Ann. N.Y. Acad. Sci. ISSN 0077-8923

A N N A L S O F T H E N E W Y O R K A C A D E M Y O F SC I E N C E S Issue: The Year in Cognitive Neuroscience

Mapping the functional neuroanatomy of spatial neglect and human parietal lobe functions: progress and challenges Patrik Vuilleumier Laboratory for Behavioral Neurology and Imaging of Cognition, Department of Neuroscience, Medical School, and Department of Neurology, University Hospital of Geneva, University of Geneva, Geneva, Switzerland Address for correspondence: Patrik Vuilleumier, LABNIC/NEUFO, University Medical Center, University of Geneva, Michel-Servet 1, 1211 Geneva, Switzerland. [email protected]

Spatial neglect is generally defined by various deficits in processing information from one (e.g., left) side of space contralateral to focal (e.g., right) hemisphere damage. Although classically associated with parietal lobe functions, there is now compelling evidence that neglect can follow lesions in many different cortical and subcortical sites, suggesting a dysfunction in distributed brain networks. In addition, neglect is likely to result from a combination of distinct deficits that co-occur due to concomitant damage affecting juxtaposed brain areas and their connections, but the exact nature of core deficits and their neural substrates still remains unclear. The present review describes recent progress in identifying functional components of the neglect syndrome and relating them to distinct subregions of parietal cortex. A comprehensive understanding of spatial neglect will require a more precise definition of cognitive processes implicated in different behavioral manifestations, as well as meticulous mapping of these processes onto specific brain circuits, while taking into account functional changes in activity that may arise in structurally intact areas subsequent to damage in distant portions of the relevant networks. Keywords: spatial neglect; attention; lesion mapping; parietal cortex; network; connectivity

Introduction Hemispatial neglect is a frequent and fascinating, but still poorly understood, neuropsychological disorder. Although it has hitherto defied a comprehensive theoretical account, research on neglect in the past three decades has yielded a vast and rich body of knowledge concerning various domains of perception, attention, and spatial cognition. Moreover, this research has laid a fertile ground for integrating clinical neuropsychology with basic neuroscience by linking particular behavioral phenomena and highlevel mental functions, such as awareness and attention, with specific neuronal properties measured at the single-cell level. More recent progress in neuroimaging techniques in humans, and parallel neurophysiology approaches in nonhuman primates, has further encouraged a meticulous mapping of the neural circuits whose damage may lead to neglect. However, many questions remain open concerning

the exact nature of the functions subserved by these circuits and their link with specific neglect symptoms. This review will provide a selective overview of current approaches and challenges to defining the functional neuroanatomy of spatial neglect, with a particular focus on parietal lobe functions, which have classically been associated with this disorder.1,2 Hemispatial neglect is frequently observed after unilateral hemispheric brain damage in humans, most often due to stroke.3,4 It is clinically defined as a failure to perceive, report, and orient to sensory stimuli on the side of space opposite the brain lesion. Most often, such lesions affect the right hemisphere and neglect concerns the left space. For example, patients with this syndrome may not see people or objects located on their left side, fail to hear voices or sounds coming from that side, miss words on the left half of a page or even the left half of words when reading, or eat only from the right side of their plate (Fig. 1). These symptoms cannot be doi: 10.1111/nyas.12161

50

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

Spatial neglect and the parietal lobe

Figure 1. Examples of tests used to assess spatial neglect. (A) Line bisection requires marking the midpoint of lines of various lengths, presented one or many at a time. Neglect patients typically deviate rightward away from the true midpoint, especially when the line is long. They may also omit lines on the left side of the page. (B) Cancellation requires searching and marking a target among distractors, for example, letters or shapes like the bells illustrated here. Neglect patients typically fail to explore and detect targets on the left side of the display. (C) The Ota cancellation task requires encircling all rings without a gap and crossing out all those with a gap (on the circle’s left or right side). This may reveal misses of targets on the left side of the sheet (space-based or egocentric neglect) as well as misses of targets with a left gap (object-based or allocentric neglect). Object-based neglect is also observed in other tasks, for example, when patients omit (or transform) only the left part of compound words (e.g., house instead of doghouse). (D) Copy of line drawing of a multi-item scene. This task may also reveal both omissions in the left space and omissions of the left side of items, as illustrated here by the performance of two distinct patients (top vs. bottom sample). Drawings also often show an apparent compression of space with items clumped together on the right side, but underuse of the left side of the sheet. (E) Perceptual extinction is elicited by comparing the detection of two stimuli presented simultaneously on either side of space, relative to the detection of the same stimuli presented alone. Patients without hemianopia (or sensory loss) often fail to perceive the contralesional (left) stimulus in bilateral but not unilateral stimulation. Such extinction can be observed in different sensory modalities.

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

51

Spatial neglect and the parietal lobe

Vuilleumier

explained by elementary losses such as visual field cuts (hemianopia), deafness, or paralysis—although the latter deficits can influence or exacerbate some manifestations of neglect. Neglect for contralesional space may affect all sensory modalities (e.g., vision, audition, touch) to various degrees, but may also arise in mental imagery (e.g., when describing familiar places from memory as seen from a given standpoint5 ), in estimation of magnitude along a mental number line,6,7 and even in mental representation of time (e.g., when attributing events to past or future along a time line8 ). Neglect may also be observed in motor behavior in the absence of any external stimuli: for example, when searching for a potential target in complete darkness or blindfolded,9,10 patients move their eyes or hands into the ipsilesional (right) half of space but fail to explore the contralesional (left) side. Patients may neglect parts of their own body and fail to use their contralesional arm or leg, even though it is not paralyzed. Perceptual extinction is another frequent symptom whereby a stimulus in contralesional space (in any sensory modality) is not detected when presented simultaneously with another stimulus in ipsilesional space (in the same or different modality), while an identical contralesional stimulus is detected when presented alone.11,12 Strikingly, neglect patients typically do not realize that they miss information or ignore one portion of space, suggesting that anosognosia for neglect is an intrinsic feature of the disorder which can dissociate from other forms of anosognosia.13 Because these patients have no direct experience of their abnormal spatial biases, they fail to spontaneously compensate for such impairment (even when they can verbally report having been told about it by, for example, doctors or therapists). Not only does spatial neglect entail puzzling deficits in patients and raise fascinating questions on the neural bases of consciousness and space representation, but it also has important clinical implications for the management of brain-damaged patients and their rehabilitation. First, the presence of neglect predicts worse recovery of concomitant neurological deficits (e.g., paralysis), reduced independence in everyday life, and hence heavier burden for caregivers and relatives. Second, therapies are still limited and lead to partial improvements, often restricted to the training situation without generalizing to everyday life. Fortunately, the severity of neglect and associated symptoms tend to show 52

some degree of progressive spontaneous recovery in the months following the acute brain insult, even though some symptoms may persist for many years. Understanding neural dysfunctions that underlie neglect symptoms and their recovery is therefore an important step to envision more efficient interventions and improve rehabilitation approaches. In recent years, neuroscience research on neglect has begun to yield several new mechanistic insights that will hopefully open new opportunities for therapeutic interventions. Although there is no unified theoretical framework to explain neglect, and controversies still exist concerning its exact neural underpinnings (as will be discussed below), there is now general agreement that a core deficit of neglect involves brain mechanisms controlling the orientation of attention in space.14,15 A profound impairment in directing attention to the contralesional side of space, objects, or one’s own body might account for the striking loss in perception and action toward that side, as well as the modulation of such deficits by various procedures enhancing attention toward that side. In the healthy brain, attention encompasses a number of distinct processes that play critical roles in selecting sensory information and motor plans for conscious awareness and goal-oriented behavior.16,17 Abundant research has shown that normal people may also fail to perceive and respond to stimuli when attention is not directed to them, just like neglect patients ignore information in their contralesional space.16 Neuroscience work has substantially broadened our knowledge of these attentional mechanisms and begun to dissect their anatomical and physiological components. This research has contributed to a better understanding of neglect symptoms and their specific neural substrates,3,17,18 while neuropsychological studies of neglect have enriched and motivated research on normal attentional mechanisms, using neuroimaging in humans15 and neurophysiological methods in nonhuman primates.19 However, attentional accounts alone are not sufficient to explain all neglect manifestations, particularly dissociations between different sectors of space (e.g., near vs. far,20 front vs. back21 ) or the existence of nonspatial/nonlateralized deficits.22 Much therefore remains to be determined about the exact nature of the attentional deficits underlying neglect and its diverse symptoms, their functional relationship to various representations of space in

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

the brain, as well as the specific role of different brain regions or circuits in each of these processes. Increasing evidence indicates that spatial neglect may involve a combination of impairments in different component processes, and that behavioral deficits emerge as a result of dysfunction within large-scale brain networks whose activity is disrupted beyond local areas of structural damage. Yet, these functional components and the corresponding networks are still largely unresolved. The present review provides a selective overview of current knowledge and highlights some outstanding questions. It is argued that describing neglect symptoms in terms of more specific cognitive processes and relating them to precise neurophysiological substrates constitute an important challenge for future neglect research. This is essential to better characterize the behavioral impairment of individual patients and design appropriate remediation strategies. This is also crucial to foster translation from animal research, where a complete model of spatial neglect has not been successfully developed and more selective manipulations of brain function can be tested. This selective overview will specifically focus on the functional neuroanatomy of putative components of neglect and recent work mapping its anatomy in the parietal lobe and interconnected brain regions. A particular emphasis will be placed on data highlighting the role of distributed brain networks and functional interactions across distant regions. These data illustrate the limitations of a strict “localizationist” approach of neuropsychological deficits and underscore the value of combining functional neurophysiological measures with more classic structural analyses of brain injury. Clinical and anatomical variability of spatial neglect The clinical presentation of neglect is heterogeneous and variable across patients. Accordingly, in neuropsychology practice, the presence of neglect is typically assessed using several different tests rather than with a single measure. Standardized batteries have been designed to probe neglect symptoms across a range of behaviors and tasks and are routinely used for diagnosis (e.g., BIT23 or GEREN;24 Fig. 1), often providing a global score of neglect severity based on the sum of deficits across several tests. However, it is common that the degree of neglect differs between tests within the same patient,

Spatial neglect and the parietal lobe

even though the overall severity of deficits tends to correlate between different tests across patients, perhaps simply due to lesion extent.25 Moreover, dissociations between different types of neglect symptoms are frequent and have often been taken to make important theoretical inferences on the architecture of particular cognitive systems (e.g., related to spatial representations or attention). These clinical observations have led to a number of dichotomies in the characterization of different manifestations of spatial neglect. For example, dissociations were observed between perceived versus imagined space, perceptual versus motor space, egocentric (trunk-, head-, or eye-centered) versus allocentric space (scene- or object-based), personal versus peri- and extra-personal space (including near vs. far), and global versus local space.18 This diversity suggests that different symptoms or tests might reflect at least partly distinct cognitive processes, which can be affected to varying degrees in different patients. Yet the nature of these processes and their neural substrates still remain far from being elucidated. It is also possible, in principle, that a single core neglect deficit might produce different effects on different spatial domains, due to particularities or additional dysfunctions in some of these domains. In keeping with the diversity of behavioral manifestations, neuropsychological studies have also highlighted a diversity of brain lesions causing neglect symptoms, particularly of the inferior parietal lobe,26,27 but also lateral prefrontal areas,28,29 posterior thalamus (pulvinar),30,31 basal ganglia (putamen and caudate),28,31 and even the superior lateral32 and medial temporal lobe.33,34 Other studies have emphasized subcortical damage in paraventricular white matter.35,36 This anatomical variability has long led to the view that neglect may reflect damage to a distributed network implicated in attention and space awareness, within which different nodes may play partly distinct roles.14 For example, on the basis of insightful analysis of anatomical connectivity as well as clinical dissociations,37 Mesulam proposed a predominant role for parietal areas in perceptual components of spatial attention and a role for prefrontal areas in motor exploratory components.14 This heterogeneity in both the clinical and anatomical presentation of neglect makes it hard to map spatial awareness onto specific neural

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

53

Spatial neglect and the parietal lobe

Vuilleumier

networks in the brain, and is likely to cause some of the discrepancies between studies attempting to determine the neural substrates of spatial neglect in unselected groups of patients (e.g., see Refs. 27, 32, 34, and 36). Such discrepancies might be aggravated by several factors. First, different investigators use different tests to establish the presence of neglect, including, for example, line bisection27 or not,32 or excluding hemianopia32 or not.27 Further, different studies use different kinds of line bisection or cancellation tasks, and/or apply different cutoff scores for diagnosing neglect, potentially selecting patients with different patterns of brain damage. Second, in most studies, neglect diagnosis is based on a composite score, typically retaining patients with deficits on at least two tests among a battery of several tests. This selection procedure could lead to pooling patients with different deficits and different patterns of brain lesion in the same group (e.g., some patients being impaired on tests A and B, others on tests C and D), hence unreliable anatomical overlap. Third, traditional lesion mapping methods based on regional overlap or lesion frequency may not be appropriate to delineate the substrates of functions subtended by multiple equipotent sites or relying on long-range connections within distributed networks.38 For instance, similar motor paralysis can occur after damage at various points along the cortico-spinal tract. Likewise, if similar neglect symptoms arise after anterior or posterior brain lesions, the most frequent overlap could potentially be found in an intermediate but irrelevant location damaged in both populations, simply due to the variable extent of vascular territories affected by stroke. The latter problem may be avoided by more recent lesion mapping approaches based on a correlation of behavioral deficits with a voxel-by-voxel analysis of brain tissue damage (as used for image contrast change in functional neuroimaging), which allow one to examine lesion–symptom relationships at a more graded and distributed whole-brain level.34,39,40 Thus, past research on the neural mechanisms of neglect is somewhat characterized by a striking paradox. On the one hand, a rich collection of behavioral dissociations has been carefully documented and studied; while on the other hand, a quest for a critical neuroanatomical substrate with unique impact on spatial awareness has relentlessly been pursued and debated. However, it is now increasingly recognized 54

that these divergences might be partly overcome by an approach focusing on the neural mechanisms of more specific components of neglect, characterized in terms of more basic cognitive processes rather than broad clinical categories only. Further, it is also increasingly considered that the neural substrates for such components should be characterized at the level of functional networks of interconnected regions, rather than in relation to particular regions of interest. Nevertheless, besides a well-established relation to attentional processes,15,17 the exact cognitive constituents underlying various neglect symptoms still remain poorly specified, hence limiting the efficacy of anatomical lesion mapping endeavors. Attention and functional segregation in superior parietal cortex One of the best known cognitive functions associated with neglect and parietal lobe function is spatial attention, for which abundant research exists at both the behavioral and neurobiological levels. However, attention itself is not a unitary process. For instance, a classic dichotomy has been established between exogenous versus endogenous mechanisms of attention orienting in space.41 Exogenous orienting occurs when attention is drawn to one location or stimulus due to its sudden appearance or intrinsic saliency, through automatic/reflexive mechanisms. Endogenous orienting corresponds to more voluntary/controlled processes that orient attention on the basis of task goals or instructions. Behavioral neuropsychology studies have suggested that exogenous orienting of spatial attention toward the contralesional side might be most consistently impaired in patients with spatial neglect, whereas endogenous orienting might be relatively spared and does not correlate with main neglect symptoms.42,43 Furthermore, both behavioral and electrophysiological evidence suggests that exogenous (more than endogenous) attention is particularly critical to modulate the access of sensory information to conscious awareness.44 This distinction in attention has therefore also been applied to dissect neglect anatomy. In line with classic lesion studies showing a predominant impact of inferior parietal damage on contralesional exogenous orienting,42 recent work with functional magnetic resonance imaging (fMRI) points to a key role of the temporo-parietal junction (TPJ) in the detection of unexpected but relevant events,

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

Spatial neglect and the parietal lobe

Figure 2. Fronto-parietal networks controlling spatial attention. (A) Parietal and frontal areas involved in spatial attention are linked by three distinct white-matter tracts in the superior longitudinal fasciculus (SLF). (B) Their projection sites in both the parietal and frontal cortex correspond to the division into a dorsal attention network (DAN) typically associated with endogenous and spatial components of orienting, and a ventral attention network (VAN) associated with more reflexive, stimulus-driven, and nonspatial components. The SLF II might contribute to integrate activity between the VAN and the DAN by connecting inferior parietal with superior frontal areas. (C) Depiction of SLF in relation to another major white tract connecting posterior with anterior brain regions: inferior longitudinal fasciculus (ILF) and inferior fronto-occipital fasciculus (IFOF; unpublished data). A and B adapted, with permission, from Ref. 159.

including orienting and disengaging from current focus of attention in exogenous attention conditions. By contrast, more superior regions in the intraparietal sulcus (IPS) have been linked with voluntary shifting and/or maintaining attention to relevant locations or objects.45 This anatomical distinction has therefore led to an influential model of spatial attention (Fig. 2) comprising both a dorsal fronto-parietal network linking the IPS with superior frontal cortex (frontal eye fields, FEF), and a ventral network linking the TPJ with the inferior frontal gyrus (operculum/insula). According to this model, damage to the ventral system would not only

impair contralesional exogenous orienting but also induce subsequent dysfunction in the dorsal system and hence additional endogenous orienting biases toward the ipsilesional side, leading to full-blown spatial neglect.15 Moreover, damage to the TPJ reliably correlates with the presence of extinction on bilateral visual stimulation,46,47 consistent with the notion that this phenomenon may reflect exogenous capture by ipsilesional distractors and impaired reorienting to the contralesional side.42 However, recent work suggests that this functional–anatomical dichotomy in attention between dorsal and ventral attention areas

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

55

Spatial neglect and the parietal lobe

Vuilleumier

Figure 3. Multiple areas within the human parietal cortex. Both the superior parietal lobe (SPL) and inferior parietal lobe (IPL) or temporo-parietal junction (TPJ) can be divided in different cortical regions on the basis of their cytoarchitectonic characteristics and functional response profile during attention or other cognitive tasks. In particular, the intraparietal sulcus (IPS) contain distinct subareas, showing either increased activation during bilateral relative to unilateral visual stimulation when attention is directed to the contralateral side (middle segment), or decreased activation during bilateral relative to unilateral visual stimulation when attention is directed to another stimulus at fixation (posterior segment near the transverse occipital sulcus, TOS). In the TPJ, the supramarginal gyrus (SMG) is activated during spatial reorienting and other task-resetting conditions but unaffected by completion between the two hemifields, whereas activity in the angular gyrus (AG) is not only modulated by attentional reorienting and spatial working memory tasks, but also decreased by bilateral relative to unilateral contralateral stimulation. Other areas in the TPJ include the parietal operculum (POp) and the posterior superior temporal gyrus (STG), which are involved in vestibular information processing.

(Fig. 2) may not be as simple as it was initially thought. Dysfunction in the IPS induced by transcranial magnetic stimulation (TMS) may also produce deficits in exogenous attention,48 and lesions in the IPS correlate with extinction-like deficits for contralesional visual stimuli when these are presented with a simultaneous ipsilesional distractor (particularly in symmetrical position).49,50 The latter extinction-like deficits were found in a visual discrimination task, rather than detection, while attention was endogenously directed to the contralesional side in the presence of a competing stimulus on the ipsilesional side. This may not necessarily reflect the same loss in awareness as observed for clinical extinction, where attention must be directed to two targets simultaneously.11,51 Moreover, whereas patients with contralesional extinction and neglect most often have right-hemisphere lesions,52 competition costs in the discrimination of contralesional stimuli during bilateral stimulation can occur after focal lesion restricted to the IPS in either hemi-

56

sphere without other signs of neglect.50 In keeping with these neuropsychological findings, fMRI in healthy participants show selective activation in the IPS (particularly in its middle horizontal segment; see Fig. 3) when attending to a contralateral visual stimulus in the presence of ipsilateral distractors, relative to the same stimulus presented alone.49 No such increase in the IPS is seen for bilateral versus unilateral visual stimulation when these are irrelevant and attention is maintained at fixation.53 The middle IPS might therefore play an important role in controlling selective attention to contralateral locations in the presence of competing inputs, perhaps by contributing to the computation of a saliency map,54 which may integrate both top–down goaldriven mechanisms and bottom–up sensory signals. Damage to this region may thus be an important component of attentional deficit in neglect and contribute to perceptual extinction in some patients. On the other hand, a more posterior region of the IPS (near the transverse parieto-occipital

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

Spatial neglect and the parietal lobe

A

B

C

Figure 4. Dissociable effects of bilateral visual competition in superior and inferior parietal cortex in healthy volunteers. (A) Visual stimulation paradigm with either unilateral right, unilateral left, bilateral, or no stimulation in the peripheral visual field, while attention is maintained on central targets at fixation to perform either a demanding (high load) or easy (low load) discrimination task. (B) Bilateral areas in the posterior IPS show decreased response to bilateral versus unilateral peripheral stimulation, with stronger suppression during higher attentional demands at fixation. Symmetrical effects occur in the right (plotted in bar graphs) and left posterior IPS. (C) Bilateral areas in the IPL (overlapping with angular gyrus) also show decreased response to bilateral versus unilateral peripheral stimulation but irrespective of attentional demands at fixation. In addition, only the right IPL responds to unilateral ipsilateral stimulation (right panel), whereas the left IPL responds to contralateral stimulation only (right panel). Adapted, with permission, from Ref. 55.

sulcus, TPOS; see Fig. 3) has been found to exhibit a suppressive effect (rather than enhancement) due to competition between the two visual hemifields when peripheral stimuli are task irrelevant.55 In healthy subjects, this more posterior region shows reduced fMRI activation to bilateral relative to unilateral peripheral stimulation, but this suppressive effect arises only when attentional demands at fixation are high (difficult detection task), presumably reflecting a reduction in the representation of surrounding visual space when attention is focused (Fig. 4A and B). This competition effect indicates suppressive interactions between bilateral sensory

inputs, rather than active filtering or disengaging from distractors, unlike the competition enhancement seen in the middle IPS when attended stimuli are presented with distractors. The posterior IPS might thus mediate a capacity-limited representation of salient visual locations based on bottom– up inputs, rather than behavioral goals. Note that, by contrast, sensory competition across the two visual hemifields does not seem to modulate stimulusdriven responses of earlier retinotopic visual areas in striate and extrastriate occipital cortex.55 Such competition in low-level visual areas may, however, occur when peripheral stimuli are attended,56

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

57

Spatial neglect and the parietal lobe

Vuilleumier

suggesting that the latter effects result from changes in top–down attentional factors rather than in a purely sensory-driven manner. Taken together, these findings converge with other results indicating that several subregions exist in and around the IPS, each with a distinct functional profile of activity during attentional tasks.53,57 Moreover, in addition to the above, a more superior and medial parietal region appears critically involved in attention shifts across different domains, including spatial shifts between locations, stimuli, or tasks,58 while it is not influenced by sensory competition between concurrent stimuli. Other more anterior regions in and around the IPS are also involved in coding spatial locations for eye and limb movements, integrating not only visual inputs but also other sensory modalities.57 Accordingly, in nonhuman primates, many different subareas have been identified in posterior parietal cortex along the IPS, implicated in attention control and sensorimotor transformation across different sensory and motor domains (such as lateral (LIP) and ventral (VIP) areas of the IPS). It is likely that large brain lesions (as typically observed after stroke) will affect more than one of these parietal areas, and hence the clinical manifestations of neglect will likely result from a combination of deficits in different kinds of parietal computations. Furthermore, variations in the extent of damage will destroy or disconnect different subregions in different patients, and thus presumably lead to corresponding variations in performance across different clinical tasks. For example, one might speculate that a loss of contralesional visual information in the posterior IPS/TPOS, in the presence of intact inputs from the ipsilesional field, may contribute to the apparent compression of space or line length typically observed in neglect patients,59 without disrupting the ability to voluntarily direct attention to the contralesional extremity of such lines (and even extend them toward the neglected side60 ) when more anterior or inferior regions in parietal cortex are preserved. In contrast, a combination of damage extending from the posterior to middle or anterior IPS might cause additional deficits in coding for contralesional locations (and hence orient attention there) when attention is captured by concurrent information on the ipsilesional side (e.g., in multi-item scenes). Conversely, sparing of the IPS may account for intact endogenous orienting of attention in neglect patients43,61 when their lesion 58

involves other areas. Increased knowledge on the functional parceling of superior parietal cortex will certainly enhance our understanding of deficits in attention, spatial competition, and sensory–motor transformations that potentially contribute to spatial neglect. Functional segregation in the inferior parietal cortex Like damage to the IPS, lesions in the inferior parietal lobe (IPL) can also produce behavioral deficits in processing a contralesional target that is simultaneously presented with ipsilesional distractors,42,50 as exemplified by the common presence of perceptual extinction in these patients.46,62 However, this impairment may have a different cause than the competition effects caused by IPS lesions. Indeed, fMRI data in healthy participants suggest that activity in the TPJ for contralateral visual stimuli is not modified by the presence of contralateral stimuli per se.53,63 Rather, the TPJ activates whenever attention is (exogenously or endogenously) drawn to one side and must be reoriented toward task-relevant information in the opposite side (invalid vs. valid attention),15 even in the absence of a competing stimulus.49,64 Interestingly, these reorienting effects appear more strongly associated with the right than left TPJ,65 in contrast to attentional effects observed in superior parietal areas that are most often symmetrical in both hemispheres. However, here again, the TPJ is likely to be fractionated into several distinct functional subareas (Fig. 3). More anterior regions in the right supramarginal gyrus are selectively recruited by reorienting attention to unattended or unexpected targets, with no effects of sensory competition between multiple stimuli.53 The same regions are also activated in other nonspatial conditions involving breaches in expectations and adaptive changes in current task contingencies.66 It has therefore been suggested that these regions may function as a circuit breaker that interrupts and resets ongoing activity in brain networks to respond to novel and unexpected information. In contrast, a more posterior area in the right angular gyrus (Fig. 3) shows a combination of effects due to attention reorienting and stimulus competition,53 but with responses significantly reduced by the presence of bilateral relative to unilateral stimuli when these are unattended and presented with a central target.55 These responses

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

in the IPL are also asymmetrical between the two visual hemifields: the left parietal cortex activates to unilateral right distractors, whereas the right IPL activates to either left or right unilateral distractors, more than to bilateral stimuli in both hemispheres (Fig. 4A and C). This asymmetry in contralateral responses accords with the longstanding view that the right parietal lobe may represent both sides of space, while the left represents only the contralateral right side.14,26,67 In such patients, the intact left IPL may still allow (or favor) responses to right stimuli; whereas in left parietal patients, intact regions in the right hemisphere can monitor stimuli from either left or right hemifield. Nonetheless, the mechanisms for a suppression of IPL response to bilateral peripheral distractors (compared to unilateral distractors) remain unclear. Other findings also indicate greater responses in right IPL to a single target presented alone as compared to multiple targets or a single target presented with distractors.63 Speculatively, the pattern of effects in the angular gyrus might reflect a complex role of these cortical areas for the individuation of behaviorally relevant or salient objects, perhaps contributing to the formation of token representations that bind together different spatiotemporal attributes into a unique and distinctive event. No such individuation may occur when bilateral and symmetric stimuli are presented simultaneously at unattended locations, as they may appear as part of the background rather than outstanding elements in a scene. Token representations are conceptualized as a temporary episodic registration of the current instantiation of a particular object or event, in terms of what was where and when, allowing the maintenance of perceptual unity and continuity of objects when they move or change in the scene, but also linking different images of the same object across body or eye movements.68,69 Capacity limitation in token representation is thought to explain perception failures in situations where similar sensory information must be bound to different items in either space or time, such as repetition blindness.70 Thus, when two instances of a particular object are seen in rapid succession or close proximity, the brain often forms only a single token and people perceive only one of the two objects. In vision, the formation of token representation might require the integration of inputs from both the dorsal (occipito-parietal) and ventral (occipito-temporal)

Spatial neglect and the parietal lobe

streams, processing information about object location and identity, respectively, which would be consistent with neuropsychological accounts proposing a key role for the IPL in binding both streams.71 A deficit in token formation and event individuation might contribute to the phenomenon of contralesional perceptual extinction typically seen in patients with IPL damage, possibly due to disruption in such dorsal–ventral integration.51,72 This phenomenon might be exacerbated when combined with an extension of the TPJ lesions to supramarginal cortex, causing additional impairment in reorienting or resetting mechanisms after capture of attention by ipsilesional distractors (i.e., impaired attentional disengagement).15,42,43 Deficient token formation may explain the exacerbation of perceptual extinction or neglect symptoms when bilateral stimuli are perceptually similar11,51 and appear simultaneously in time at symmetrical locations.72,73 These effects can extend beyond vision to other sensory modalities, as suggested by auditory and tactile extinction,74 consistent with a supramodal or crossmodal function of the angular gyrus.75,76 Even without a concomitant impairment in reorienting, deficient tokenization processes after IPL damage might also account for nonspatial disturbances commonly observed in these patients, with or without clinical neglect. For instance, neglect patients show abnormal temporal dynamics of attention, with difficulties in distinguishing between two targets appearing in rapid succession at a central location,77 judging the relative onset and offset of visual events,78,79 and integrating discontinuous stimuli in a coherent percept of apparent motion.80 Interestingly, the latter deficits are not spatially lateralized but observed in both hemifields, even though they might exacerbate contralesional spatial biases when combined with other lateralized (e.g., reorienting) impairments. Interestingly, deficits in temporal discrimination abilities for two concurrent targets were found to correlate with damage in right angular gyrus, unlike spatial biases in the same temporal order judgment tasks that correlate with more anterior damage in the TPJ (supramarginal gyrus), in addition to other differences in white matter and cerebellum.79 Similar or nearby regions in right angular gyrus might subserve other cognitive functions crucial for attention and spatial cognition, including working memory maintenance and updating. Accordingly,

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

59

Spatial neglect and the parietal lobe

Vuilleumier

nonlateralized deficits in spatial working memory are commonly observed in neglect patients and may contribute to their behavioral deficits in various tasks, particularly search (cancellation) tasks in which patients often tend to recursively explore the same location and revisit previously marked targets.81 Such deficits in spatial short-term memory after IPL lesion have been observed in both sides of space81,82 but may predominate on the contralesional side.83 Thus, in a visual search task, neglect patients showed no difference in target detection latencies when targets changed locations within the left hemifield during search, relative to when they remained at the same locations, while location changes within the right hemifield disrupted performance, suggesting a lack of stable representation for explored location in contralesional but not ipsilesional space.83 Such impairment in spatial memory might partly be secondary to impaired spatial remapping processes that allow maintaining and updating the representation of a stimulus location when its perceived position (e.g., on retina) is modified by changes in body (e.g., eye) position.84,85 For example, a visual location (even in the intact right hemifield) may vanish from memory when neglect patients make a single saccade to a more rightward location, whereas locations in either hemifield can be remembered across similar time delays without intervening movement.84 This pattern has been interpreted as an inability of spatial memory to maintain or transfer contralesional visual information in gaze-centered coordinates. Moreover, spatial memory deficits correlate with neglect severity and lesion extension in IPL and subcortical white matter.81,82,84 Nonlateralized deficits in sustained attention and alertness have also been linked with IPL lesions and neglect.86,87 Finally, other areas in the TPJ may correspond to cortical projections sites for the vestibular system, integrating information critical for the coding of body position in space, with a well-established righthemisphere dominance.88 Recent evidence from brain imaging in healthy participants suggests that several subareas in the TPJ may process vestibular inputs, including the parietal operculum in particular (Fig. 3), but also the posterior insula and superior temporal gyrus.89,90 Damage to these regions might therefore also contribute to some of the wellknown anomalies in spatial representations in neglect patients, including deviation in the subjective 60

perception of straight-ahead orientation and trunk midline position.91,92 Even if it remains nonexhaustive, this overview indicates that not only the superior parietal cortex around the IPS, but also the inferior parietal lobe and the TPJ contain a number of distinct subregions that make different contributions to attention and spatial cognition. While conjoint damage to this collection of cortical regions and their connections is likely to account for many typical aspects of neglect, variations in the exact impact of lesion across patients may help explain some dissociations in clinical presentation, although the precise understanding and mapping of these different functions within parietal cortex still remain far from resolved. Furthermore, parietal areas do not function in isolation but are connected to several prefrontal and subcortical areas to form distributed networks whose coordinated functioning can also be impaired by focal damage in parietal cortex, as further discussed in subsequent sections. Behavioral dissociations and neuroanatomy of neglect components Even though deficits in spatial attention and damage to its neural substrates in parietal cortex constitute a major component of the neglect syndrome, it is clear that purely attentional disturbances (even those affecting exogenous mechanisms43 ) are not sufficient to account for all neglect phenomena. For example, neglect of contralesional space during manual or ocular exploration in complete darkness9 cannot be explained by a capture of exogenous attention by ipsilesional stimuli (at least in a straightforward manner). Likewise, dissociations affecting a selective domain (such as far vs. near space, front vs. back space, extrapersonal vs. personal space, and locations within vs. between objects) all imply additional brain systems’ supporting stimulus awareness in a more elaborate format than just based on asymmetry in the control of exogenous attention orienting. These behavioral observations have been taken to indicate the role of higher order representations of space, which might integrate sensory (multimodal) inputs with motor planning for different effectors (e.g., eyes, limbs) or different action sectors, such as reaching or avoiding (near space), and looking or walking (far space).20,21,57,93 Several studies have investigated the neuroanatomical substrates underlying some of these

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

differences in the spatial distribution of neglect deficits. A frequent approach is to compare lesions in patients with and without deficits in particular clinical tests. For example, neglect on line bisection has been reported to be more common after posterior (parietal) damage, whereas neglect on cancellation tasks is more severe after anterior (frontal) or subcortical lesions.94,95 Likewise, omissions of targets in left space during cancellation tasks (space-based or egocentric neglect) have been linked with damage (or hypoperfusion) in inferior parietal lobe, whereas omissions of the leftsided features of targets irrespective of their location in space during cancellation (stimulus-based or allocentric neglect) were associated with dysfunction in lateral and inferior temporal lobe.96,97 The latter findings were interpreted in terms of distinct neural systems for controlling attention between and within objects in the dorsal-parietal and ventral-temporal visual stream, respectively. However, other lesion studies used different clinical tests to define neglect in egocentric (viewer-centered) or allocentric (either stimulus- or scene-centered) coordinates and found different neuroanatomical correlates,98,99 or instead implicated partly overlapping brain regions.62,99,100 In keeping with this, functional neuroimaging work in healthy volunteers also suggests a differential recruitment of fronto-parietal versus occipito-temporal regions in egocentric versus allocentric spatial tasks, as well as shared processing in posterior parietal cortex.101–103 A similar distinction between a dorsal action-related stream and a ventral perception-related stream of visual processing has been made to account for dissociable forms of neglect in near and far space,20,103,104 but no systematic lesion study has been performed in a large patient group yet. Taken together, these anatomical data converge with the idea that different representations of visual space might be distinguished in terms of their use for action or perception, even though the exact computations and neural systems supporting these representations, and their interaction with attention and awareness, still need to be more precisely determined. Moreover, egocentric neglect deficits can be further subdivided with respect to different reference frames centered on the eye, head, trunk, or even limb,105 suggesting distinct contributions of representations controlling and/or monitoring the position or movement of specific body parts.

Spatial neglect and the parietal lobe

Other dissociations between extrapersonal (space-centered) and personal (body-centered) neglect have been investigated using voxelwise lesion mapping.106,107 In one study,106 each spatial domain was assessed using a battery of different tasks (including cancellation and reading, but not line bisection, for extrapersonal space), and impairment was diagnosed when patients performed below a given cut-off score on at least two tasks in the same domain. Personal neglect was found to correlate with lesions in the postcentral somatosensory cortex, adjacent white matter, and inferior parietal lobule. This pattern differed from extrapersonal neglect whose severity correlated with lesions mainly affecting the inferior frontal cortex and superior temporal regions, partly overlapping with areas previously associated with exogenous attention and vestibular processing. Motivated by the heterogeneity of anatomical findings and symptoms examined across past studies, a few meta-analyses were recently performed using an activation likelihood estimation (ALE) procedure originally developed for functional neuroimaging studies. By combining results from more than 20 lesion mapping studies published in the past decade, these meta-analyses have generally confirmed the notion that different manifestations of spatial neglect can be linked to both distinct and common neural substrates.47,108 The most consistent findings were found for extinction (in the TPJ), line bisection (in posterior parietal cortex), and to a lesser degree, cancellation tasks (in prefrontal more often than parietal cortex), whereas other neglect symptoms (such as allocentric or personal neglect) showed greater dispersion across brain sites (Fig. 5). In addition, all studies converged to show a critical involvement of hemispheric whitematter structures, whose damage is not only frequent but often shared between different neglect deficits, particularly for the superior longitudinal fasciculus (SLF), inferior fronto-occipital fasciculus (IFOF), and thalamic radiations (see also below and Fig. 2C). Such meta-analyses are extremely useful and promising, as they provide a data-driven statistical tool to assess consistency across studies. Yet, they remain limited due to the fact that the clinical tests and diagnostic criteria may vary substantially between studies, even when considering a specific task (e.g., cancellation) or a specific spatial domain (e.g., allocentric). While this variability might help

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

61

Spatial neglect and the parietal lobe

Vuilleumier

Figure 5. Meta-analysis of 20 lesion-mapping studies of neglect symptoms. The likelihood of lesion peaks is illustrated on an inflated brain template for six different neglect measures, including (A) line bisection (purple spheres); (B) cancellation tasks (red spheres); (C) global scores from a combination of tasks (green spheres); (D) allocentric object-centered tests (blue spheres); (E) personal body-centered tests (black spheres); and (F) perceptual extinction on double stimulation (orange spheres). Adapted, with permission, from Ref. 47.

to highlight common processes implicated in a given neglect component, differences in sensitivity or additional demand characteristics across clinical tests might blur the role of more specific processes. Differences in the delay since lesion onset and the interval between radiological and behavioral assessment may further complicate the comparison between studies.40,109 A more general limitation of many studies attempting to delineate lesions associated with different types of neglect symptoms is that these symptoms are defined according to broad and clinically defined dichotomies (e.g., personal vs. extrapersonal or egocentric vs. allocentric space), rather than more specific cognitive processes. However, it is plausible (and even likely) that some of these 62

spatial abilities rely on more than one unitary cognitive function. For example, allocentric judgment tasks may involve object shape recognition, symmetry detection, and disengagement from competing information, each underpinned by distinct neural substrates (just like perceptual extinction may reflect both spatial and temporal deficits in attention subserved by different areas in parietal cortex). Furthermore, neglect symptoms are most often measured with a single test, adapted from clinical practice. However, these clinical tests have been developed to be sensitive to the presence of deficits, rather than specific symptoms, and hence tend to tap into different abilities. One approach to identifying specific functional components (or building blocks) contributing to

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

Spatial neglect and the parietal lobe

Figure 6. Lesion mapping of neglect factors. Factorial analysis of performance across various tests identified three separate functional components that accounted for neglect in different domains and were associated with different lesion sites. Factor 1 correlated with the lesion in posterior parietal cortex (and to a lesser degree in superior frontal cortex) and predicted neglect in line bisection, text reading, and drawing. Factor 2 correlated with the lesion in inferior frontal cortex (and to a lesser degree in parietal cortex) and predicted neglect in cancellation tasks and drawing. Factor 3 correlated with the lesion in the temporal lobe (with a peak in the deep medial region and extension toward the inferior later cortex) and predicted neglect in single-word reading and object-based tasks (allocentric neglect). Adapted, with permission, from Ref. 34.

one or many neglect symptoms has been to apply factorial analyses or clustering statistics on the performance of patients across a large range of different tasks.24,34,110,111 This approach exploits concomitant variations among a collection of correlated variables to reduce these variables to a lower number of potentially underlying, but unobserved, factors. For example, it is possible that variations of neglect severity across a battery of clinical tests might reflect deficits in, say, two distinct cognitive factors (e.g., exogenous and endogenous attention), each associated with a particular neural system (e.g., ventral and dorsal parieto-frontal networks). Using a range of standard tests in large groups of braindamaged patients (n = 40–200), factorial analysis studies have indeed pointed to the role of at least two or three separate factors explaining neglect behavior across different kinds of tests. These factors appear remarkably consistent across studies,24,34,110,111 and cluster along the following dimensions: (1) exploratory functions involving overt motor behavior and distractor interference (search component), particularly associated with performance on cancellation tasks, drawing, and writing; (2) perceptual functions recruited for encoding, maintaining, and/or shifting across multiple locations in space (deploy component), and associated with performance on line bisection, text reading, and figureground segregation; (3) allocentric representations (object-based component) associated with performance on single-word reading and omissions of contralesional features within objects. Of course, neglect is likely to involve more than these three fac-

tors, which altogether explained only between 50% and 85% of the variance in tests across the different studies. Hence, other factors must explain the residual variance. Nevertheless, these data demonstrate that different neglect deficits may implicate at least partly distinct cognitive processes, while the same cognitive process may contribute to different deficits in different tasks. For example, in one study,34 neglect on drawing from copy was found to be predicted by a combination of deficits in two different components (search and deploy). Furthermore, this data-driven approach may also allow delineating the neuroanatomical substrates associated with higher level factors explaining behavior across different tasks (with or without clinically overt neglect), rather than mapping the correlates of neglect in a single test that may be underpinned by multiple cognitive processes. For example, in the study mentioned above,34 the exploratory search factor was predominantly associated with damage to the inferior and posterior middle prefrontal gyri; whereas the perceptual deploy factor was associated with damage to both inferior and superior parietal lobe, and to a lesser degree, to posterior frontal areas (Fig. 6). These data accord with recent meta-analyses suggesting differential contribution of frontal and parietal lesions to cancellation and line bisection tasks, respectively (see Refs. 47 and 108), but also show that similar cognitive processes may be implicated in other behavioral manifestations. Frontal lobe damage may contribute to greater distractor interference29 and perseveration112 during cancellation tasks. In

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

63

Spatial neglect and the parietal lobe

Vuilleumier

addition, the same study34 revealed that a third object-based factor correlated with damage to both medial (parahippocampal) and lateral (inferior sulcus) temporal lobe areas, again converging with other findings from studies on allocentric neglect.96–98 Interestingly, Verdon et al.34 found that patients with more severe and pervasive neglect, involving more than one of the three factors identified above, had more common extension of their lesion to subcortical regions in paraventricular white matter,34 suggesting that disconnection of white-matter tracts might influence multiple domains of spatial cognition simultaneously (see also below). Additional research is needed to refine these factors and understand the exact cognitive processes underlying them. Even though the heterogeneity of neglect symptoms might be compatible with the existence of a core syndrome, with optional satellite deficits in other domains,113 the nature of this core disorder still remain unclear beyond purely descriptive clinical features. Moreover, this core syndrome itself might encompass a combination of more elementary deficits that do not lead to clinical neglect when damaged in isolation. To better dissect and define the specific cognitive components contributing to neglect behavior, and to understand their role in different deficits, future research should use a large variety of tests tapping onto distinct domains and large group of patients, both with and without a clinical diagnosis of neglect. Distributed networks and connections In keeping with the view that spatial awareness and its disturbances in neglect result from a combination of interactive processes, recent research has highlighted the role of distributed brain networks in the control of attention,14,15 as well as the widespread impact of damage to subcortical connections in neglect patients.34,36,114,115 It has long been known that damage to deep white-matter regions may lead to long-lasting and severe forms of neglect, but it is only during the past few years that advances in neuroimaging techniques (such as diffusion tensor imaging, DTI) have fostered more systematic investigations concerning the precise anatomical connectivity of brain networks implicated in neglect. Converging evidence from several lesion mapping studies34,114,116,117 and meta-analyses47,108 points to a greater frequency of lesions along intrahemi64

spheric tracts in right brain-damaged patients with neglect relative to those without, in particular in the SLF (Fig. 2A) and inferior fronto-occipital fasciculus (IFOF; Fig. 2C), reciprocally connecting posterior and anterior brain regions. In addition, the thalamic radiations projecting to fronto-parietal cortex are also frequently involved. These white-matter lesions appear to be shared across a range of different deficits (e.g., egocentric or allocentric neglect) and are more frequent when patients show neglect across many different tests,34,99,108 suggesting that subcortical disconnections may influence several cognitive subcomponents simultaneously and thus produce a common pattern of deficits across different spatial domains. Moreover, precise white-matter tract-based statistics in patient groups, as well as single-case dissection analysis, suggest that specific portions of the SLF might differentially be implicated.36,117 SLF has been proposed to be segregated into three distinct bundles: SLF1 connects superior parietal cortex with more superior frontal areas (overlapping with the dorsal endogenous attention network), whereas SLF3 connects the inferior parietal cortex with more inferior frontal areas (overlapping with the ventral exogenous attention network), and SLF2 connects inferior parietal with superior frontal regions (possibly implicated in cross-talks between networks; see Fig. 2). Damage to SLF2 (or nearby superior occipito-frontal fibers, SOF) appears as the best predictor among white-matter tracts for the presence and/or severity of spatial neglect, regardless of test.116,117 Nevertheless, voxelwise regression analysis suggests that line bisection deficits tend to correlate more with lesions in superior paraventricular and deep parietal lobe white matter, whereas visual cancellation deficits correlate with more anterior lesions under the precentral and middle frontal gyri, near the FEF,117 in agreement with previous lesion data suggesting distinct roles for frontal and parietal cortical areas in neglect symptoms.34,94,95 In addition, transient neglect on line bisection can also be induced by electrical stimulation of whitematter fibers in the depth of the inferior parietal lobe, corresponding to SLF2 or SOF.118 These findings are consistent with diffusion tensor imaging work in healthy volunteers supporting a crucial involvement of these fronto-parietal tracts in normal visuospatial attention. Indeed, some portions of SLF show strong hemispheric asymmetry, with a right

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

preponderance that is the most pronounced for SLF3.119 In addition, the degree of asymmetry in SLF2 volume between right and left hemispheres in individual participants correlates with asymmetries in their performance on visual attention tasks, including the magnitude of right-side “pseudo neglect” on line bisection and a left hemifield advantage in the Posner cueing paradigm. Taken together, these data suggest that disrupted communication between frontal and parietal areas may play a major role in the occurrence of neglect, perhaps due to abnormal integration of dorsal and ventral attention systems within and between hemispheres.15,115 Even though it remains debated whether the white-matter lesions alone can lead to persistent neglect after stroke and whether cortical damage is more critical,116 the extension of lesions to specific white-matter tracts clearly appear to contribute to the severity of neglect and to the range of clinical manifestations, perhaps by impacting multiple functional systems simultaneously. It is also possible that particular symptoms may involve distinct white-matter tracts. For example, a single-case study suggested that additional damage to interhemispheric fibers in the posterior corpus callosum (splenium) might be responsible for representational neglect in mental imagery.120 Interestingly, abnormal white matter in posterior corpus callosum has been found to predict the severity of neglect on clinical tests (BIT) in stroke patients,121 as well as the degree of spatial biases in temporal order judgments.79 This differs from other studies reporting no association between callosal lesion and neglect,116,117 but the latter studies used different tests to diagnose neglect (mainly cancellation and line bisection). Of note, the total BIT score shown to correlate with anomalies in posterior corpus callosum included representational drawing tests as well as behavioral measures from everyday life,121 but unfortunately this study did not examine correlations for specific subtests. A role for callosal damage would be consistent with one of the rare lesion models of neglect in monkey,122 in which neglect was observed when combining optic tract section with commissurotomy, or after deep leucotomy disconnecting the parietal cortex from both ipsilateral and contralateral visual inputs, but not after other lesions including parietal cortical ablation alone or combined with frontal eye–field ablation. Likewise,

Spatial neglect and the parietal lobe

simultaneous interruption of both inter- and intrahemispheric communication might result from deep brain lesions extending to both paraventricular and paracallosal fibers. This may interrupt or weaken the access of contralesional sensory information to attentional networks in the right hemisphere, while at the same time disrupting frontoparietal interactions that control attention orienting in space and the selection of behaviorally relevant events for conscious awareness. In sum, recent connectivity approaches to neglect anatomy support the view that parietal and frontal areas dynamically interact within large-scale networks whose activity is crucial to subserve spatial attention and awareness.14,15,114 One may speculate that reciprocal fronto-parietal communication might implement winner-take-all threshold mechanisms that allow the modulation and selection of high-order spatial representations (e.g., saliency maps) and/or sensory event representations (e.g., tokens) computed in posterior parietal areas.123–125 Interruption of white-matter fibers in SLF (and possibly other tracts) might contribute to the abnormal spatial biases arising in the selection of these representations for conscious awareness.12 Furthermore, besides fronto-parietal and callosal disconnections, subcortical brain lesions may disrupt other whitematter fibers that are also likely to play an important role in space representation and attention. These include connections between parietal cortex and early visual areas (or other sensory areas),126 as well as connections with subcortical nuclei, such as the pulvinar, or with superior colliculus, both of which have direct interactions with parietal cortex.127 For example, a functional imbalance within subcortical circuits is suggested by the classic Sprague effect in cats, whereby an apparent contralesional neglect (i.e., blindness or deafness) after unilateral parietooccipital lesion can be nullified by additional destruction or deactivation of the superior colliculus on the other side128 —an effect suggesting that impaired spatial orienting may be primarily caused by losses of descending signals on structurally intact collicular circuits (see also Ref. 129). Future studies using advanced diffusion imaging in patients and animal models should help clarify the functional role of these different cortical-subcortical pathways in spatial cognition and their role in neglect disorders.

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

65

Spatial neglect and the parietal lobe

Vuilleumier

Distant functional effects of focal lesions Disconnection between areas, either due to whitematter lesions or destruction of one cortical or subcortical gray-matter structure, can produce a dysfunction in structurally intact portions of the corresponding brain network, by interrupting or distorting the flow of communication between distant regions. Such dysfunction may correspond to the phenomenon of diaschisis, a transient suppression of neuronal activity and cerebral blood flow in anatomically preserved circuits following damage in a separate but functionally related neuronal region, which is often observed in the acute stage after stroke and tends to progressively recede during subsequent recovery. Diaschisis in cortical areas has sometimes been proposed to account for neglect symptoms arising after subcortical lesions in white matter or deep gray nuclei, and can be assessed with perfusion imaging measures.130 Diaschisis in distinct cortical territories may also account for differences in the clinical manifestations of neglect (e.g., egocentric or allocentric deficits).97 Distant dysfunction following focal damage can also influence the functional dynamics of welldefined networks during particular task demands as well as during resting state conditions.115,131,132 In particular, functional MRI in stroke patients has shown that acute damage in the right TPJ is associated with reduced activity in intact regions of the ipsilateral IPS, which might in turn cause an interhemispheric imbalance with a relative increase of activity in the opposite left IPS, possibly accounting for the typical rightward spatial biases in attention orienting in neglect patients.131 Moreover, such interhemispheric asymmetry in the IPS was reported to correlate with spatial asymmetry in attention performance on the Posner orienting task, and to abate with neglect recovery in more chronic stages. In addition, disrupted functional connectivity between right inferior parietal and frontal areas was also found to predict left neglect severity, and these functional changes in connectivity were more frequent in patients with damage to SLF,115 in keeping with the view that the latter white-matter tract may play a crucial role in coordinating frontoparietal areas (see above). However, these anomalies in the functional connectivity of the TPJ persisted after recovery in the chronic stage, whereas intra- and interhemispheric functional connectivity of the IPS

66

was altered during the acute stage after stroke only, and then fully recovered in parallel to full-blown neglect symptoms.133 These data have therefore been interpreted as evidence that widespread dysfunction in the ventral fronto-parietal attention network may cause left spatial neglect by inducing secondary disturbances in the structurally intact dorsal attention network. However, increased activity in left parietal cortex does not always correlate with contralesional deficits but might also reflect compensatory processes.134 Neglect deficits in the acute stage after subcortical or inferior parietal lesions are also associated with more diffuse and bilateral hypo-activity in areas of the dorsal attention network, including the IPS and FEF.135,136 Conversely, focal damage in the IPS can produce spatial biases in selective visual attention without concomitant changes in activity for intact areas in the TPJ or frontal lobe.50 Distant functional effects of parietal lesions also arise in intact sensory regions during specific task demands. For instance, fMRI in patients with right parietal damage shows a reduced activation to left visual stimuli in intact retinotopic visual cortex of the damaged hemisphere (Fig. 7), but only when attention is actively engaged by a demanding task at fixation (detection of prespecified targets), whereas retinotopic responses are normal and symmetric under passive visual stimulation (with fixation maintained at screen center without a concurrent detection task).137 This reduction in retinotopic areas also correlates with reduced ability of patients to report left-sided stimuli during the central attention task. These data suggest that although visual inputs are preserved for both hemifields, a functional imbalance between hemispheres with reduced perceptual processing in the damaged side occurs as soon as attention resources are allocated to competing information (even when the latter is presented at a central location in space). Furthermore, such attentiondependent reduction in stimulus-evoked responses appeared to be progressively exacerbated from earlier to later retinotopic areas along the visual stream (i.e., with small effects in V1 but much larger effects in V4, where visual responses to contralateral stimuli were actually abolished during higher attention demand at fixation). Reduced activity in intact right occipital cortex is also observed during attentional tasks in right brain-damaged patients with left neglect relative to those without.134 Similar decreases

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

Spatial neglect and the parietal lobe

Figure 7. Attention-dependent changes in visual cortex activity after right posterior parietal damage. Responses to visual stimulation in the contralesional left (LVF) and ipsilesional right (RVF) hemifields were compared either (A, B) during a passive condition while fixation was maintained at screen center without any task, or (C) during an active attentional task requiring the detection of infrequent red target at fixation. In two patients (JC, upper row; AH, lower row) with damage restricted to the right posterior parietal lobe, visual responses in structurally intact right occipital cortex were preserved during the passive condition, but markedly reduced during the central attention tasks. Retinotopic mapping analysis showed that this suppression arose in all visual areas but with increasing effects from V1 to V4. No differential effect was observed in these retinotopic areas for bilateral versus unilateral visual stimulation. Adapted, with permission, from Ref. 137.

in occipital activity may even occur at baseline, without any visual stimuli in peripheral hemifields, but arising when a central fixation cross is presented during catch trials while the patient is expecting a possible target on one or the other side.138 These findings demonstrate that structurally intact visual areas fail to activate normally after parietal lesion, presumably due to the loss of top–down or reentrant modulatory signals, but the exact source of these signals remains unresolved. One possible source might arise in superior parietal regions computing saliency maps of space that can then be used to bias sensory processing in lower visual areas through direct or indirect feedback connections. A modulation of intact sensory areas in neglect patients has also been observed during manipulations of egocentric spatial coordinates while sensory stimulation was actually kept constant. Thus, contralateral occipital activation and visual detection performance for left visual stimuli were found to be reduced in a right parietal patient when his gaze was directed straight ahead or leftward, but restored when gaze was directed rightward such that stimuli in the left retinal hemifield now fell in right egocentric space (unpublished observations). Likewise,

tactile detection on the right hand and contralateral activation of left somatosensory cortex were reduced in a right parietal patient with neglect when his intact right hand was placed in the left (contralesional) egocentric space.139 These modulations of sensory cortical areas accord with modulations of visual and tactile extinction by posture reported in previous behavioral studies.105,140 The presence of visual and tactile extinction after stroke has also been found to correlate with changes in resting perfusion values in occipital and inferior parietal cortical areas, respectively.141 These findings indicate that disturbances in spatial attention and spatial representation due to focal parietal damage may have a profound impact on the functioning of structurally intact sensory areas at lower stages of perceptual processing. Furthermore, these changes have direct consequences for perceptual awareness and thus provide a proximate neural substrate for at least some of the clinical manifestations associated with neglect. These phenomena blur a strict functional distinction between primary sensory losses (e.g., hemianopia after occipital damage) and secondary attention-dependent or spacerelated suppression of sensory processing (e.g.,

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

67

Spatial neglect and the parietal lobe

Vuilleumier

Figure 8. False positive in anatomical lesion mapping. Lesion maps show a voxelwise contrast between structural brain damage in patients with hemianopia and those without hemianopia, as clinically assessed by confrontation and noted in neurological clinical records (n = 30 unselected patients with a focal right hemisphere ischemic stroke). Lesion peaks (blue to cyan) can be observed in occipital and ventral temporal areas, as expected, but also in posterior parietal areas, suggesting that damage in the latter may induce pseudo-hemianopia due to attentional disturbances rather than true visual field cut (unpublished data).

extinction after parietal damage). On the one hand, this further highlights the role of distributed and interactive networks in subserving spatial awareness. On the other hand, this also calls for caution when interpreting the results of anatomical lesion mapping studies in neglect patients, since focal lesion in a given cortical or subcortical site may cause symptoms through distant influences on another distant, but connected site. A striking illustration is provided by voxelwise lesion-symptom mapping (VLSM) analysis of visual field deficits (hemianopia) as assessed by confrontation tests during clinical examination, which may reveal lesions in the occipital lobe (as expected) but also in parietal cortex (see Fig. 8). The latter false-positive result suggests that visual dysfunction caused by parietal damage might occasionally result in a pseudohemianopia that is mistakenly diagnosed as true hemianopia.105,142,143 More precise anatomical and DTI analysis might be necessary to verify that such findings are not caused by interruption of optic radiations in the depth of the parietal lobe, although this seems unlikely given the frequent subsequent recovery of pseudo-hemianopia and its modulation by postural changes.105 Moreover, similar functional effects might explain that (apparent) sensorimotor disorders are more frequently reported after right than left hemisphere lesions.144 However, impaired awareness in extinction (and other neglect-related phenomena) cannot exclusively be accounted for by attenuated neuronal responses in sensory pathways. Several fMRI studies of extinction on double stimulation in parietal patients have compared brain responses to identical contralesional stimuli when these are consciously perceived or extinguished (relative to an absence of stimulation), and these studies consistently found preserved activation in early visual145,146 or

68

somatosensory147,148 areas despite a lack of stimulus awareness on extinction events. Critically, awareness of the same sensory events (relative to extinction) correlated with enhanced activation at higher stages along sensory pathways coupled with concomitant activation of intact parietal and frontal areas in the intact left hemisphere145–147 (see also Ref. 134). Neuroimaging studies of perceptual awareness in healthy people have reported a similar pattern of isolated activation in sensory areas during unconscious processing, but more widespread recruitment of frontoparietal areas during conscious processing.149 Thus, findings in neglect patients converge with other data to support the notion that awareness is dependent on distributed neural mechanisms in fronto-parietal networks that allow the broadcasting of stimulus information in large-scale brain networks associated with attention, working memory, and goal-directed behavior.12,14 More generally, these findings demonstrate that neglect and related disorders (such as extinction) do not reflect the destruction of a single module specialized for spatial awareness, and that their neural substrates extend beyond the site of structural brain damage. Instead, several aspects of neglect may arise from a complex interplay between defective processes due to the lesion and abnormal residual activity in anatomically preserved regions, as observed for the IPS during spatial orienting131 or early visual cortex during extinction.137 Similar functional changes leading to imbalanced activity between or within hemispheres might underlie other neglect components, such as egocentric spatial biases, deviation of subjective body midline, and common postural anomalies.113,150 However, the exact nature of the modulatory or interactive signals exerted between areas is still poorly understood.

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

Conclusions Significant progress has been made in the past decade regarding the cognitive and neuroanatomical systems involved in spatial neglect, but many questions and challenges remain. There is now growing consensus that neglect cannot be considered as a monolithic disorder with a unique neural substrate in the parietal lobe (or elsewhere in the brain). Rather, it is probably best understood as a clinical syndrome consisting of different components, which may constitute some core features, but also variable manifestations depending on the site and extent of brain damage.18,113 However, the suspected core features are still too often purely descriptive (e.g., referring to general contralesional spatial deficits in behavior), and need to be better defined in terms of underlying cognitive processes. Different researchers tend to emphasize different impairments as the characteristic signature of spatial neglect, such as deviations in egocentric coordinates, distortions in internal representations of space, or asymmetries in the control of attention, among others. Moreover, these putative core symptoms also occur in isolation without qualifying for full-blown neglect, including, for example, spatial biases in egocentric straight-ahead,151 exogenous attention orienting,42 or bilateral stimulus competition.50 This suggests that the definition of a core syndrome of neglect itself is likely to incorporate a combination of elementary deficits, implicating distinct cognitive and neural systems, whose impairment will not lead to typical neglect when present singly. In fact, this view echoes previous accounts proposing that neglect might result from the co-occurrence of two deficits (e.g., contralesional spatial biases in attention combined with concomitant local biases152 or with nonlateralized deficits in working memory153 ), although the latter components are probably insufficient to explain other neglect features. An important challenge for future research is therefore to develop a coherent component process model of spatial neglect, to identify the basic ingredients of the core syndrome, if any, and clarify whether clinically distinct entities may exist or not (e.g., extinction vs. neglect). Similar componential approaches have already provided useful theoretical frameworks in other neuropsychological domains, for instance in relation to memory processes

Spatial neglect and the parietal lobe

underlying both distinct and partly overlapping forms of amnesia. To be comprehensive, however, a componential model of neglect should go beyond dichotomous categories of symptoms (e.g., egocentric vs. allocentric, endogenous vs. exogenous, local vs. global), but identify independent cognitive processes that can account for behavior across different tasks and correspond to specific neural systems. Converging data from factorial analyses and anatomical studies suggest that these basic component processes are likely to include spatial biases in attention due to impaired resistance to exogenous interference and stimulus competition, distortions in egocentric representations, deficits in spatial remapping, as well as nonlateralized losses in working memory or temporal event resolution, among others. A functional fractionation of the neglect syndrome into component processes also dovetails with the progress made in delineating the complex anatomical parceling of the human parietal lobe and its interconnections with other brain areas. Studies in humans and monkeys demonstrate that the parietal cortex is constituted of many distinct subregions with different functions contributing to attention and space representation,53,57 such that parietal lesions in humans will inevitably affect several of these regions and hence cause a combination of different component deficits even when damage is relatively focal and restricted to the parietal lobe, as succinctly reviewed above. For instance, damage to neuronal populations in the mid-IPS might impair the ability to maintain representation of salient or task-relevant locations in contralesional space,49 while extension to more posterior IPS might abolish the representation of unattended visual locations on the contralesional side in the presence of competing inputs,55 and damage to superior IPS may additionally disrupt the ability to voluntarily shift attention to previously unattended information,45,58 such that a combination of these deficits due to a single lesion could potentially account for a defective representation of line length during bisection tasks and impaired computing of newline returns during text reading, hence a common factor shared by these different tasks in posterior parietal regions.34 Conversely, extension of damage to anterior parietal areas or sensorimotor regions may disrupt space representations in relation to different body-centered or action-based coordinates,57,154 and hence

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

69

Spatial neglect and the parietal lobe

Vuilleumier

contribute to modulation by body or limb position,105 or to deficits in spatial remapping during body or eye movements;84 whereas damage to inferior parietal and superior temporal regions may lead to anomalies in cortical vestibular processes and egocentric deviations.88,90,92 Further extension to distinct subregions in inferior parietal cortex will add to clinical manifestations, for instance, contributing to perceptual extinction through a combination of deficits in reorienting or resetting mechanisms,42,66 alertness,86,87 and spatiotemporal tokenization of sensory events,72,78 each dependent on distinct subregions in IPL.53 This functional parceling of the parietal lobe clearly argues against a single module whose damage would be responsible for parietal neglect. Furthermore, besides a mixture of deficits due to damage in parietal lobe, neglect symptoms will be further nuanced when lesions extend to other cortical areas (e.g., leading to greater distractor interference during search after frontal lesion, object-based deficits after temporal lesion) or to subcortical nuclei (e.g., pulvinar, basal ganglia) and white-matter tracts (inter- or intrahemispheric). Moreover, beyond structural damage to specific brain regions, functional changes may occur in anatomically intact regions, and these will also directly bear on perceptual and behavioral symptoms (such as reduced sensory responses of retinotopic visual cortex to contralesional stimuli when attention is engaged at fixation137 or following reduced activation of the IPS after IPL damage131 ). Altogether, these data show that a strict lesion localization approach is insufficient to fully account for complex neuropsychological deficits, such as spatial neglect, and that understanding the latter requires a comprehensive model of brain function in terms of dynamically interactive networks wherein damage in one area can cause changes in one or many other interconnected areas. Even without being exhaustive, the current review highlights that understanding the functional neuroanatomy of neglect is a tremendously complex endeavor and undoubtedly remains a major challenge for neuroscience, despite many exciting new insights on spatial cognition, attention, and parietal lobe function. Future work will need to dissect with ever increasing precision the respective contributions and interactions of many different neuronal populations within and between large70

scale networks. However, time is ripe to address this challenge and make new progress by exploiting the methodological advances and complementarities of current neuroimaging techniques, in both patients and healthy volunteers, in combination with a careful analysis of neglect components and underlying cognitive processes. Precise lesion mapping studies should be integrated with structural connectivity analysis and functional measures assessing distant consequences of focal brain destruction and/or disconnection, as well as high resolution imaging studies of cortical responses in well-defined conditions. Although not covered in this review, electroencephalography (EEG) and magnetoencephalography (MEG) recordings can also provide valuable information on intact versus altered dynamics of neural processing within spared networks after focal brain damage,146,155–158 but are too rarely used in patients with neglect or extinction. Such electrophysiological measures would be particularly useful to interpret functional changes observed in distant areas137 or distributed networks115 after parietal damage, and to determine whether such changes reflect early, late, or feedback/reentrant stages in stimulus processing,158 or even baseline changes arising before stimulus-evoked responses.138 Ultimately, better defining the intricate neural architecture implicated in neglect will not only provide fascinating knowledge on brain processes mediating our representation of space, selective attention, and conscious awareness, but it is also crucial to better understanding the handicapping disorders observed in brain-damaged patients and offer appropriate therapeutic interventions corresponding to individual deficits.

Acknowledgments This work was partly supported by Grants from the Swiss National Science Foundation (No. 114014) and the Geneva Academic Society (Foremane fund). This paper is dedicated to the memory of Jon Driver with whom many stimulating discussions about neglect were initiated and still continue today in the author’s work. Special thanks go to Arnaud Saj, Vincent Verdon, and Roland Vocat, who contributed to some of the work reviewed here and made helpful comments; and to Marteen Vaessen for creating Figure 2C.

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

Conflicts of interest The author declares no conflicts of interest.

References 1. Critchley, M. 1953. The Parietal Lobes. New York: Hafner. 2. Heilman, K.M., D.M. Pandya & N. Geschwind. 1970. Trimodal inattention following parietal lobe ablations. Trans. Am. Neurol. Assoc. 95: 259–268. 3. Driver, J. & P. Vuilleumier. 2002. “Spatial neglect.” In Encyclopedia of Cognitive Sciences. N. P. Group, Ed.: 211–218. New York: Macmillan Publishers Ltd. 4. Vuilleumier, P. & A. Saj. 2013. “Hemispatial neglect.” In The Behavioral and Cognitive Neurology of Stroke. Godefroy, O., Ed.: 126–157. Cambridge: Cambridge University Press. 5. Bisiach, E. & C. Luzzatti. 1978. Unilateral neglect of representational space. Cortex 14: 129–133. 6. Zorzi, M., K. Priftis & C. Umilta. 2002. Brain damage: neglect disrupts the mental number line. Nature 417: 138– 139. 7. Vuilleumier, P., S. Ortigue & P. Brugger. 2004. The number space and neglect. Cortex 40: 399–410. 8. Saj, A. et al. 2013. Left spatial neglect patients also neglect the left side of time. Psychol. Sci. In press. 9. Karnath, H.O. & M. Fetter. 1995. Ocular space exploration in the dark and its relation to subjective and objective body orientation in neglect patients with parietal lesions. Neuropsychologia 33: 371–377. 10. Karnath, H.O. & M.T. Perenin. 1998. Tactile exploration of peripersonal space in patients with neglect. Neuroreport 9: 2273–2277. 11. Vuilleumier, P.O. & R.D. Rafal. 2000. A systematic study of visual extinction. Between- and within-field deficits of attention in hemispatial neglect. Brain 123: 1263–1279. 12. Driver, J. & P. Vuilleumier. 2001. Perceptual awareness and its loss in unilateral neglect and extinction. Cognition 79: 39–88. 13. Vocat, R. et al. 2010. Anosognosia for hemiplegia: a clinicalanatomical prospective study. Brain 133: 3578–3597. 14. Mesulam, M.M. 1990. Large-scale neurocognitive networks and distributed processing for attention, language, and memory. Ann. Neurol. 28: 597–613. 15. Corbetta, M. & G.L. Shulman. 2011. Spatial neglect and attention networks. Annu. Rev. Neurosci. 34: 569–599. 16. Driver, J. 2001. A selective review of selective attention research from the past century. Br. J. Psychol. 92: 53–78. 17. Rafal, R.D. 1994. Neglect. Curr. Opin. Neurobiol. 4: 231– 236. 18. Driver, J., P. Vuilleumier & M. Husain. 2004. “Spatial neglect and extinction.” In The New Cognitive Neurosciences. Gazzaniga, M., Ed.: 589–606. Cambridge, MA: MIT Press. 19. Kastner, S. & M.A. Pinsk. 2004. Visual attention as a multilevel selection process. Cogn. Affect. Behav. Neurosci. 4: 483–500. 20. Vuilleumier, P. et al. 1998. Near and far visual space in unilateral neglect. Ann. Neurol. 43: 406–410. 21. Saj, A. & P. Vuilleumier. 2007. Neglect: remembering the space left behind. Curr. Biol. 17: R1060–1062.

Spatial neglect and the parietal lobe

22. Husain, M. & C. Rorden. 2003. Non-spatially lateralized mechanisms in hemispatial neglect. Nat. Rev. Neurosci. 4: 26–36. 23. Halligan, P.W., J.C. Marshall & D.T. Wade. 1989. Visuospatial neglect: underlying factors and test sensitivity. Lancet 2: 908–911. 24. Azouvi, P. et al. 2002. Sensitivity of clinical and behavioural tests of spatial neglect after right hemisphere stroke. J. Neurol. Neurosurg. Psychiatr. 73: 160–166. 25. Stone, S.P. et al. 1998. Unilateral neglect: a common but heterogeneous syndrome. Neurology 50: 1902–1905. 26. Vallar, G. & D. Perani. 1986. The anatomy of unilateral neglect after right-hemisphere stroke lesions. A clinical/CTscan correlation study in man. Neuropsychologia 24: 609– 622. 27. Mort, D.J. et al. 2003. The anatomy of visual neglect. Brain 126: 1986–1997. 28. Damasio, A.R., H. Damasio & H.C. Chui. 1980. Neglect following damage to frontal lobe or basal ganglia. Neuropsychologia 18: 123–132. 29. Husain, M. & C. Kennard. 1997. Distractor-dependent frontal neglect. Neuropsychologia 35: 829–841. 30. Rafal, R.D. & M.I. Posner. 1987. Deficits in human visual spatial attention following thalamic lesions. Proc. Natl. Acad. Sci. USA. 84: 7349–7353. 31. Karnath, H.O., M. Himmelbach & C. Rorden. 2002. The subcortical anatomy of human spatial neglect: putamen, caudate nucleus and pulvinar. Brain 125: 350– 360. 32. Karnath, H.O., S. Ferber & M. Himmelbach. 2001. Spatial awareness is a function of the temporal not the posterior parietal lobe. Nature 411: 950–953. 33. Bird, C.M. et al. 2006. Visual neglect after right posterior cerebral artery infarction. J. Neurol. Neurosurg. Psychiatr. 77: 1008–1012. 34. Verdon, V. et al. 2010. Neuroanatomy of hemispatial neglect and its functional components: a study using voxel-based lesion-symptom mapping. Brain 133: 880–894. 35. Samuelsson, H. et al. 1997. Anatomical and neurological correlates of acute and chronic visuospatial neglect following right hemisphere stroke. Cortex 33: 271–285. 36. Doricchi, F. et al. 2008. White matter (dis)connections and gray matter (dys)functions in visual neglect: gaining insights into the brain networks of spatial awareness. Cortex 44: 983–995. 37. Bisiach, E. et al. 1990. Perceptual and premotor factors of unilateral neglect. Neurology 40: 1278–1281. 38. Godefroy, O. et al. 1998. Brain–behaviour relationships. Some models and related statistical procedures for the study of brain-damaged patients. Brain 121: 1545–1556. 39. Bates, E. et al. 2003. Voxel-based lesion-symptom mapping. Nat. Neurosci. 6: 448–450. 40. Saj, A. et al. 2012. The anatomy underlying acute versus chronic spatial neglect also depends on clinical tests. Brain 135: e207; author reply e208. 41. Posner, M.I., Y. Cohen & R.D. Rafal. 1982. Neural systems control of spatial orienting. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 298: 187–198. 42. Friedrich, F.J. et al. 1998. Spatial attention deficits in humans: a comparison of superior parietal and temporal– parietal junction lesions. Neuropsychology 12: 193–207.

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

71

Spatial neglect and the parietal lobe

Vuilleumier

43. Bartolomeo, P. & S. Chokron. 2002. Orienting of attention in left unilateral neglect. Neurosci. Biobehav. Rev. 26: 217– 234. 44. Chica, A.B. et al. 2011. Spatial attention and conscious perception: the role of endogenous and exogenous orienting. Atten. Percept. Psychophys. 73: 1065–1081. 45. Corbetta, M. & G.L. Shulman. 2002. Control of goaldirected and stimulus-driven attention in the brain. Nat. Rev. Neurosci. 3: 201–215. 46. Karnath, H.O., M. Himmelbach & W. Kuker. 2003. The cortical substrate of visual extinction. Neuroreport 14: 437– 442. 47. Molenberghs, P., M.V. Sale & J.B. Mattingley. 2012. Is there a critical lesion site for unilateral spatial neglect? A metaanalysis using activation likelihood estimation. Front. Hum. Neurosci. 6: 78. 48. Chica, A.B., P. Bartolomeo & A. Valero-Cabre. 2011. Dorsal and ventral parietal contributions to spatial orienting in the human brain. J. Neurosci. 31: 8143–8149. 49. Molenberghs, P. et al. 2008. Convergence between lesionsymptom mapping and functional magnetic resonance imaging of spatially selective attention in the intact brain. J. Neurosci. 28: 3359–3373. 50. Gillebert, C.R. et al. 2011. Lesion evidence for the critical role of the intraparietal sulcus in spatial attention. Brain 134: 1694–1709. 51. Baylis, G., R. Rafal & J. Driver. 1993. Visual extinction and stimulus repetition. J. Cogn. Neurosci. 5: 453–466. 52. Stone, S.P., P.W. Halligan & R.J. Greenwood. 1993. The incidence of neglect phenomena and related disorders in patients with an acute right or left hemisphere stroke. Age Ageing 22: 46–52. 53. Gillebert, C.R. et al. 2013. Cytoarchitectonic mapping of attentional selection and reorienting in parietal cortex. Neuroimage 67: 257–272. 54. Bogler, C., S. Bode & J.D. Haynes. 2011. Decoding successive computational stages of saliency processing. Curr. Biol. 21: 1667–1671. 55. Schwartz, S. et al. 2005. Attentional load and sensory competition in human vision: modulation of fMRI responses by load at fixation during task-irrelevant stimulation in the peripheral visual field. Cereb. Cortex 15: 770–786. 56. Fink, G.R. et al. 2000. Neural consequences of competing stimuli in both visual hemifields: a physiological basis for visual extinction. Ann. Neurol. 47: 440–446. 57. Grefkes, C. & G.R. Fink. 2005. The functional organization of the intraparietal sulcus in humans and monkeys. J. Anat. 207: 3–17. 58. Chiu, Y.C. & S. Yantis. 2009. A domain-independent source of cognitive control for task sets: shifting spatial attention and switching categorization rules. J. Neurosci. 29: 3930– 3938. 59. Milner, A.D. & M. Harvey. 1995. Distortion of size perception in visuospatial neglect. Curr. Biol. 5: 85–89. 60. Bisiach, E. et al. 1996. Beyond unilateral neglect. Brain 119: 851–857. 61. Ladavas, E., M. Carletti & G. Gori. 1994. Automatic and voluntary orienting of attention in patients with visual neglect: horizontal and vertical dimensions. Neuropsychologia 32: 1195–1208.

72

62. Molenberghs, P. & M.V. Sale. 2011. Testing for spatial neglect with line bisection and target cancellation: are both tasks really unrelated? PloS One 6: e23017. 63. Vandenberghe, R. et al. 2005. Attentional responses to unattended stimuli in human parietal cortex. Brain 128: 2843– 2857. 64. Astafiev, S.V., G.L. Shulman & M. Corbetta. 2006. Visuospatial reorienting signals in the human temporo-parietal junction are independent of response selection. Eur. J. Neurosci. 23: 591–596. 65. Shulman, G.L. et al. 2010. Right hemisphere dominance during spatial selective attention and target detection occurs outside the dorsal frontoparietal network. J. Neurosci. 30: 3640–3651. 66. Corbetta, M., G. Patel & G.L. Shulman. 2008. The reorienting system of the human brain: from environment to theory of mind. Neuron 58: 306–324. 67. Heilman, K.M. & T. Van Den Abell. 1980. Right hemisphere dominance for attention: the mechanisms underlying hemispheric asymmetries of inattention (neglect). Neurology 30: 327–330. 68. Kahneman, D., A. Treisman & B.J. Gibbs. 1992. The reviewing of object files: Object-specific integration of information. Cogn. Psychol. 24: 175–219. 69. Hollingworth, A., C.C. Williams & J.M. Henderson. 2001. To see and remember: visually specific information is retained in memory from previously attended objects in natural scenes. Psychon. Bull. Rev. 8: 761–768. 70. Kanwisher, N. & M.C. Potter. 1989. Repetition blindness: The effects of stimulus modality and spatial displacement. Memory Cogn. 17: 117–124. 71. Milner, A.D. 1998. Streams and consciousness: visual awareness and the brain. Trends Cogn. Sci. 2: 25–30. 72. Baylis, G.C. et al. 2002. Visual extinction with double simultaneous stimulation: what is simultaneous? Neuropsychologia 40: 1027–1034. 73. Rorden, C. et al. 2009. Visual extinction: the effect of temporal and spatial bias. Neuropsychologia 47: 321–329. 74. Shisler, R.J., C.L. Gore & G.C. Baylis. 2004. Auditory extinction: the effect of stimulus similarity and task requirements. Neuropsychologia 42: 836–846. 75. Kamke, M.R. et al. 2012. Parietal disruption alters audiovisual binding in the sound-induced flash illusion. Neuroimage 62: 1334–1341. 76. Seghier, M.L. 2013. The angular gyrus: multiple functions and multiple subdivisions. Neuroscientist 19: 43–61. 77. Husain, M. et al. 1997. Abnormal temporal dynamics of visual attention in spatial neglect patients. Nature 385: 154– 156. 78. Battelli, L. et al. 2003. Bilateral deficits of transient visual attention in right parietal patients. Brain 126: 2164–2174. 79. Roberts, K.L. et al. 2012. Spatial and temporal attention deficits following brain injury: a neuroanatomical decomposition of the temporal order judgement task. Cogn. Neuropsychol. 29: 300–324. 80. Battelli, L., A. Pascual-Leone & P. Cavanagh. 2007. The “when” pathway of the right parietal lobe. Trends Cogn. Sci. 11: 204–210. 81. Malhotra, P. et al. 2005. Spatial working memory capacity in unilateral neglect. Brain 128: 424–435.

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

Vuilleumier

82. Pisella, L., N. Berberovic & J.B. Mattingley. 2004. Impaired working memory for location but not for colour or shape in visual neglect: a comparison of parietal and non-parietal lesions. Cortex 40: 379–390. 83. Kristjansson, A. & P. Vuilleumier. 2010. Disruption of spatial memory in visual search in the left visual field in patients with hemispatial neglect. Vision Res. 50: 1426–1435. 84. Vuilleumier, P. et al. 2007. Impaired perceptual memory of locations across gaze-shifts in patients with unilateral spatial neglect. J. Cogn. Neurosci. 19: 1388–1406. 85. Pisella, L. & J.B. Mattingley. 2004. The contribution of spatial remapping impairments to unilateral visual neglect. Neurosci. Biobehav. Rev. 28: 181–200. 86. Robertson, I.H. et al. 1998. Phasic alerting of neglect patients overcomes their spatial deficit in visual awareness. Nature 395: 169–172. 87. Finke, K. et al. 2012. How does phasic alerting improve performance in patients with unilateral neglect? A systematic analysis of attentional processing capacity and spatial weighting mechanisms. Neuropsychologia 50: 1178–1189. 88. Dieterich, M. 2007. Functional brain imaging: a window into the visuo-vestibular systems. Curr. Opin. Neurol. 20: 12–18. 89. Janzen, J. et al. 2008. Neural correlates of hemispheric dominance and ipsilaterality within the vestibular system. Neuroimage 42: 1508–1518. 90. zu Eulenburg, P. et al. 2012. Meta-analytical definition and functional connectivity of the human vestibular cortex. Neuroimage 60: 162–169. 91. Richard, C. et al. 2004. Straight ahead in spatial neglect: evidence that space is shifted, not rotated. Neurology 63: 2136–2138. 92. Karnath, H.O. 1994. Subjective body orientation in neglect and the interactive contribution of neck muscle proprioception and vestibular stimulation. Brain 117: 1001–1012. 93. Rizzolatti, G. & R. Camarda. 1987. “Neural circuits for spatial attention and unilateral neglect.” In Neurophysiological and Neuropsychological Aspects of Spatial Neglect, Vol. 45. Jeannerod, M., Ed.: 289–314. Amsterdam: North-Holland. 94. Binder, J. et al. 1992. Distinct syndromes of hemineglect. Arch. Neurol. 49: 1187–1194. 95. Rorden, C., M. Fruhmann Berger & H.O. Karnath. 2006. Disturbed line bisection is associated with posterior brain lesions. Brain Res. 1080: 17–25. 96. Medina, J. et al. 2009. Neural substrates of visuospatial processing in distinct reference frames: evidence from unilateral spatial neglect. J. Cogn. Neurosci. 21: 2073–2084. 97. Hillis, A.E. et al. 2005. Anatomy of spatial attention: insights from perfusion imaging and hemispatial neglect in acute stroke. J. Neurosci. 25: 3161–3167. 98. Grimsen, C., H. Hildebrandt & M. Fahle. 2008. Dissociation of egocentric and allocentric coding of space in visual search after right middle cerebral artery stroke. Neuropsychologia 46: 902–914. 99. Chechlacz, M. et al. 2010. Separating neural correlates of allocentric and egocentric neglect: distinct cortical sites and common white matter disconnections. Cogn. Neuropsychol. 27: 277–303. 100. Rorden, C. et al. 2012. Allocentric neglect strongly associated with egocentric neglect. Neuropsychologia 50: 1151– 1157.

Spatial neglect and the parietal lobe

101. Galati, G. et al. 2000. The neural basis of egocentric and allocentric coding of space in humans: a functional magnetic resonance study. Exp. Brain Res. 133: 156–164. 102. Neggers, S.F. et al. 2006. Interactions between ego- and allocentric neuronal representations of space. Neuroimage 31: 320–331. 103. Chen, Q. et al. 2012. Neural interaction between spatial domain and spatial reference frame in parietal–occipital junction. J. Cogn. Neurosci. 24: 2223–2236. 104. Lane, A.R. et al. 2013. Near and far space: Understanding the neural mechanisms of spatial attention. Hum. Brain Mapp. 34: 356–366. 105. Vuilleumier, P. et al. 1999. To see better to the left when looking more to the right: effects of gaze direction and frames of spatial coordinates in unilateral neglect. J. Int. Neuropsychol. Soc. 5: 75–82. 106. Committeri, G. et al. 2007. Neural bases of personal and extrapersonal neglect in humans. Brain 130: 431–441. 107. Baas, U. et al. 2011. Personal neglect-a disorder of body representation? Neuropsychologia 49: 898–905. 108. Chechlacz, M., P. Rotshtein & G.W. Humphreys. 2012. Neuroanatomical dissections of unilateral visual neglect symptoms: ALE meta-analysis of lesion-symptom mapping. Front. Hum. Neurosci. 6: 230. 109. Khurshid, S. et al. 2012. Reperfusion of specific cortical areas is associated with improvement in distinct forms of hemispatial neglect. Cortex 48: 530–539. 110. Kinsella, G. et al. 1993. Analysis of the syndrome of unilateral neglect. Cortex 29: 135–140. 111. McGlinchey-Berroth, R. et al. 1996. Assessment of neglect reveals dissociable behavioral but not neuroanatomical subtypes. J. Int. Neuropsychol. Soc. 2: 441–451. 112. Ronchi, R. et al. 2009. Perseveration in left spatial neglect: drawing and cancellation tasks. Cortex 45: 300–312. 113. Karnath, H.O. & C. Rorden. 2012. The anatomy of spatial neglect. Neuropsychologia 50: 1010–1017. 114. Bartolomeo, P., M. Thiebaut de Schotten & F. Doricchi. 2007. Left unilateral neglect as a disconnection syndrome. Cereb. Cortex 17: 2479–2490. 115. He, B.J. et al. 2007. Breakdown of functional connectivity in frontoparietal networks underlies behavioral deficits in spatial neglect. Neuron 53: 905–918. 116. Karnath, H.O., C. Rorden & L.F. Ticini. 2009. Damage to white matter fiber tracts in acute spatial neglect. Cereb. Cortex 19: 2331–2337. 117. Thiebaut de Schotten, M. et al. 2012. Damage to white matter pathways in subacute and chronic spatial neglect: a group study and 2 single-case studies with complete virtual “in vivo” tractography dissection. Cereb. Cortex. Nov 15. [Epub ahead of print]. 118. Thiebaut de Schotten, M. et al. 2005. Direct evidence for a parietal-frontal pathway subserving spatial awareness in humans. Science 309: 2226–2228. 119. Thiebaut de Schotten, M. et al. 2011. A lateralized brain network for visuospatial attention. Nat. Neurosci. 14: 1245– 1246. 120. Rode, G. et al. 2010. Representation and disconnection in imaginal neglect. Neuropsychologia 48: 2903–2911. 121. Bozzali, M. et al. 2012. Microstructural damage of the posterior corpus callosum contributes to the clinical severity of neglect. PloS One 7: e48079.

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74 

73

Spatial neglect and the parietal lobe

Vuilleumier

122. Gaffan, D. & J. Hornak. 1997. Visual neglect in the monkey: representation and disconnection. Brain 120: 1647– 1657. 123. Hahnloser, R.H., R.J. Douglas & K. Hepp. 2002. Attentional recruitment of inter-areal recurrent networks for selective gain control. Neural. Comput. 14: 1669–1689. 124. Lee, D.K. et al. 1999. Attention activates winner-take-all competition among visual filters. Nat. Neurosci. 2: 375– 381. 125. Lumer, E.D. 2000. Effects of spike timing on winner-takeall competition in model cortical circuits. Neural. Comput. 12: 181–194. 126. Greenberg, A.S. et al. 2012. Visuotopic cortical connectivity underlying attention revealed with white-matter tractography. J. Neurosci. 32: 2773–2782. 127. Saalmann, Y.B. et al. 2012. The pulvinar regulates information transmission between cortical areas based on attention demands. Science 337: 753–756. 128. Lomber, S.G., S. Malhotra & J.M. Sprague. 2007. Restoration of acoustic orienting into a cortically deaf hemifield by reversible deactivation of the contralesional superior colliculus: the acoustic “Sprague Effect.” J. Neurophysiol. 97: 979–993. 129. Vuilleumier, P., D. Hester & G. Assal. 1996. Unilateral spatial neglect recovery after sequential strokes. Neurology 46: 184–189. 130. Hillis, A.E. et al. 2002. Subcortical aphasia and neglect in acute stroke: the role of cortical hypoperfusion. Brain 125: 1094–1104. 131. Corbetta, M. et al. 2005. Neural basis and recovery of spatial attention deficits in spatial neglect. Nat. Neurosci. 8: 1603– 1610. 132. Carter, A.R. et al. 2010. Resting interhemispheric functional magnetic resonance imaging connectivity predicts performance after stroke. Ann. Neurol. 67: 365–375. 133. He, B.J. et al. 2007. The role of impaired neuronal communication in neurological disorders. Curr. Opin. Neurol. 20: 655–660. 134. Umarova, R.M. et al. 2011. Acute visual neglect and extinction: distinct functional state of the visuospatial attention system. Brain 134: 3310–3325. 135. Thimm, M., G.R. Fink & W. Sturm. 2008. Neural correlates of recovery from acute hemispatial neglect. Restor. Neurol. Neurosci. 26: 481–492. 136. Saj, A. et al. 2013. Prism adaptation enhances activity of intact fronto-parietal areas in both hemispheres in neglect patients. Cortex 49: 107–119. 137. Vuilleumier, P. et al. 2008. Abnormal attentional modulation of retinotopic cortex in parietal patients with spatial neglect. Curr. Biol. 18: 1525–1529. 138. Vuilleumier, P. 2005. “Visual extinction and hemispatial neglect after brain damage: Neurophysiological basis of residual processing.” In: Neurobiology of Attention. Itti, L. G., Rees, & J. Tsotsos, Eds.: 351–357. San Diego: Elsevier. 139. Valenza, N. et al. 2004. Tactile awareness and limb position in neglect: functional magnetic resonance imaging. Ann. Neurol. 55: 139–143. 140. Aglioti, S., N. Smania & A. Peru. 1999. Frames of reference for mapping tactile stimuli in brain-damaged patients. J. Cogn. Neurosci. 11: 67–79.

74

141. Hillis, A.E. et al. 2006. Neural correlates of modalityspecific spatial extinction. J. Cogn. Neurosci. 18: 1889–1898. 142. Kooistra, C.A. & K.M. Heilman. 1989. Hemispatial visual inattention masquerading as hemianopia. Neurology 39: 1125–1172. 143. Saj, A. et al. 2012. Time since stroke influences the impact of hemianopia and spatial neglect on visual–spatial tasks. Neuropsychology 26: 37–44. 144. Sterzi, R. et al. 1993. Hemianopia, hemianaesthesia, and hemiplegia after right and left hemisphere damage. A hemispheric difference. J. Neurol. Neurosurg. Psychiatr. 56: 308– 310. 145. Rees, G. et al. 2002. Neural correlates of conscious and unconscious vision in parietal extinction. Neurocase 8: 387– 393. 146. Vuilleumier, P. et al. 2001. Neural fate of seen and unseen faces in visuospatial neglect: a combined event-related functional MRI and event-related potential study. Proc. Natl. Acad. Sci. USA 98: 3495–3500. 147. Sarri, M., F. Blankenburg & J. Driver. 2006. Neural correlates of crossmodal visual-tactile extinction and of tactile awareness revealed by fMRI in a right-hemisphere stroke patient. Neuropsychologia 44: 2398–2410. 148. Beversdorf, D.Q., J.D. Hughes & K.M. Heilman. 2008. Functional MRI of the primary somatosensory cortex in extinction to simultaneous bilateral tactile stimuli due to right temporal lobe stroke. Neurocase 14: 419– 424. 149. Dehaene, S. et al. 2006. Conscious, preconscious, and subliminal processing: a testable taxonomy. Trends. Cogn. Sci. 10: 204–211. 150. Saj, A. et al. 2008. Reducing rightward bias of subjective straight ahead in neglect patients by changes in body orientation. J. Neurol. Neurosurg. Psychiatr. 79: 991–996. 151. Saj, A. et al. 2013. Where is straight ahead to a patient with unilateral vestibular loss? Cortex 49: 1219–1228. 152. Marshall, J.C. & P.W. Halligan. 1994. Independent properties of normal hemispheric specialization predict some characteristics of visuo-spatial neglect. Cortex 30: 509– 517. 153. Wojciulik, E. et al. 2001. Spatial working memory deficit in unilateral neglect. Neuropsychologia 39: 390–396. 154. Balslev, D. & R.C. Miall. 2008. Eye position representation in human anterior parietal cortex. J. Neurosci. 28: 8968– 8972. 155. Verleger, R. et al. 1996. On-line brain potential correlates of right parietal patients’ attentional deficit. Electroencephalogr Clin. Neurophysiol. 99: 444–457. 156. Marzi, C.A. et al. 2000. Electrophysiological correlates of conscious vision: evidence from unilateral extinction. J. Cogn. Neurosci. 12: 869–877. 157. Driver, J. et al. 2001. Functional MRI and evoked potential correlates of conscious and unconscious vision in parietal extinction patients. Neuroimage 14: 68–75. 158. Di Russo, F. et al. 2008. Impaired visual processing of contralesional stimuli in neglect patients: a visual-evoked potential study. Brain 131: 842–854. 159. Bartolomeo, P., M. Thiebaut de Schotten & A.B. Chica. 2012. Brain networks of visuospatial attention and their disruption in visual neglect. Front. Hum. Neurosci. 6: 110.

C 2013 New York Academy of Sciences. Ann. N.Y. Acad. Sci. 1296 (2013) 50–74