ناتوانی صریح و ضمنی در ادراک بیماری و یا اختلال حرکتی اندام فوقانی

Abstract If asked directly, anosognosic patients deny or seriously underestimate their motor difficulties. However explicit denial of hemiplegia does not necessarily imply a lack of insight of the deficit. In this study we explored explicit and implicit awareness for upper limb motor impairment in a group of 30 right-brain damaged patients. Explicit awareness was assessed using a questionnaire (the VATAm) in which patients are asked to rate their motor abilities, whereas implicit awareness was assessed by means of a newly developed test (BMT – bimanual task). This test requires the performance of a series of bimanual tasks that can be better performed using two hands, but could also be performed using one hand only. With the BMT, patients’ performance rather than their verbal reports is evaluated and scored as an index of awareness. Paretic patients with anosognosia tend to approach these tasks as if they could use both hands. Our findings showed that explicit and implicit awareness for motor deficits can be dissociated, and they may be differently affected by feedback suggesting that different underlying mechanisms may account for the multi-factorial phenomenon of anosognosia.

1. Introduction There is growing agreement amongst researchers to consider anosognosia as a multi-factorial phenomenon (Cocchini et al., 2009, Cocchini et al., 2002, Davies et al., 2005, Marcel et al., 2004, McGlynn and Schacter, 1989, Orfei et al., 2007, Vallar and Ronchi, 2006 and Vuilleumier, 2004; see also Prigatano, 2010). This suggests that different mechanisms may be responsible for lack of awareness, and that different aspects of anosognosia should be considered (Heilman et al., 1998 and Marcel et al., 2004). Despite this possible heterogeneous scenario, most of the researchers’ attention has been directed to systematically investigating explicit forms of unawareness, mainly relying on meta-cognition tasks. In this context patients are typically asked to respond to more or less direct questions about their deficit (e.g. Berti, Làdavas, & Della Corte, 1996), rate their ability in performing motor tasks (e.g. Della Sala, Cocchini, Beschin, & Cameron, 2009) or estimate their performance before and after having performed specific tasks (e.g. Marcel et al., 2004). More subtle aspects of anosognosia, which also may provide relevant information about the nature of patients’ unawareness, have received only qualified attention. Implicit processing of the information about one's own condition is one such neglected aspect. Considering patients’ incidental comments or their behaviour in specific situations may reveal their actual beliefs (insight) about their conditions. For example, Ramachandran and Blakeslee (1998) anecdotally described a patient oblivious of his severe paresis who nevertheless commented, “I can’t wait to go back to two-fisted beer drinking” (Ramachandran & Blakeslee, 1998, p. 143). Similarly Berti, Làdavas, Stracciari, Giannarelli, and Ossola (1998) described a hemiplegic patient who denied any motor impairment but whose “tacit knowledge of her physical condition was apparent in most conversations” (p. 21). Marcel et al. (2004) compared patients’ ratings of their own motor abilities with their own ratings in judging how well the examiner would perform the same tasks had he been in the patient's current situation. The authors observed that up to 50% of right-brain damaged patients rated the examiner's ability consistently lower than their own. Nardone, Ward, Fotopoulou, and Turnbull (2007) described 2 anosognosic patients who showed an increment of simple reaction times when a word related to movement was presented with the target. These findings suggest that even patients who firmly deny any motor problem may have access to some type of implicit knowledge about their motor impairment. A possible dissociation between implicit and explicit knowledge, or “dissociation of knowledge” (Bisiach & Geminiani, 1991), has strong theoretical implications about the nature of anosognosia. Information about implicit awareness is fundamental to provide relevant support to theoretical approaches; indeed some of these imply some insight into the deficit. For example, the motivational theory suggests that anosognosia is a psychological defence mechanism (Turnbull and Solms, 2007a, Turnbull and Solms, 2007b, Weinstein and Kahn, 1955 and Weinstein, 1991); however in order to “trigger” the denial process the patient must have processed some information about the deficit and have some “knowledge” of the deficit. On the contrary, other theories postulate that anosognosic patients lack the necessary information about their motor deficit, or the feedback information is incorrectly interpreted, resulting in lack of explicit and implicit awareness of the deficit. For example, the feed-forward theory by Gold, Adair, Jacobs, and Heilman (1994; see Fotopoulou et al., 2008 for a revised interpretation) identifies in the intention to move the “defective” stage that leads to lack of awareness. According to this interpretation, anosognosics’ intentional systems do not formulate expectations about movements and the information about the feedback is correctly processed (i.e. no movement occurred); however since there is no mismatch between intention to move and feedback, the lack of movement is not interpreted as a failure. Other such theories suggest that anosognosia is linked to a malfunction in monitoring the impairment (Jenkinson et al., 2009 and Rubens and Garrett, 1991) or of the self-evaluation process (Levine, 1990). According to these theories, a lack of explicit knowledge of the motor impairment would be likely associated with lack of insight. Therefore, considering that anosognosia seems to be a multi-factorial phenomenon, different patients’ unawareness may underline different causes and further knowledge about patients’ implicit knowledge about the motor impairment would contribute to a better understanding of the nature of each patient's anosognosia and guide theoretical interpretation. Clearly patients’ knowledge of the deficit, at an explicit or implicit level, depends on the individual ability to perceive and process information from the surrounding environment and from the internal status of the body (Craig, 2010). Some authors have observed how simple exposure to a failure may considerably increase some patients’ explicit awareness. We could call this phenomenon “empirical learning”. Berti et al. (1996; see also Marcel et al., 2004) reported that some patients, but not all, tend to update the evaluation of their own motor ability after having performed bimanual or bipedal motor tasks. However, these authors, as many others, have described patients who vigorously denied any motor impairment despite clear evidence to the contrary (see Prigatano, 2010). Yet, some patients may still deny their deficit but react to the examiner's request to move the paretic limb in quite a defensive way (e.g. “You are beginning to ask me too many things!” Berti et al., 1998, p. 29). This suggests that anosognosia cannot be invariably accounted for as due to monitoring deficits or incorrect self-evaluation or lack of motor intention. Indeed, it seems that some patients have some knowledge of their deficit but that this information fails to reach more explicit levels of awareness. More systematic investigations about the effect of empirical learning on explicit and implicit forms of awareness could provide additional understanding of the nature of anosognosia. The aim of the present study is to identify different levels of awareness and investigate possible factors that may prevent patients from becoming fully aware of their motor impairment.

Results 3.1. Explicit anosognosia 3.1.1. The VATAm upper limb All 30 brain-damaged patients completed the VATAm and provided the expected rating to all the check questions. Thirteen (43%) showed evidence of anosognosia for their motor impairment. Of them, 3 (10%) showed mild, 7 (23%) moderate and 3 (10%) severe anosognosia. Of these 13 patients, 8 (61.5%) had lesions encompassing the frontal or the parietal lobe. None of the control patients showed evidence of anosognosia. 3.2. Implicit anosognosia 3.2.1. The bimanual task The first attempt with the BMT represents the patients’ knowledge about their own motor impairments, with no previous direct experience of failure or partial failure at that specific task. The following two attempts indicate the patients’ awareness following a recent direct experience of failure or difficulty in performing the required task. Therefore, performance during first attempt was considered separately for second and third attempts. 3.2.2. BMT – first attempt The control groups’ average error score was 2.31/24 (sd = 2.46; range 0–9). Healthy volunteers obtained an average error score of 2.84/24 (sd = 2.56; range = 0–9), and no effect of the hand used was found (F(1,18) = 1.85; n.s.). Control patients obtained an average error score of 0.86/24 (sd = 1.46; range = 0–4). Brain damaged patients obtained an average error score of 5.27/24 (sd = 6.16; range = 0–19). Performance of the three groups differed significantly (F(2, 55) = 3.06; p < .05). Differences in performance amongst the three groups were further investigated by means of Dunnett post hoc analysis, which showed a significant difference between brain damaged patients and control patients (p < .005). The trend of a poorer performance of the healthy volunteers compared to the control patients might be due to the fact that the healthy controls were relatively new to this condition, while the control patients have been exposed to similar situations before, hence they could benefit from previous experience. However, the performance between the two control groups did not differ significantly suggesting no obvious effect of education in performing the BMT. In order to interpret each individual brain damaged patient's performance as normal or pathological, we identified a cut-off value, which indicated the threshold between normal and pathological performance. According to Crawford's modified t-test ( Crawford & Garthwaite, 2007), an error score of 9 is significantly (with p < .01) different from the control groups’ performance. Note that none of the control participants performed above this cut-off. Hence an error score equal to or over 9 was considered as evidence of implicit anosognosia. Seven (23%) brain damaged patients showed evidence of implicit anosognosia, and their average error score was15.29 (sd = 2.93; range = 10–19). Five of these 7 patients (71.4%) had brain lesions encompassing the fronto-parietal areas. 3.3. Explicit and implicit anosognosia dissociations 3.3.1. VATAm and BMT We next compared the performance of the 30 brain damaged patients on the VATAm upper limb and the first attempt of the BMT. Five (17%) patients showed anosognosia on both tests, while 15 (50%) did not show signs of unawareness on either test. However, 2 (7%) patients (cases no. 1 and 3) showed anosognosia on the BMT only, whereas 8 (27%) patients (cases no. 5, 6, 12, 15, 20, 22, 23, 24) showed anosognosia on the VATAm only. To overcome the possible criticism of a weak double dissociation mainly resulting from the patients’ scores close to cut-offs, we considered also very overt cases of anosognosia, i.e. VATAm upper limb scores over 8.1, which is the cut-off for moderate anosognosia (Della Sala et al., 2009), and BMT error scores over 12.22, which is the error score at 4 standard deviations from our control groups’ mean. Similarly, we considered only cases of obviously normal performance, i.e. VATAm upper limb scores lower than 2.56, which is the score within 1 standard deviation from the normative group reported in Della Sala et al. (2009), and BMT error scores lower than 5.13, which is the value within 1 standard deviation from our control groups’ mean. According to this stricter criterion, 7 cases of double dissociations could still be singled out: 2 patients showed severe anosognosia solely on the BMT, while 5 patients showed severe anosognosia on the VATAm only. 3.4. The effect of empirical learning 3.4.1. BMT – second and third attempts Both control groups generally performed close to ceiling on the first attempt, however they improved even further on the second and third attempts. Control patients obtained an identical performance on second and third attempts with an average error score of .57/24 (sd = 1.51; range = 0–4); whereas healthy volunteers obtained an average error score of .42/24 (sd = .96; range = 0–4) and of .32 (sd = .75; range = 0–4) on second and third attempts, respectively. The difference between the two attempts was not significant (F < 1). None of the 23 brain damaged patients performing within the normal range during the first attempt showed anosognosia on second and third attempts. The group performance on the second (error score = 4.20/24; sd = 5.31; range = 0–17) and the third (error score = 2.37/24; sd = 4.55; range = 0–17) attempts was better than on the first attempt (see Section 3.2.2: average error score = 5.27/24; sd = 6.16; range = 0–19). Error scores of the three test attempts were entered in a repeated measures ANOVA that showed a significant improvement across the three attempts (F(2,58) = 11.24; p < .001). Post hoc paired t-tests showed that each attempt significantly differed from the previous one (first–second attempts with p < . 05; second–third attempts with p < .005; first–third attempts with p < .001). To better investigate the impact of empirical learning, we considered individual performance over the three attempts of the 7 brain damaged patients who showed implicit anosognosia during the first attempt. Despite all of them showing at least some improvement across the three attempts, only 5 (cases no. 1, 3, 7, 10, 18) scored within the normal range on the second or the third attempts. Interestingly, only 3 of them (cases no. 1, 7 and 10) performed within the normal range on a follow-up assessment 3 days later, suggesting that some patients regained awareness only temporarily. Both group and individual analyses showed that most of the brain damaged patients tend to improve their performance (modifying their behaviour accordingly to the motor deficit), and then increase their awareness, following the simple exposure to a total or partial failure while they attempt the task.