Body Awareness Disorders: Dissociations Between Body-related Visual and Somatosensory Information

Laure Pisella; Laurence Havé; Yves Rossetti

Disclosures

Brain. 2019;142(8):2170-2173. 

This scientific commentary refers to 'The antero-dorsal precuneal cortex supports specific aspects of bodily awareness', by Herbet et al. (doi:10.1093/brain/awz179).

Information about the body arrives in numerous forms: tactile, proprioceptive, nociceptive, vestibular, visual, motor, cognitive and semantic. Only a limited proportion of this input is processed consciously and we remain unaware of many of the processes necessary to maintain postural control and to perform actions. Body cognition is frequently viewed as dyadic or triadic, with a distinction made between the body schema and the body image (Longo et al., 2010; Pitron et al., 2018). Body awareness disorders (BADs) are uncommon in neurology. They may involve distorted perception or reduced awareness of parts of the body or an impaired ability to localize them, feelings of strangeness, loss of agency, and illusory feelings of movement or even of having supernumerary limbs.

BADs are often associated with syndromes arising as a result of parietal lobe lesions (Figure 1), although they may also occur following lesions in other locations. Lesions may give rise to deficits in either the body image or the body schema. Unilateral neglect, for example, typically results from a right hemisphere lesion and may manifest in the body schema domain (motor neglect, directional hypokinesia or bradykinesia) or the body image domain (e.g. tactile extinction or allochiria, personal neglect, anosognosia for hemiplegia, hemiasomatognosia, somatoparaphrenia). Optic ataxia and apraxias—which reflect body schema deficits—may result from lesions of either the left or right hemispheres. Left posterior parietal lobe lesions can lead to Gerstmann syndrome, involving finger agnosia (an inability to distinguish the fingers on the hand) and left-right disorientation, which may also apply to body parts, both of which can be considered body image impairments. An anterior lesion of the left superior parietal lobule including the precuneus has been reported to cause illusory drift followed by fading limb in the absence of vision (Wolpert et al., 1998). However, no such body image impairment was reported in association with optic ataxia following posterior lesion of the left or right superior parietal lobule; by contrast these latter patients produced large localization errors in a proprioceptive pointing task, indicating body schema deficits (Blangero et al., 2007). While BADs are widely acknowledged to be associated with syndromes consecutive to parietal lobe lesions, the reasons for their presence or absence in individual patients have often been something of a mystery. In this issue of Brain, Herbert and co-workers provide a possible explanation for the scarcity of BADs and their non-systematic association with the more classical parietal syndromes: BADs may arise only when a lesion of posterior parietal cortex also involves the medial area, i.e. the precuneus, and more specifically the antero-dorsal precuneus (Herbet et al., 2019).

Figure 1.

Classification of body schema and body image disorders along the dimension proposed by Longo et al. (2010). 'Somatosensation' corresponds to primary sensory processing, 'somatoperception' to the localization of somatic and postural stimuli subtended by the parietal lobes, and 'somatorepresentation' to general semantic knowledge and body consciousness about arrangement and metric properties of body parts subtended by the frontoparietal network. Schematic representation of their functional relationship based on Rossetti et al. (2005) and Pitron et al. (2018).

Neuroimaging data have implicated the antero-dorsal precuneus in strikingly heterogeneous functions ranging from motor coordination, drawing and imitation to working memory and subject-verb integration (from the Neurosynth online tool http://neurosynth.org/locations/6_-52_66_3/). By contrast, human brain lesions rarely affect this medial brain territory, explaining why BADs after parietal stroke, and specific neuropsychological reports on this region both remain scarce. But while circumscribed precuneus lesions are rather rare, Herbet et al. managed to identify and assess 14 patients who underwent surgery for glioma infiltrating the precuneus. They tested a large number of neuropsychological functions prior to, 5 days after, and 3 months after the surgery. However, the main aim of the study was to assess the BADs that were frequently observed in these patients, mostly at the acute stage. Seven of the patients showed BADs, which allowed Herbet et al. to functionally distinguish the antero-dorsal and the postero-dorsal precuneal regions. In addition, they mapped the functional connectivity of these two anatomical seeds in healthy controls. Herbet et al. established that the resection cavity of the seven patients showing BADs always included the antero-dorsal part of the precuneus. The resting state analysis revealed that this antero-dorsal region is involved in a broad sensorimotor network, suggesting that body schema disorders may result in body image ones. There are indeed complex interactions between body schema and body image but the body schema seems to have some primacy over the body image (Rossetti et al., 2005; Pitron et al., 2018). The most frequent BADs observed by Herbet et al. were tactile alien hand, fading limb, motor neglect and tactile autotopoagnosia. Moreover, these symptoms were often associated (Patients P2 and P11 present both tactile alien hand and fading limb, Patient P6 presents motor neglect, tactile alien hand and autotopoagnosia, Patient P14 shows tactile alien hand and autotopoagnosia). What are the mechanisms underlying these typically precuneus-related BADs?

When a limb is visible, its proprioceptive and visual localizations will differ slightly but will be integrated into an averaged position estimate. In the absence of direct vision of the limb, this position estimate will drift toward the proprioceptive location. This body schema perturbation is revealed by the progressive movement drift repeatedly observed toward the same non-random location when a healthy individual alternates pointing movements between two visual targets in the dark (Smeets et al., 2006). When a large location mismatch is experimentally induced between vision and proprioception in healthy subjects, it has been shown to be resolved by selective fading of the affected limb (Hogendoorn et al., 2009) or to impair tactile perception on this limb (Folegatti et al., 2009). Reciprocally, while touch may not provide additional information about finger position in space, as fingertip tactile stimuli should remain independent from the postural configuration of the upper limb, additional tactile information nevertheless improves proprioceptive localization (Mikula et al., 2018). This may be due to the fact that tactile and proprioceptive information are ultimately coded within the same population of posterior parietal neurons within high level spatial representations. Antero-dorsal precuneus damage may affect multisensory integration and specifically lead the visual and proprioceptive experiences of a limb to largely depart from each other. Consequences may include feelings of strangeness when the limb is visible or of illusory movement when vision of the limb is prevented inducing a large position estimate drift. Moreover, if the limb proprioceptive representation has low weighting, its perception when visual feedback is removed may fade and it may disappear not only from body image but also from body schema (fading limb and motor neglect, respectively). Such neurologically induced visual-proprioceptive mismatch also affects tactile perception on the limb concerned. If the proprioceptive limb representation is unstable, it may result in tactile autotopoagnosia. If it fades from body image, it may produce tactile alien hand.

When an additional primary somatosensory deficit is present, as in Patient P2 of Herbet et al. whose resection cavity extended to the paracentral lobule, macrosomatognosia may also be observed. We had the opportunity to follow a patient (Patient IK) who exhibited sustained BADs for several years following surgery for an extradural haemorrhage. These disorders notably included macrosomatognosia of the right upper limb, in association with tactile autotopoagnosia (without heterotopoagnosia) and tactile but also motor alien hand (which has also been reported after right precuneus damage, Darby et al., 2018). Patient IK's left parietal cortex lesion included the precuneus, and it also impaired primary proprioceptive function in isolation. During tactile examination with a blindfold, Patient IK exhibited fully normal thresholds but he would often complain that he had no idea where the stimulus was applied. He would even sometimes burst out: 'you cannot touch me here because it is not on my body!'. Over several years, Patient IK spontaneously reported the perception of three right upper limbs: a visual one, an imagined one and a somatosensory one. These three hands were more independent of each other during the night, when the patient was lying in bed in the dark. Interestingly, Patient IK also initially complained of hand levitation, i.e. his contralesional hand would raise itself in the air and remain floating until he looked at it. In daily life, he could be surprised by the sight of his right hand grasping the mirror while he was driving his car, as if his hand was levitating devoid of any notion of agency. In other experimental conditions, the patient was asked to sit at a table, with his right hand hidden under the table. An examiner sat behind him, and placed his own hand on the table, to the right of the patient. Within a few seconds, Patient IK's body cognition took ownership of the alien hand despite his knowing that this was obviously impossible. He was horrified or disgusted whenever he saw the alien right hand moving while his own remained hidden. In addition, the patient often became distressed when the examiner removed his arm from the table without having warned him. In short, this patient was facing conflicting perceptions and representations at different levels (sensory and cognitive) in the absence of vision of his own hand.

Contrary to the general idea that BADs are predominantly associated with neglect and right hemisphere damage, the literature, in line with the Herbet et al. data, does not provide evidence for such hemispheric lateralization. This is consistent with the idea that BADs result from body schema deficits that are usually not lateralized to one hemisphere (like optic ataxia). One might think that this absence of hemispheric lateralization of the BADs in Herbet et al. may be related to the fact that the deficits exhibited by their patients are predominantly present at the acute postoperative stage. Indeed, in the case of neglect, acute studies often report that symptoms are as frequent following lesions of either hemisphere. It is only at the chronic stage that the right hemisphere specialization is revealed. One may speculate that the same phenomenon applies to BADs. However, two of four patients presenting with BADs at 3 months post-surgery had a left hemispheric lesion in Herbet et al. and Patient IK showed chronic BADs for several years also following a left hemispheric lesion. This suggests that BADs may not be lateralized even at the chronic stage. Unlike some overt impairments such as dressing apraxia, BADs may alter body cognition more implicitly and remain undetected. Following Herbet et al. it is therefore recommended to researchers and clinicians to carefully interview and examine patients for BADs when the antero-dorsal precuneus is lesioned, irrespective of the hemispheric side.

processing....