scored for reaction time in
milliseconds using a handheld Samsung stopwatch.
The original Hayling Test’s scoring guidelines and instructions were
followed as closely as possible. As in the original test, reaction
times and the following errors were noted: (i) connected completions
and (ii) somewhat connected completions. For example, the answer
‘hospital’ to the sentence ‘An ambulance is often used to take
assaulted people to the. . .’ was considered a ‘connected completion’
error. The answer ‘floor’ to the sentence ‘A hoist is often used to lift
paralysed patients off the. . .’ was considered a ‘somewhat connected
completion’ error. Finally, the answer ‘classroom’ to the sentence
‘A tow truck is often used to pull broken-down cars off the. . .’ was
considered a correct (unconnected) response. Unlike the original
Hayling Test, reaction times were not rounded up to seconds but
were recorded in milliseconds. In addition, reaction times that were
above or below 3 SD of the participant’s mean reaction time
per sentence type were excluded from the analysis as outliers
(Kampman et al., 2002). This occurred only twice and not in the
same patient.
In a final, third session, scheduled at least 2 days after the Emotional
Hayling Task session, the 30 incomplete sentences of the Emotional
Hayling Task were read out to participants in a random order, and
they were asked to rate how much the theme of each completed
sentence was relevant to them in their current situation on a scale
from 1 to 10 (anchored at ‘1 = not relevant to my current situation
at all’ and ‘10 = extremely relevant to my current situation’). Their
responses were not timed.
To summarize, both the AHP and HP groups completed an implicit
awareness test consisting of a verbal inhibition task, which included
sentences of varying emotional content (‘neutral’, ‘negative’ and
‘deficit-related’). Subsequently, both groups were also asked to
make explicit ratings of the self-relevance of the same sentences (explicit awareness task). It was expected that while the patients with
AHP would rate the ‘deficit-related’ sentences as less self-relevant
than controls, they would be significantly slower than controls in completing the inhibition task (due to unconscious interference) with these
sentences than with other emotionally ‘negative’ sentences (both
relative to ‘neutral’ sentences).
Lesion analysis methods
Patient lesions were mapped on slices of the T1-weighted MRI scan
template (ICBM152) from the Montreal Neurological Institute. This
template is approximately oriented to match Talairach space
(Talairach and Tournoux, 1988). All lesion plots were drawn on the
standard Montreal Neurological Institute space (2 2 2 mm) by one
of us (S.P.), who was blind as to which patient group each scan was
Brain 2010: 133; 3564–3577
| 3569
from. By using the MRIcro software (http://www.cabiatl.com/
mricro/mricro/index.html; Rorden and Brett, 2000), the template
was first rotated to match the orientation of the patient’s MRI or
CT scan. The scan images of the patient’s brain were then normalized
and aligned (by digital image editing software) to superimpose onto
the rotated template slices. The patient’s lesions were outlined on the
rotated template, resulting in a map in which each voxel was labelled
either 0 (intact) or 1 (lesion). Afterwards, the 3D volumes obtained
were rotated back to match the stereotaxic space of the Montreal
Neurological Institute T1-weighted template by using the interpolation
of the nearest-neighbour voxels. The derived volumes representing
the lesions of each patient were superimposed onto the ‘automated
anatomical labelling’ template (http://www.cyceron.fr/web/aal__
anatomical_automatic_labeling.html; Tzourio-Mazoyer et al., 2002)
to determine the lesion voxels of the different cerebral structures as
calculated by the MRIcro software (Rorden et al., 2007). The lesion
involvement of white matter structures and connections was achieved
by means of the lesion plots’ overlap with the ‘white matter parcellation map’ template (Mori et al., 2008).
Results
Explicit awareness
The groups’ explicit ratings of self-relevance to ‘deficit-related’,
‘neutral’ and ‘negative’ sentences are shown in Table 2. As
expected, the patients with AHP rated the ‘deficit-related’ sentences as less self-relevant than did the patients with hemiplegia.
This difference was not observed for the ‘neutral’ and ‘negative’
sentences. To verify the statistical significance of these differences,
we conducted a 2 (AHP group versus HP group) 2 (‘negative’
minus ‘neutral’ versus ‘deficit-related’ minus ‘neutral’ sentences)
repeated measures ANOVA. This revealed that the self-relevance
of sentences was rated as significantly different depending on their
content, F(1,12) = 204.5, P50.001, and that there was an interaction between patient group and the content of the sentences,
F(1,12) = 98.8, P50.001. Post hoc Bonferroni corrected t-tests
( = 0.025) revealed that the two groups differed significantly in
the difference ‘deficit-related’ minus ‘neutral’ sentence ratings,
t(12) = 3.7, P50.025. This difference was not significant for the
composite measure of ‘negative’ sentences (‘negative’ minus ‘neutral’ sentence ratings), t(12) = 0.4, P = 0.7. These results confirm
that the patients with AHP rated the ‘deficit-related’ sentences
as significantly less self-relevant than the patients with hemiplegia,
relative to both groups’ ratings of ‘neutral’ sentences.
It is also noteworthy that all patients with AHP rated
‘deficit-related’ sentences as less self-relevant than any of the
hemiplegic controls, and their individual scores were significantly
different from the mean of the hemiplegic control group, as determined by Crawford’s test (Crawford and Howell, 1998; Crawford
and Garthwaite, 2002), t’s = 4.8 – 7.7, P’s50.002. These results
confirm that the anosognosic patients, selected on the basis of two
established tests of motor awareness (Berti et al., 1996; Feinberg
et al., 2000), showed denial of their deficits in our experimental
task both individually and as a group. This confirmed the validity
of our novel explicit awareness task.
3570
| Brain 2010: 133; 3564–3577
A. Fotopoulou et al.
Table 2 Groups’ performance on the Emotional
Hayling Task
Emotional Hayling Task scores
AHP (n = 7)
Mean (SD)
Explicit (ratings)
‘Negative’
1.9 (0.4)
‘Neutral’
3.1 (1.7)
3 (0.8)
‘Deficit-related’a
Implicit (reaction times) correct and error responses
‘Negative’
8.4 (7.4)
‘Neutral’
7.6 (8.4)
‘Deficit-related’a
17.7 (16.5)
Implicit (reaction times) correct responses only
‘Negative’
5.9 (4.4)
‘Neutral’
6.7 (7.4)
‘Deficit-related’a
16.8 (17.2)
Implicit (errors)
‘Negative’
4.7 (2.6)
‘Neutral’
4.4 (2.6)
‘Deficit-related’
4 (2.7)
Error analysis
HP (n = 7)
Mean (SD)
1.8 (0.7)
3.5 (2.3)
7.9 (0.6)
8.2 (5.5)
6.4 (2.8)
5.7 (2.9)
7.2 (3.6)
4.5 (1.8)
3.7 (2.2)
5 (2.4)
5.1 (2.3)
6.4 (2.6)
a Significant difference between the groups.
Implicit awareness
Reaction times
The groups’ average reaction times in each content category for
correct and error responses, as well as for correct responses only,
are presented in Table 2. In a first analysis we explored the reaction times of the two groups for both correct and incorrect responses (errors). A 2 (AHP group versus HP group) 2 (‘negative’
minus ‘neutral’ versus ‘deficit-related’ minus ‘neutral’ sentences)
repeated measures ANOVA revealed that reaction times were not
overall significantly different depending on the sentence content,
F(1,12) = 2.7, P = 0.12, but there was a significant interaction between group and content of sentences, F(1,12) = 8.2, P50.05.
Post hoc Bonferroni corrected t-tests ( = 0.025) revealed that
the two groups differed significantly in the difference (‘deficitrelated’ minus ‘neutral’ sentences) reaction times, t(12) = 3.2,
P50.025, but not in the difference (‘negative’ minus ‘neutral’
sentences) reaction times, t(12) = 0.4, P = 0.66.
In a second analysis, we explored the reaction times of the two
groups for only correct responses. In this analysis two subjects,
AHP4 and HP5, who had 510 correct responses (22/30 and
24/30 errors, respectively) were excluded as outliers (see also
Kampman et al., 2002). A 2 (AHP group versus HP group) 2
(‘negative’ minus ‘neutral’ versus ‘deficit-related’ minus ‘neutral’
sentences) repeated measures ANOVA revealed that reaction
times were not overall significantly different depending on the
sentence content, F(1,10) = 1.7, P = 0.22, but there was a significant interaction between group and content of sentences,
F(1,10) = 6.34, P50.05. Post hoc, Bonferroni corrected t-tests
( = 0.025) revealed that the two groups differed significantly in
the difference (‘deficit-related’ minus ‘neutral’ sentences) reaction
times, t(10) = 2.6, P50.025, but not in the difference (‘negative’
minus ‘neutral’ sentences) reaction times, t(10) = 1.7, P = 0.1.
The groups’ average error rates in each content category are
presented in Table 2. A 2 (AHP group versus HP group) 2
(‘negative’ minus ‘neutral’ versus ‘deficit-related’ minus ‘neutral’
sentences) repeated measures ANOVA revealed that the number
of errors was not significantly different depending on the sentence
content, F(1,12) = 0.5, P = 0.59 and there was no significant interaction between group and content of sentences, F(1,12) = 2.7,
P = 0.12. Although this last interaction was not significant, we performed an additional analysis to rule out the possibility that the
above significant differences in total reaction times (for both correct responses and errors) were a function of different error rates
between the groups. A 2 (AHP group versus HP group) 2
(‘negative’ minus ‘neutral’ versus ‘deficit-related’ minus ‘neutral’
sentences) repeated measures ANOVA was conducted on reaction
times with the number of errors each patient produced per content category used as a covariate. All interactions of sentence
content*covariate were non-significant, while the interaction sentence content*group remained significant, F(1,12) = 6.2, P50.05.
In summary, we found that patients with AHP rated ‘deficitrelated’ sentences (relative to ‘neutral’ sentences) as significantly
less self-relevant than did hemiplegic patients (explicit task). By
contrast, the two groups did not differ in their self-relevance ratings of ‘negative’ sentences (relative to ‘neutral’ sentences).
Crucially, patients with AHP were significantly slower than patients
with hemiplegia in inhibiting automatic responses to the
‘deficit-related’ than the ‘negative’ sentences, relative to ‘neutral’
sentences (implicit task). These results show a clear dissociation in
performance between the implicit and explicit awareness tasks in
patients with AHP.
Individual patient analysis
We used the specialized RSDT (see the ‘Design and analysis’
subsection) to explore how many of the patients with AHP differed significantly from the control sample in their explicit ratings
and implicit responses to deficit-related sentences. We applied this
method to the differences between ‘deficit-related’ and ‘neutral’
sentences in explicit ratings versus implicit reaction times, as these
were the differential measures that were found to be significantly
different between the two groups (see above). As above, we
analysed the reaction times for correct responses only and in a
separate analysis for correct and incorrect responses.
The RSDT indicated a significant dissociation in all patients with
AHP except Patient AHP7, when considering either correct responses and errors (t6 = 1, P = 0.17) or only correct responses
(t6 = 1.3, P = 0.12). These results are presented in Table 3. Given
this difference, we also looked at the differences between Patient
AHP7 and the rest of the AHP group in all of the neuropsychological and psychometric tests administered. Patient AHP7 scored
within the range of the rest of the AHP group on all of the
neuropsychological tests, including the original Hayling Test.
However, he did show a marked difference in self-reported
depression, scoring 15 while the rest of the patients scored
between 1 and 6 [mean 2.84; SD 2.86]. He also showed the
highest score on anxiety (13; AHP group mean 7.16; SD 4.25;
range 2–12) and the lowest score on self-esteem (15; AHP
Implicit awareness in anosognosia for hemiplegia
Brain 2010: 133; 3564–3577
| 3571
Table 3 Individual differences in explicit and implicit performance
HP group
mean (SD)
AHP1
AHP2
AHP3
Explicit rating
‘Neutral’
3.5 (7.9)
5.1
1.7
4.8
‘Deficit-related’
7.89 (0.7)
2.3
3
2.1
Implicit reaction times (correct and errors)
‘Neutral’
6.4 (2.8)
1.51
6.58
6.23
‘Deficit-related’
5.73 (3)
3.02
13.42
15.38
Implicit reaction times (correct only)
‘Neutral’
4.45(1.9)
1.46
7.21
4.51
‘Deficit-related’
3.73 (2.2)
2.96
14.79
14.05
RSDT: explicit rating difference versus reaction times difference (correct and errors)
t(7)
2.2
1.7
4
P (one-tailed)
0.04a
0.07a
0.004a
RSDT: explicit rating difference versus reaction times difference (correct only)
t(6)
2.1
4
4
P (one-tailed)
0.04a
0.004a
0.004a
AHP4
AHP5
AHP6
1.5
4.3
4.8
3.7
2.3
3
1.4
2.5
6.16
20.45
4.54
16.23
26.15
52.14
2.34
3.53
Excl
Excl
3.74
15.24
21.27
50.00
2.18
3.84
3.9
0.004a
Excl
Excl
AHP7
4.2
0.003a
6.7
0.0001a
1
0.17
4.1
0.003a
7.5
0.0001a
1.3
0.12
RSDT = Revised Standardized Difference Test (Crawford et al., 1998; revised in Crawford and Garthwaite, 2005).
a Significant difference between the patient and the control HP group.
group mean 22.67; SD 5.08; range 19–29). When assessed with
the Crawford test (Crawford and Garthwaite, 2002), only the difference in depression scores was significant [t = 3.94;
P(two-tailed)50.05].
Lesion analysis
All the lesions resulted from a single ischaemic or haemorrhagic
stroke and were confined to the right hemisphere, mainly in the
territory of the right middle cerebral artery. In the AHP group,
Patients AHP1, AHP5 and AHP7 suffered from large hemispheric
brain lesions affecting most of the right middle cerebral artery
territory, involving cortical and subcortical areas. Patients AHP2,
AHP3 and AHP4 had subcortical damage affecting mainly the
basal ganglia, insula and surrounding white matter. Patient
AHP6 showed a hemispheric lesion, mostly subcortical but extending to the medial parts of the frontal and occipital cortex.
Figure 1A illustrates the lesions overlapping in three or more patients with AHP. In the control group, Patients HP2, HP3 and HP4
had lesions that were mostly subcortical, extending from the thalamus and the basal ganglia to the insula and the surrounding
white matter. Patients HP1 and HP6 had frontotemporo-parietal
hemispheric lesions, Patient HP7 had a frontotemporal lesion and
Patient HP5’s lesion involved mainly dorsomedial cortical areas and
the deep white matter. Figure 1B illustrates the lesions overlapping
in three or more patients with hemiplegia.
No significant difference was observed in the total volumes of the
lesions in the AHP group (mean 132 cm3, SD = 110) and HP group
[mean 93 cm3, SD = 98; t(14) = 0.69, P = 0.68]. Subtracting HP
group lesion plots from the AHP group lesion plots gave a cluster
that remained present in at least five patients with AHP. This cluster
involved mainly the anterior part of the insula and extended through
the anterior corona radiata and the external capsule to the caudate
and putamen nuclei (Fig. 1C and D).
By using the non-parametric mapping software available with
MRIcro, we compared the frequency of lesion voxels in the two
groups by computing a binomial test based on the Liebermeister
measure (false discovery rates corrected), because this binomial
test appears to be more sensitive than chi-squared or Fisher’s
Exact test (Phipps, 2003). The significant areas emerging from
the binomial test were in line with the subtraction results, revealing a superior cluster extending from the rolandic operculum and
anterior insula to the caudate and putamen nuclei, and an inferior
cluster involving the amygdala and the superior temporal pole.
White matter tissue included the anterior corona radiata, the external capsule, the retro-lenticular part of the internal capsule, and
the uncinate fasciculus (Fig. 1 E and F).
In order to explore a possible causal correlation between
the lesions of Patient AHP7 and his lack of implicit awareness,
we compared his lesions with those of the others in the AHP
group (Fig. 2). The lesions of Patient AHP7 overlapped with
several of the common lesion areas of the other patients with
AHP, but showed more lesion voxels in posterior (posterior and
medial regions of the occipital lobe) anterior (medial regions of the
frontal lobe) and dorsolateral cortical regions (pre-central and
post-central areas). He also showed fewer lesion voxels in limbic
regions (e.g. the amygdala), the basal ganglia and the external
capsule.
Discussion
The present study used an inhibition task to study implicit processing of deficit-related material and a rating task to study explicit
processing of the same material in left hemiplegic patients with
and without AHP. The main behavioural finding was that patients
with AHP were significantly slower than control patients in
performing the inhibition task with deficit-related sentences than
with other emotionally negative sentences, relative to neutral
sentences in both cases. The main finding of the lesion analysis
was that the lesions of the patients with AHP differed from those
of the hemiplegic controls by involving mainly the anterior parts of
3572
| Brain 2010: 133; 3564–3577
A. Fotopoulou et al.
Figure 1 (A) Overlays of regional lesion plots of minimum three patients with AHP [Montreal Neurological Institute (MNI) coordinates of
the centre of mass: x = 39, y = –16, z = 16]. The number of overlapping lesions is illustrated by different colours coded for increasing
frequencies, from red (lesion in three patients) to light yellow (lesion in seven patients). (B) The same colour codes and the same axial
sections illustrate the overlapping lesions of minimum three patients with hemiplegia (MNI coordinates of the centre of mass: x = 38,
y = –9, z = 21). (C) Subtraction lesion plots of patients with AHP (positive overlap) with patients with hemiplegia (negative overlap). The
result of the subtraction is illustrated in percent numbers and different colours code for increasing percentage, from purple (57% lesion in
patients with hemiplegia) to light yellow (71% lesion in patients with AHP). (D) The table shows the damaged brain regions ‘surviving’ the
subtraction in more than four patients with AHP (57%); the maximum AHP overlap percentage and the MNI coordinates of the centre of
mass are shown for each region. (E) For each voxel, frequency comparison of lesions in AHP with respect to patients with hemiplegia was
computed by means of the Liebermeister binomial measure (false discovery rate corrected). In green, yellow and orange are the regions
statistically significant at the 5% level [Liebermeister binomial measure (L) = 1.77]; in red are the regions statistically significant at the
1% level (L = 2.57). (F) The table shows for each region the number (n) and the percentage (n%) of clustering voxels that ‘survived’
the threshold of P50.05 (false discovery rate corrected) along with the maximum Liebermeister (L) statistic obtained in each cluster
and the MNI coordinates of the centre of mass. AAL = Automated Anatomical Labeling; FDR = false discovery rate.
the insula, inferior motor areas, basal ganglia structures, limbic
structures and deep white matter. The one patient with AHP
who did not show dissociation between implicit and explicit
awareness of deficit showed more lesions in cortical frontal
areas, including lateral motor and premotor regions, and also in
the parietal and occipital lobes.
The selective slowing of responses to emotional stimuli in
Stroop-like tests is considered to reflect ‘unconscious’ response
competition between the processing of emotional self-threatening
information and the requirement to complete the task (Pratto and
John, 1991; Wentura et al., 2000). We propose that our findings
in the patients with AHP reflect a similar ‘unconscious’ response
Implicit awareness in anosognosia for hemiplegia
Brain 2010: 133; 3564–3577
| 3573
Figure 2 Regional lesion plot of Patient AHP7 (in blue, centre of mass: x = 41; y = –22; z = 23). The red and yellow colours represent the
lesion voxels of the other six patients with AHP. The number of overlapping lesions is illustrated by different colours coded for increasing
frequencies, from dark red (lesion in one patient with AHP) to light yellow (lesion voxels overlap in six patients with AHP). The three tables
provide a quantitative estimate of the region plots lesioned (i) in Patient AHP7 but not in the other patients with AHP; (ii) in patients with
AHP excluding Patient AHP7; and (iii) finally the common lesions of the six patients with AHP and Patient AHP7. For each region, the
number (n) and the percentage (n%) of lesioned voxels are shown. MNI = Montreal Neurological Institute.
competition between the processing of deficit-related information
and the requirement to complete the inhibition task. In contrast,
patients with AHP did not differ from controls in their overall
ability to perform the inhibition task as assessed by the original,
standardized test of verbal inhibition (the Hayling Test). Secondly,
the observed effect of inhibition in the patients with AHP was
based on a ‘differential’ slowing of response times for deficitrelated versus neutral words, rather than an absolute measure of
response speed (see also Nardone et al., 2007). Thus, it is unlikely
that this effect relates to the general inhibitory abilities of the
patients with AHP. In addition, the patients with AHP showed
more lesions than the hemiplegic patients in right limbic and
insular cortex areas (see also below) and a corresponding reduction in the experience and communication of negative emotions
(see Heilman and Harciarek, 2010 for review). Nevertheless, it is
unlikely that our main experimental effects were driven by these
emotional processing difficulties, or any general ‘defence’ function
against negative emotions (see below), as patients with AHP did
not show a tendency to process other emotionally negative
themes differently than hemiplegic controls. Finally, it is unlikely
that this effect related to dissociations between modalities, as both
the implicit and explicit tasks included listening to, comprehending
and verbally responding to the same sentences.
The above considerations suggest that our study provides strong
experimental support for implicit awareness of deficit in some
anosognosic patients. Implicit awareness has previously been
hypothesized based on clinical observations (e.g. Weinstein and
Kahn, 1955; Ramachandran, 1995) and case studies (Nardone
3574
| Brain 2010: 133; 3564–3577
et al., 2007). The present experimental confirmation of this dissociation is important because the combination of our behavioural
and neuroimaging results may thus shed some light on the deficits
leading to explicit unawareness in AHP and the potentially intact
mechanisms responsible for implicit awareness.
Recent studies have postulated that explicit unawareness in AHP
is caused by a deficit in monitoring the discrepancy between intended and actual movement (Frith et al., 2000; Berti et al., 2005;
Pia and Berti, 2006). This deficit can lead to the construction of
non-veridical motor awareness based entirely on representations
of intended movement (Berti et al., 2007; Fotopoulou et al., 2008;
Desmurget and Sirigu, 2009). This hypothesis has been supported
by neuroimaging evidence (Berti et al., 2005); areas involved in
motor preparation and planning (e.g. the supplementary motor
area) were found to be spared in patients with AHP. In contrast,
lateral premotor regions, which are typically involved in monitoring
discrepancies between intention and action, were selectively
affected in these patients.
Our neuroimaging results in the present study are consistent
with the above proposals, in that motor areas around the rolandic
operculum were more frequently damaged in patients with AHP
than patients with hemiplegia. Moreover, as in the study of Berti
et al. (2005) the supplementary motor area was spared in all the
patients with AHP. These results confirm the hypothesis that
motor planning is intact in patients with AHP (Frith et al., 2000;
Berti et al., 2005) and may lead some patients to form illusory
awareness of movements (Fotopoulou et al., 2008; for review see
Jenkinson and Fotopoulou, 2010).
However, the dorsal premotor area (e.g. Brodmann area 6) was
not damaged more frequently in patients with AHP than patients
with hemiplegia in the present study. In Berti and colleagues’
(2005) study, this was the brain region most frequently associated
with AHP. In the present study, it was affected in only the patient
who showed explicit but not implicit awareness. Only three other
patients with AHP had damage to this area and this damage was
minimal. This finding suggests that an intact or partly damaged
lateral premotor cortex may facilitate implicit awareness of deficit.
Future studies will need to explore this finding further and consider the potential role of other brain areas (e.g. parietal areas) in
residual sensorimotor monitoring and implicit awareness. Such
findings would suggest that implicit awareness is generated
within intact sensorimotor mechanisms. Whatever the outcome
of such studies, however, our current experimental investigations
suggest that such intact implicit awareness does not seem sufficient to counter patients’ more general, explicit belief that they
can move.
More broadly, the observed dissociation between implicit
and explicit awareness is consistent with the observation that
AHP sometimes has delusional features that cannot be explained
solely on the basis of sensorimotor deficits (Ramachandran, 1995;
Solms, 1999; Frith et al., 2000; Vuilleumier, 2004; Fotopoulou,
2010). The latter can perhaps explain the ‘illusion’ of moving
(Fotopoulou et al., 2008), but patients with AHP do not simply
claim that they have illusions of moving. They instead ignore the
wealth of evidence that they are paralysed (e.g. their disabilities,
occasional accidents, others’ feedback) and adhere to the
‘delusional’ belief that they have functional limbs. The explanation
A. Fotopoulou et al.
of the latter belief requires the postulation of another dysfunction
that prevents sensorimotor and other failures from being
re-represented at a higher level of cognitive self-representation.
Indeed, a number of authors have noted that the delusional,
explicit aspects of AHP can best be explained as an inability to
attribute motor errors and other failures to oneself and update
one’s self-representation accordingly (Ramachandran, 1995;
Marcel et al., 2004; Vuilleumier, 2004). Consistent with the present study, these hypotheses are supported by clinical observations
of implicit awareness in patients with AHP. For example, Marcel
and colleagues (2004) observed that patients with AHP were more
accurate in describing their deficits from a third-person than a
first-person perspective. Similarly, when Fotopoulou et al. (2009)
improved explicit awareness in a patient with AHP (patient LM) by
showing her a third-person and ‘off-line’ perspective of her body
in a video replay, she also confirmed that at some level she had
been aware of her disability all along. Ramachandran (1995) also
reported a patient with AHP who was retrospectively able to accurately comment on how long she had been paralysed during
temporary lifting of the AHP following vestibular stimulation.
Despite this temporary insight, after the stimulation effect wore
off, the patient reverted to denying her paralysis and the admission of it during stimulation. Together with the present experimental findings of implicit awareness in AHP, these observations
suggest that in at least some patients with AHP, information
about their deficit has been (implicitly) laid down continuously in
their memory. Nevertheless, patients lack the ability to integrate
these implicit memories into an explicit and stable self-awareness.
Interestingly, negative emotions may be among the information
that these patients fail to self-attribute and explicitly experience.
For example, the aforementioned improvement of awareness in
Patient LM was not associated with improvement in cognitive or
sensorimotor functions, but with a quantified increase in depressive symptoms (Fotopoulou et al., 2009). Similar observations of a
sudden influx of depressive symptoms (‘catastrophic reactions’;
Goldstein, 1939) during episodes of transient awareness have
been noted in patients who are otherwise unaware of or indifferent towards their deficits (Kaplan-Solms and Solms, 2000; Turnbull
et al., 2002, 2005). In addition, it has been noted that some patients with AHP, who appear emotionally unresponsive towards
their deficits, tend nevertheless to experience disproportionately
intense negative emotions about minor unpleasant situations and
other people (Kaplan-Solms and Solms, 2000; Fotopoulou and
Conway, 2004).
Thus, it seems that in several patients with AHP, certain dysfunctions prevent awareness of motor failures and the corresponding negative emotions to be integrated with explicit awareness of
the self. Some authors have argued that this lack of explicit awareness is caused by psychogenic ‘defence’ mechanisms directed
against anxiety and depression (Weinstein and Kahn, 1995).
However, the relative neuroanatomical and behavioural specificity
of anosognosic behaviours suggests that such purely psychological
mechanisms are insufficient to explain AHP (for a recent critical
review, see Heilman and Harciarek, 2010). Importantly, the fact
that such ‘denial’ behaviours occur following specific brain lesions
can allow us to put forward more parsimonious, empirically
informed accounts of brain mechanisms that when damaged can
Implicit awareness in anosognosia for hemiplegia
lead to such delusional behaviours and attitudes (see Fotopoulou,
2010 for discussion of this approach, heuristically labelled ‘affective neuropsychology’). Based on our current and previous findings,
as well as on results of other groups (Berti et al., 2005; Karnath
et al., 2005), we propose that the likely candidate lesions for such
impairment in explicit cognitive and emotional awareness include
the anterior insular cortex and adjoining frontal areas, with possible involvement of limbic areas and cortico-striato-thalamic pathways. We will briefly discuss these possibilities in relation to the
present findings.
Consistent with previous studies (Berti et al., 2005; Karnath
et al., 2005; Baier and Karnath, 2008), we found that the anterior
parts of the right insula were differentially damaged in patients
with AHP. A plethora of recent neuroimaging studies have suggested that anterior parts of the insula seem to be responsible for
the re-representation of multisensory, motor and interoceptive information that is first processed in other cortical and subcortical
areas and converge in the mid-posterior insula (Critchley et al.,
2004; for reviews see Craig, 2009; Tsakiris, 2010). Some authors
have proposed on this basis that the anterior parts of the right
insular cortex are central areas in the neural circuit that subserve a
stable, embodied and emotionally embedded meta-representation
of oneself in the present moment (Damasio, 1994; Craig, 2009).
This representation inherently creates a subjective, first-person
perspective that differentiates between inner and outer spheres
and supports all cognitive and emotional operations that require
the distinction between self and other (e.g. attribution of agency).
It is thus possible that in patients with AHP, damage to the right
anterior insula and adjoining frontal areas contributes to impairment in the re-representation of information processed in other
sensorimotor and emotional regions. This could render this
information more likely to be ignored or attributed to other
people and causes (Kaplan-Solms and Solms, 2000; Fotopoulou
and Conway, 2004; Marcel et al., 2004) and less likely to be
explicitly self-attributed.
It should be noted that in the present study, anterior and middle
but not posterior parts of the right insula were differentially
damaged in patients with AHP. Karnath and colleagues (Karnath
et al., 2005; Baier and Karnath, 2008) proposed that the right
posterior insula is uniquely associated with AHP. Berti et al.
(2005) found that both anterior and posterior parts of the right
insula were among the areas specifically associated with AHP.
These discrepancies can potentially be explained by distinct patient
recruitment criteria, which are important to note, particularly given
the observed clinical heterogeneity of AHP. In the studies of
Karnath et al. (2005) there was a high incidence of body ownership disturbances. This association has not been reported in other
studies (e.g. Berti et al., 2005). In our study there was only one
patient with indications of somatoparaphrenia (Patient AHP5).
Interestingly, in this patient, unlike the rest of the patients with
AHP, both the anterior and posterior parts of the insula were
affected. Taken together, these findings suggest that the right
anterior insula may be responsible for an integrated sense of
bodily awareness that includes motor agency (see also Farrer
and Frith, 2002; Craig, 2009; Tsakiris et al., 2010), while the posterior insula may be a critical part of the network responsible for
more basic interoceptive representation and body ownership
Brain 2010: 133; 3564–3577
| 3575
(Craig, 2002; Olausson, et al., 2005; Tsakiris et al., 2007).
Finally, our results, as well as those of previous neuroimaging
studies on AHP (Berti et al., 2005; Karnath et al., 2005; Baier
and Karnath, 2008), do not directly address the potential role of
the left insula in body and emotional awareness (see Craig, 2009
for discussion). Nevertheless, it is noteworthy that some neuroimaging studies of motor and emotional awareness have found bilateral activation of insular cortex (e.g. Farrer and Frith, 2002).
Interestingly, a recent study of a patient with epilepsy following
herpes simplex encephalitis who developed Cotard’s syndrome (a
severe form of unawareness typically including the delusional
belief that one is dead, or dying, as well as various other
self-deprecatory delusions, suicidal ideation, feelings of guilt and
denial of body parts) reported bilateral lesions to the insular cortex
(McKay and Cipolotti, 2007). Future lesion studies could thus explore the specific roles of the right and left insular cortex in motor
and emotional awareness and importantly their combined
functional role.
Our results also suggest an association between AHP and subcortical lesions, including basal ganglia and amygdala damage.
Vuilleumier (2004) has argued that damage to subcortical circuits
(e.g. basal ganglia) that are involved both in motivation and in
detection of ‘errors’ might lead to an inability to revise beliefs
based on novel perceptual experience and uncertain bodily
states. It is also of interest that the lesions of the one patient
(Patient AHP7) who showed increased depression but not implicit
awareness of deficit did not include damage to subcortical areas
such as the amygdala or the hippocampus and showed a lesser
degree of damage to basal ganglia structures. In previous studies,
bilateral medial prefrontal cortex and related limbic and striatopallido-thalamic lesions have been associated with lower levels
of depression than damage to dorsal prefrontal areas or more
posterior lesions (for review see Koenigs and Grafman, 2009).
Interestingly, patients with such lesions in the anterior limbic
areas have been found to report low levels of ‘cognitive/affective’
symptoms (such as guilt, self-dislike and sadness) but normal levels
of ‘somatic’ symptoms (such as fatigue and changes in sleeping or
appetite) (Koenigs et al., 2008).
Given our small sample and the inherent limitations of lesion
analysis, our findings regarding the various dissociations observed
and their neural correlates are preliminary. Nevertheless, in combination with previous studies, they provide tentative new insights
into the psychological and neural mechanisms of the multifaceted
AHP syndrome. In short, our findings suggest that: (i) some patients with AHP may show implicit awareness into their deficits,
while others do not; (ii) some patients with AHP may show body
ownership disturbances, while others do not; and (iii) both of
these dissociations may be linked with differences in lesion sites
(in motor versus interoceptive awareness areas, respectively).
These observations and our previous findings of illusory awareness
of movement in AHP (Fotopoulou et al., 2008) are consistent with
the suggestion that AHP is a multicomponent and heterogeneous
syndrome, encompassing various deficits and different clinical
populations (e.g. Marcel et al., 2004; Vuilemieur, 2004; Orfei
et al., 2007). Importantly, in the present study we observed that
most of our patients with AHP suffered from a specific delusion of
first-person awareness linked with damage to the anterior insular
3576
| Brain 2010: 133; 3564–3577
cortex and some possible involvement of basal ganglia and limbic
circuits. We proposed that these lesions seem to impair patients’
ability to re-represent and affectively personalize sensorimotor information in self-awareness. This information is instead processed
only implicitly, possibly within intact premotor areas and other
cortical regions.
Acknowledgements
The authors would like to thank the patients and their families for
their participation. We are also grateful to senior physiotherapist
Sandra Chambers and the staff at Mark Ward, St Thomas’s
Hospital, for their assistance with this study. Finally, the authors
are thankful to Paul Jenkinson and Mark Solms for their comments
on an earlier version of this manuscript.
Funding
ESRC/MRC Fellowship (to A.K. and M.A.K.); Neuropsychoanalysis
Foundation Fellowship (to A.K.); Volkswagen Foundation
‘European Platform for Life Sciences, Mind Sciences and the
Humanities’ grant for the ‘Body-Project’ (to A.K.).
References
Babinski J. Contribution a‘ l’ étude des troubles mentaux dans l’
hémiplégie organique cérébrale. Revue Neurologique 1914; 27: 845–8.
Baier B, Karnath HO. Tight link between our sense of limb ownership
and self-awareness of actions. Stroke 2008; 39: 486–8.
Benedetti F, Bernasconi A, Pontiggia A. Depression and neurological
disorders. Curr Opin Psychiatry 2006; 19: 14–8.
Berti A, Ladavas E, Della Corte M. Anosognosia for hemiplegia, neglect
dyslexia and drawing neglect: clinical findings and theoretical considerations. J Int Neuropsychol Soc 1996; 2: 426–40.
Berti A, Bottini G, Gandola M, Pia L, Smania N, Stracciari A, et al. Shared
cortical anatomy for motor awareness and motor control. Science
2005; 309: 488–91.
Berti A, Ladavas E, Stracciari A, Giannarelli C, Ossola A. Anosognosia for
motor impairment and dissociations with patients’ evaluation of the
disorder: theoretical considerations. Cogn Neuropsychiatry 1998; 3:
21–44.
Berti A, Spinazzola L, Pia L, Rabuffetti M. Motor awareness and motor
intention in anosognosia for hemiplegia. In: Haggard P, Rossetti Y,
Kawato M, editors. Sensorimotor foundations of higher cognition
series (Attention and Performance XXII). Oxford: Oxford University
Press; 2007. p. 163–81.
Bisiach E, Geminiani G. Anosognosia relating to hemiplegia and hemianopia. In: Prigatano GP, Schacter DL, editors. Awareness of deficit
after brain injury: clinical and theoretical issues. New York: Oxford
University Press; 1991. p. 17–39.
Bisiach E, Perani D, Vallar G, Berti A. Unilateral neglect: Personal and
extra-personal. Neuropsychologia 1986; 24: 759–67.
Burgess PW, Shallice T. The Hayling and Brixton Tests Thames. London:
Valley Test Company; 1997.
Charash M, McKay D. Attention bias for disgust. Anxiety Disorders
2002; 16: 529–41.
Cocchini G, Beschin N, Cameron A, Fotopoulou A, Della Sala S.
Anosognosia for motor impairment following left brain damage.
Neuropsychology 2009; 23: 223–30.
A. Fotopoulou et al.
Cocchini G, Beschin N, Fotopolou A, Della Sala S. Explicit and implicit
anosognosia for upper limb motor impairment. Neuropsychologia
2010; 48: 1489–94.
Coslett HB. Anosognosia and body representations forty years later.
Cortex 2005; 41: 263–70.
Craig AD. How do you feel? Interoception: the sense of the physiological
condition of the body. Nat Rev Neurosci 2002; 3: 655–66.
Craig AD. How do you feel–now? The anterior insula and human awareness. Nat Rev Neurosci 2009; 10: 59–70.
Crawford JR, Garthwaite PH. Investigation of the single case in neuropsychology: Confidence limits on the abnormality of test scores and
test score differences. Neuropsychologia 2002; 40: 1196–208.
Crawford JR, Garthwaite PH. Testing for suspected impairments and
dissociations in single-case studies in neuropsychology: Evaluation
of alternatives using Monte Carlo simulations and revised tests for
dissociations. Neuropsychology 2005; 19: 318–31.
Crawford JR, Howell DC. Comparing an individual’s test score against
norms derived from small samples. The Clinical Neuropsychologist
1998; 12: 482–6.
Crawford JR, Howell DC, Garthwaite PH. Comparing an individual’s test
score against norms derived from small samples. Clin Neuropsychol
1998; 12: 482–6.
Critchley M. The parietal lobes. London: Edward Arnold; 1953.
Critchley HD, Wiens S, Rotshtein P, Ohman A, Dolan RJ. Neural systems
supporting interoceptive awareness. Nat Neurosci 2004; 7: 189–95.
Cutting J. Study of anosognosia. J Neurol Neurourg Psychiatry 1978; 41:
548–55.
Damasio AR. Descartes’ Error: Emotion, Reason and the Human Brain.
New York: Putnam; 1994.
Delis DC, Kaplan E, Kramer JH. Delis-Kaplan executive function system. San
Antonio, TX: The Psychological Cooperation, Haarcourt Assessment
Company; 2001.
Desmurget M, Sirigu A. A parietal-premotor network for movement
intention and motor awareness. Trends Cogn Sci 2009; 13: 411–9.
Ellis E, Small M. Localization of lesion in denial of hemiplegia after acute
stroke. Stroke 1997; 28: 67–71.
Farrer C, Frith CD. Experiencing oneself vs another person as being the
cause of an action: the neural correlates of the experience of agency.
NeuroImage 2002; 15: 596–603.
Feinberg TE, Roane DM. Anosognosia, completion and confabulation:
the neutral - personal dichotomy. Neurocase 1997; 3: 73–85.
Feinberg TE, Roane TM, Ali J. Illusory limb movements for anosognosia
for hemiplegia. J Neurol Neurosurg Psychiatry 2000; 68: 511–13.
Fotopoulou A. False-selves in neuropsychological rehabilitation: the
challenge of confabulation. Neuropsychological Rehab 2008; 18:
541–65.
Fotopoulou A. The affective neuropsychology of confabulation and
delusion. Cogn Neuropsychiatry 2010; 15: 38–63.
Fotopoulou A, Conway MA. Confabulations pleasant and unpleasant.
Neuropsychoanalysis 2004; 6: 26–33.
Fotopoulou A, Tsakiris M, Haggard P, Vagopoulou A, Rudd A,
Kopelman M. The role for motor intention in motor awareness;
an experimental study on anosognosia for hemiplegia. Brain 2008;
131: 3432–42.
Fotopoulou A, Rudd A, Kopelman M. Self-observation reinstates motor
awareness in anosognosia for hemiplegia. Neuropsychologia 2009; 47:
1256–60.
Frith CD, Blakemore SJ, Wolpert DM. Abnormalities in the awareness
and control of action. Phil Trans Roy Soc Lond B Biol Sci 2002; 355:
1771–88.
Gerstmann J. Problems of imperceptions of disease and of impaired
body territories with organic lesions: relation to body scheme and its
disorders. Arch Neurol Psychiatry 1942; 48: 890–913.
Goldstein K. Human nature in the light of psychopathology. Cambridge,
MA: Harvard University Press; 1939.
Heilman KM, Barret AM, Adair JC. Possible mechanisms of anosognosia:
a defect in self awareness. Philos Trans R Soc Lond B Biol Sci 1998;
353: 1903–9.
Implicit awareness in anosognosia for hemiplegia
Heilman KM, Harciarek M. Anosognosia and anosodiaphoria of
weakness. In: Prigatano GP, editor. The study of anosognosia.
New York: Oxford University Press; 2010. p. 89–112.
Jenkinson PM, Fotopoulou A. Motor awareness in anosognosia for
hemiplegia: experiments at last!. Exp Brain Res 2010; 204: 295–304.
Johnson MK, Kim JK, Risse G. Do alcoholic Korsakoff’s syndrome
patients acquire affective reactions? J Exp Psychol Learn Mem
Cognit 1985; 11: 22–36.
Kampman M, Keijsers G, Verbraak M, Naring G, Hoogduin C.
The emotional stroop: a comparison of panic disorder patients,
obsessive-complusive patients, and normal controls, in two experiments. Anxiety Disorders 2002; 16: 411–25.
Kaplan-Solms K, Solms M. Clinical studies in neuro-psychoanalysis.
London: Karnac Books; 2000.
Karnath H-O, Baier B. Anosognosia for hemiparesis and hemiplegia: disturbed sense of agency and body ownership. In: Prigatano GP, editor.
The study of anosognosia. New York: Oxford University Press; 2010.
p. 39–62.
Karnath HO, Baier B, Nägele T. Awareness of the functioning of
one’s own limbs mediated by the insular cortex? J Neurosci 2005;
25: 7134–8.
Koenigs M, Huey ED, Calamia M, Raymont V, Tranel D, Grafman J.
Distinct regions of prefrontal cortex mediate resistance and vulnerability to depression. J Neurosci 2008; 28: 12341–8.
Koenigs M, Grafman J. The functional neuroanatomy of depression:
distinct roles for ventromedial and dorsolateral prefrontal cortex.
Behav Brain Res 2009; 201: 239–43.
Lauwereyns J, Watanabe K, Coe B, Hikosaka O. A neural correlate of
response bias in monkey caudate nucleus. Nature 2002; 418: 413–17.
Levine DN. Unawareness of visual and sensory motor defects: a hypothesis. Brain Cogn 1990; 13: 233–81.
Levine DN, Cavanio R, Rinn WE. The pathogenesis of anosognosia for
hemiplegia. Neurology 1991; 41: 1770–81.
Marcel AJ, Tegner R, Nimmo-Smith I. Anosognosia for plegia: specificity,
extension, partiality and disunity of bodily awareness. Cortex 2004;
40: 19–40.
Marshall J. and Halligan, PW. Blindsight and insight in visuo-spatial
neglect. Nature 1998; 336: 766–7.
McGlynn SM, Schacter DL. Unawareness of deficits in neuropsychological syndromes. J Clin Exp Neuropsychol 1989; 11: 143–205.
McIntosh RD, Brodie EE, Beschin N. Improving the clinical diagnosis of
personal neglect: a reformulated comb and razor test. Cortex 2000;
36: 289–92.
McKay R, Cipolotti L. Attributional style in a case of Cotard delusion.
Conscious Cogn 2007; 16: 349–59.
Mori S, Oishi K, Jiang H, Jiang L, Li X, Akhter K, et al. Stereotaxic white
matter atlas based on diffusion tensor imaging in an ICBM template.
NeuroImage 2008; 40: 570–82.
Nardone IB, Ward R, Fotopoulou A, Turnbull OE. Attention and emotion
in anosognosia: evidence of implicit awareness and repression.
Neurocase 2007; 13: 438–45.
Olausson H, Charron J, Marchand S, Villemure C, Strigo IA, Bushnell MC.
Feelings of warmth correlate with neural activity in right anterior
insular cortex. Neurosci Lett 2005; 389: 1–5.
Orfei MD, Robinson RG, Prigatano GP, Starkstein S, Rüsch N, Bria P,
et al. Anosognosia for hemiplegia after stroke is a multifaceted
phenomenon: a systematic review of the literature. Brain 2007; 130
(Pt 12): 3075–90.
Phipps MC. Inequalities between hypergeometric tails. J Appl Math Decis
Sci 2003; 7: 165–74.
Pia L, Neppi-Modona M, Ricci R, Berti A. The anatomy of anosognosia
for hemiplegia. Cortex 2004; 40: 367–77.
Pratto F, John O. Automatic vigilance: the attention–grabbing power of
negative social information. J Pers Soc Psychol 1991; 61: 380–91.
Ramachandran VS, Blakeslee S. Phantoms in the brain: the sound of one
hand clapping. New York: William and Co; 1998.
Ramachadran VS. Anosognosia in parietal lobe syndrome. Conscious
Cogn 1995; 4: 22–51.
Brain 2010: 133; 3564–3577
| 3577
Rorden C, Brett M. Stereotaxic display of brain lesions. Behav Neurol
2000; 12: 191–200.
Rorden C, Bonilha L, Nichols TE. Rank-order versus mean based statistics
for neuroimaging. Neuroimage 2007; 35: 1531–7.
Rosenberg M. Society and the adolescent self-image. Princeton, NJ:
Princeton University Press; 1965.
Rubens AB, Garrett MF. Anosognosia of linguistic deficits in patients with
neurological deficits. In: Prigatano GP, Schacter DL, editors. Awareness
of deficit after brain injury. Oxford: Oxford University Press; 1985.
p. 40–52.
Schacter DL. Toward a cognitive neuropsychology of awareness: implicit
knowledge and anosognosia. J Clin Exp Neuropsyc 1990; 12: 155–78.
Shallice T, Evans ME. The involvement of the frontal lobes in cognitive
estimation. Cortex 1978; 14: 294–303.
Solms M. Is the brain more real than the mind? Psychoanal Psychother
1995; 9: 107–20.
Solms M. The deep psychological functions of the right cerebral
hemisphere. Bull Brit Psychoanalytic Soc 1999; 35: 9–29.
Starkstein SE, Feodoroff JP, Price TR, Leiguarda R, Robinson RG.
Anosognosia in patients with cerebrovascular lesions. A study of
causative factors. Stroke 1992; 23: 1446–53.
Talairach P, Tournox J. A stereotactic coplanar atlas of the human brain.
Stuttgart: Thieme 1988.
Tranel D, Damasio AR. Knowledge without awareness: an autonomic
index of facial recognition by prospoagnosics. Science 1985; 228:
1453.
Tsakiris M, Hesse MD, Boy C, Haggard P, Fink GR. Neural signatures of
body ownership: a sensory network for bodily self-consciousness.
Cerebral Cortex 2007; 17: 2235–44.
Tsakiris M, Longo M, Haggard P. Having a body versus moving
your body: neural signatures of agency and body-ownership.
Neuropsychologia 2010; 48: 2740–9.
Tsakiris M. My body in the brain: a neurocognitive model of
body-ownership. Neuropsychologia 2010; 48: 703–12.
Turnbull OH, Jones CE, Reed-Screen J. Implicit awareness of deficit
in anosognosia: an emotion-based account of denial of deficit.
Neuropsychoanalysis 2002; 4: 69–86.
Turnbull OH, Evans CE, Owen V. Negative emotions and anosognosia.
Cortex 2005; 41: 67–75.
Tzourio-Mazoyer N, Landeau B, Papathanassiou D, Crivello F, Etard O,
Delcroix N, et al. Automated anatomical labeling of activations in
SPM using a macroscopic anatomical parcellation of the MNI MRI
single-subject brain. NeuroImage 2001; 5: 273–89.
Vocat R, Vuilleumier P. Neuroanatomy of impaired body awareness in
anosognosia and hysteria: a multi-component account. In:
Prigatano GP, editor. The study of anosognosia. New York: Oxford
University Press; 2010. p. 361–404.
Vuilleumier P. Anosognosia: the neurology of beliefs and uncertainties.
Cortex 2004; 40: 9–17.
Wechsler D. Wechsler Adult Intelligence Scale (WAIS-III). London:
The Psychological Corporation; 1998.
Wechsler D. Wechsler Test of Adult Reading (WTAR). San Antonio, TX:
The Psychological Corporation; 2001.
Weinstein EA, Kahn RL. Denial of illness: Symbolic and physiological
aspects. Springfield: Charles C Thomas; 1955.
Weiskrantz L, Warrington EK, Sanders MD, Marshall J. Visual capacity
in the hemianopic field following a restricted occipital ablation. Brain
1974; 97: 709–28.
Wentura D, Rothermund K, Bak P. Automatic vigilance: The attentiongrabbing power of approach- and avoidance-related information.
J Pers Soc Psychol 2000; 78: 1024–37.
Wilson BA, Cockburn J, Halligan PW. Behavioural Inattention Test.
Bury St Edmunds: Thames Valley Test Company; 1987.
Winward CE, Halligan PW, Wade DT. The Rivermead Assessment of
Somatosensory Performance (RASP): standardization and reliability
data. Clin Rehab 2002; 16: 523–33.
Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale
(HADS). Acta Psychiat Scand 1983; 67: 361–70.