ناتوانی در ادراک بیماری برای آپراکسی: شواهد تجربی برای آگاهی ناقص از حرکات صورت فرد

Abstract Anosognosia is a multifaceted, neuro-psychiatric syndrome characterized by defective awareness of a variety of perceptuo-motor, cognitive or emotional deficits. The syndrome is also characterized by modularity, i.e., deficits of awareness in one domain (e.g., spatial perception) co-existing with spared functions in another domain (e.g., memory). Anosognosia has mainly been reported after right hemisphere lesions. It is however somewhat surprising that no studies have thus far specifically explored the possibility that lack of awareness involves apraxia, i.e., a deficit in the ability to perform gestures caused by an impaired higher-order motor control and not by low-level motor deficits, sensory loss, or failure to comprehend simple commands. We explored this issue by testing fifteen patients with vascular lesions who were assigned to one of three groups depending on their neuropsychological profile and brain lesion. The patients were asked to execute various actions involving the upper limb or bucco-facial body parts. In addition they were also asked to judge the accuracy of these actions, either performed by them or by other individuals. The judgment of the patients was compared to that of two external observers. Results show that our bucco-facial apraxic patients manifest a specific deficit in detecting their own gestural errors. Moreover they were less aware of their defective performance in bucco-facial as compared to limb actions. Our results hint at the existence of a new form of anosognosia specifically involving apraxic deficits.

Introduction The term anosognosia, initially coined to indicate the denial of motor deficits contralateral to a brain lesion (Anosognosia for Hemiplegia – AHP, Babinski, 1914), refers to a multifaceted syndrome where patients who have suffered strokes, traumatic brain injury, degenerative diseases or neuropsychiatric disorders show complete or partial lack of awareness of a variety of neurological and cognitive deficits (e.g., hemianopia, blindness, hemianesthesia, neglect, aphasia, amnesia) ( Prigatano, 2010). Clinical reports and experimental studies on AHP (Cocchini et al., 2010a, Moro, 2013, Moro et al., 2011 and Ramachandran, 1994) and Alzheimer's Disease (Mograbi & Morris, 2013) show distinct types of anosognosia where implicit and emergent residual forms of awareness are present. Moreover, patients may exclusively deny their own paralysis but recognize deficits in other patients (1st person deficit) or fail to recognize motor impairment both in themselves and in other subjects (1st and 3rd person deficit) (Marcel et al., 2004 and Moro et al., 2011). Despite the steady increase of interest in anosognosia, no studies have thus far specifically set out to investigate the possible existence of anosognosia for apraxia (AA). The term apraxia refers to a wide spectrum of disorders with in common an inability to perform skilled or learned purposeful gestures. Although sometimes co-existing with motor or sensory deficits or language disorders, apraxia is not explained by any of these (Zadikoff & Lang, 2005). Conceptual and production components of gestural organization may be differentially affected, leading to ideational (i.e., defective action and object-use knowledge) or ideomotor apraxia (i.e., defective action execution in gesture pantomime and imitation) (Leiguarda & Marsden, 2000). In limb apraxia, imitation of transitive gestures (e.g., hammering a nail) is more impaired than the imitation of intransitive gestures (e.g., waving goodbye) (Buxbaum, Kyle, & Menon, 2005). Among the body-part defined subtypes of apraxia, bucco-facial apraxia (BFA) refers to an inability to voluntarily control facial, lingual, pharyngeal and masticatory actions (e.g., protruding tongue, blinking eyes) on purpose but not in ecological situations, when the movements are automatically performed. Neuropsychological and neuroanatomical results indicate that BFA and limb apraxia are at least partially independent (Raade, Rothi, & Heilman, 1991). While limb apraxia appears to be more commonly associated with left frontal and parietal brain damage (Pazzaglia, Smania, Corato, & Aglioti, 2008), BFA follows lesions in left prefrontal areas, the central operculum, the insula, the centrum semiovale, their subcortical projections and the basal ganglia (Pramstaller & Marsden, 1996). In this study we investigated the existence of a specific form of AA related to the possibility that subjects presenting with apraxia also show defective awareness of their difficulties. With this aim we asked patients with or without BFA to judge the correctness of bucco-facial and limb related gestures performed by themselves or by a gender-matched model. The patients' responses were compared with the evaluations provided by their therapist and caregiver. The comparison between discrepancies in judgment of actions performed by themselves or others allowed us to distinguish deficits in awareness from non-specific difficulties in action recognition.

4. Results Errors. Although A+ made more errors (n.42) than A− (n.35) and RBD (n.24), only the comparison between RBD and A+ was statistical significant (χ21 = 5.73, p = .017, ϕ = .151, p Bonferroni corrected = .05). Importantly, while in actions involving limbs (A+: 23, A−: 22, RBD: 18) there were no differences between the groups, in actions involving the mouth the A+ group failed in more cases than the other groups (A+ = 19, A− = 13, RBD = 6). Only the comparison between A+ versus RBD gave a statistically significant result (χ21 = 7.68, p = .006, ϕ = .151, p Bonferroni corrected = .018). Separating transitive and intransitive gestures, the same difference between these two groups was present for intransitive actions (χ21 = 9.16, p = .0025, ϕ = .27, p Bonferroni corrected = .007). The A+ patients frequently failed the mouth movements or executed unrecognizable actions (Table 3) while the RBD patients made spatial errors in trajectory and direction and the A− group made errors in configurations involving arms/hands and mouth but never executed totally different or unrecognizable actions. Table 3. Analysis of errors. Topography Typology of gesture Typology of error Transitive Intransitive Hand arm Mouth Spatial Other A+ A+1 Mouth 1 5 1.00 Limbs 3 2 .40 .60 A+2 Mouth 3 .33 .67 Limbs 1 4 .80 .20 A+3 Mouth 2 .50 .50 Limbs 2 .50 .50 A+4 Mouth 4 .25 .25 .50 Limbs 2 6 .63 .38 A+5 Mouth 4 1.00 Limbs 4 .50 .50 A− A−1 Mouth 1 1.00 Limbs 2 3 .80 .20 A−2 Mouth 2 1.00 Limbs 1 1.00 A−3 Mouth 2 .50 .50 Limbs 3 4 .29 .71 A−4 Mouth 3 1.00 Limbs 2 3 .40 .60 A−5 Mouth 2 3 1.00 Limbs 2 4 .50 .50 RBD RBD1 Mouth 1 1.00 Limbs RBD2 Mouth 2 1.00 Limbs 1 2 1.00 RBD3 Mouth 1 1.00 Limbs 4 5 .11 .89 RBD4 Mouth 2 .50 .50 Limbs 4 2 .17 .83 RBD5 Mouth No errors Limbs The number of errors are reported for Topography (Mouth, Limbs) and Typology of Gesture (Transitive, Intransitive). Types of errors (Hand/Arm: hand/arm configuration, Mouth: mouth configuration, Spatial: error in measure or trajectory, Other: not recognizable or totally different action) are computed as a proportion of the total number of errors. Table options In general, when their judgment was incorrect, both patients and observers evaluated incorrect actions as being correct. Only in a very few trials did the subjects state that a correct action was incorrect. Control videos. In the ten actions used as controls in order to exclude any general difficulties in action recognition, all three groups performed well with no differences when they were compared with the average evaluations of the caregivers/therapists (A+: z = .56, p = .58, CI = −.58, 1.04; A−: z = −.10, p = .92, CI = −.79, .79; RBD: z = 0, p = 1, CI = −.84, .75). Phase 1. On-line judgment. The meta-analyses carried out on all 28 actions indicated that when asked to judge their own gestures directly after execution, all three groups recognized fewer errors than their observers (A+: z = 2.99, p = .003, CI = .47, 2.24; A−: z = 2.96, p = .003, CI = .46, 2.28; RBD: z = 1.96, p = .05, CI = .00, 2.12). Crucial to our study, in the limb actions both the A+ and A− groups failed in the judgments of their own actions (A+: z = 2.01, p = .04, CI = .3, 2.18; A−: z = 2.57, p = .01, CI = .38, 2.82; RBD: z = 1.70 p = .09, CI = −.17, 2.32), while in the mouth actions only group A+ (z = 2.35, p = .02, CI = .23, 2.60; A−: z = 1.42, p = .15, CI = −.33, 2.07; RBD: z = 1.11, p = .26, CI = −.59, 2.16) failed in their judgments ( Fig. 3). A statistical difference was also present for intransitive actions (mouth + limb) in left damaged groups (A+: z = 3.24, p = .001, CI = .73, 2.97; A−: z = 2.56, p = .01, CI = .32, 2.38; RBD: z = 1.54, p = .12, CI = −.24, 2.03) ( Fig. 3). Judgment of actions executed by the patients. Summary effects from the ... Fig. 3. Judgment of actions executed by the patients. Summary effects from the meta-analyses of the comparison between the patients' and the observers' judgments. Effects are separately reported for each group, phase, topography and typology of action. A 0 score indicates the correspondence between patient and observer judgments. Scores below 0 indicate that the patient's judgments are more severe than those of the observers (underestimation), and score over 0 indicates that the patient judges the actions more leniently than the observers (overestimation). ° = trend towards statistical significance (p = .06); * = p ≤ .05; ** = p ≤ .01. Intervals of confidence (CI) are reported in the text. Figure options In conclusion, some degree of difficulty in judging one's own action immediately after the execution was found in all the left brain damaged patients. Importantly, however, this was not specific for the facial or limb actions. Therefore, this difficulty may be considered as a generic impairment likely due to the role of the left hemisphere in motor planning rather than to apraxia. Phase 2. Off-line judgment. When asked to judge their own actions shown in a video, only group A+ failed (z = 2.54, p = .01, CI = .23, 1.75; A−: z = .80, p = .42, CI = −.42, 1.01; RBD: z = .86, p = .39, CI = −.52, 1.34). Dividing the gestures into categories depending on topography, a trend towards statistical significance was observed, with the A+ group failing exclusively in the judgment of mouth actions (z = 1.84, p = .066, CI = −.18, 1.77; A−: z = .16, p = .87, CI = −.97, 1.14; RBD: z = .28, p = .78, CI = −1.1, 1.48). In addition, the same group failed in the evaluation of intransitive (z = 3.06, p = .002, CI = .57, 2.58; A−: z = .21, p = .83, CI = −.77, .95; RBD: z = .61, p = .54, CI = −.71, 1.36) but not transitive gestures ( Fig. 3). Self versus other. Separating judgments concerning patients' incorrect actions from judgments referred to incorrect actions performed by models, only the A+ group fail (A+: z = 2.47, p = .013, CI = .3, 2.64; A−: z = −.59, p = .56, CI = −1.56, .84; RBD: z = .42, p = .67, CI = −.96, 1.5) and this was exclusively with reference to patients' actions ( Fig. 4). Judgment of actions performed by the patient themselves and those performed by ... Fig. 4. Judgment of actions performed by the patient themselves and those performed by others. Summary effects from the meta-analyses of the comparison between the patients' and the observers' judgments are reported, separated for the three groups. Figure options Individual data are discussed in the Supplemental Material.