Masud Husain (2008), Scholarpedia, 3(2):3681. | doi:10.4249/scholarpedia.3681 | revision #170779 [link to/cite this article] |
Hemineglect, also known asunilateral neglect,hemispatial neglect orspatial neglect, is a common and disabling condition followingbrain damage in which patients fail to be aware of items to one side of space. Neglect is most prominent and long-lasting after damage to theright hemisphere of the human brain, particularly following a stroke. Such individuals with right-sided brain damage often fail to be aware of objects to their left, demonstrating neglect of leftward items.
The deficit may be so profound that patients are unaware of large objects, even people, towards their neglected orcontralesional side - the side of space opposite brain damage. They may eat from only one side of a plate, write on one side of a page, shave or make-up only the non-neglected oripsilesional side of their face (same side as brain damage). Their drawings may fail to include items towards the neglected side, for example when placing the numbers in a drawing of a clock (Fig.1). Many patients are often also unaware they have a deficit (anosognosia).
Classically, the neglect syndrome has been associated with damage to theright posteriorparietal cortex. More recent studies have begun to challenge this view, suggesting instead that a more widespread network of areas may be involved, including those that have been implicated in studies ofneuroimaging ofattention (discussed further in theNeuroanatomy of neglect below; see also Corbetta & Shulman, 2002 and Husain & Rorden, 2003). Differences in the location and extent of lesions (brain damage) across patients may contribute to the heterogeneity of the condition.
Individuals with neglect do not necessarily suffer from any primary disorder of sensation or movement. For example, their visual fields may be completely intact. So they may be able to detect a salient, bright item on their neglected side if it is flashed on a dark background. But even though they may not be blind, patients may fail to report the same stimulus if it is presented in a cluttered background. People with neglect may also have no weakness of their eye or limb muscles, yet fail to direct their gaze or hands to explore contralesional space (e.g., Behrmann et al., 1997; Husain et al., 2001). Their visual search may, instead, be directed only towards items to their ipsilesional side (Fig.2).
Neglect may be multimodal, extending to involve visual, auditory and somatosensory systems. It may also affect memories of scenes - so-calledrepresentational neglect. In their famous experiment, Bisiach and Luzzatti (1978) asked patients with left-sided visual neglect from Milan to imagine viewing the central square, the Piazza del Duomo, from the cathedral in the centre of the square. They reported that these patients often omitted to mention places or streets off to the left side of the squarefrom that viewpoint. However, when asked to imagine looking directly at the cathedral, i.e., shifting their viewpoint by 180 degrees, the same individuals now neglected to mention places on the side of the square they had just reported - ones which now fell to their left from the new imagined point of view. Instead, they now recalled places to the right of the new imagined vantage point.
Importantly, the border between neglected space and non-neglected space is:
The probability of reporting items varies in many patients (e.g., Kaplan et al., 1991; Smania et al., 1998; Chatterjee et al., 1999):
Several studies have revealed that even though a stimulus to the neglected side of space may fail to be reported, it may nevertheless be processed to a substantial degree - to the level of its identity or meaning (Berti & Rizzolatti, 1992). The most striking examples come from investigations ofsemantic priming (e.g., McGlinchey-Berroth et al., 1993). In one experiment, stimuli to the left which were unreported (neglected) by the patient, nevertheless speeded subsequent responses to a word if it was semantically related to the neglected item (e.g., tree and apple) compared to if it was unrelated (e.g., bed and apple). Another study showed a patient pictures of an undamaged house and one burning with the flames on the left side of the house (Marshall & Halligan, 1988). When asked which house they would prefer to live in, a patient with left-sided neglect, who did not report seeing the flames of the burning house, nevertheless chose the undamaged house significantly more often. Note that the level to which stimuli may be processed implicitly or unconsciously in neglect patients is far more substantial than that reported in individuals withblindsight who have primary visual field deficits. Functional imaging studies have demonstrated activation in intact early visual areas even when patients have noconscious awareness for visual stimuli (Rees et al., 2000; Rees 2001; Vuilleumier et al., 2001). See also sections inAttention and Consciousness.
Neglect may be measured in various ways. Different patients may show deficits on different types of test, so there is no single test used to diagnose the condition (Azouvi et al., 2002; Parton et al., 2004). Many patients will show neglect behaviour in their everyday life when interacting with others, so observation of the patient is crucial. In addition, batteries of neuropsychological tests have been devised to capture and quantify the degree of neglect.
Most batteries include:
Classically, neglect has been associated with lesions of the rightposterior parietal cortex, particularly the inferior parietal lobe (IPL) or temporoparietal junction (TPJ) (Vallar & Perani, 1986) (Fig.4).
More recently, some authors have proposed a critical association with damage to the right superior temporal gyrus (Karnath et al., 2001), though this view has been strongly challenged (Mort et al., 2003) and remains a subject of much debate and controversy (Doricchi & Tomaiuolo, 2003; Karnath et al., 2004a, 2004b; Mort et al., 2004). What is clear, however, is that there is no one single lesion location associated with the syndrome.
In addition to posterior cortical areas, most researchers agree thatsubcortical lesions (for example, involving thethalamus andbasal ganglia) that do not encroach directly on the cortex can lead to neglect, although this may be via indirect effects on overlying cortical regions (Hillis et al., 2002, 2005). Moreover, it has also become evident that isolated lesions of the rightfrontal lobe may be associated with neglect, without involvement of posterior parietal or temporal regions (Husain & Kennard, 1996). Recent studies also implicatewhite matter pathways, connecting, for example, posterior cortical and frontal regions in individuals with neglect (Thiebaut de Schotten et al., 2005; Bartolomeo et al., 2007).
Different patients have different combinations of posterior cortical, frontal, subcortical and white matter damage. This heterogeneity may be a key factor determining the diversity of functional deficits reported in this condition (see Mechanisms underlying neglect).
Left hemisphere damage may also lead to neglect of rightward items, though this is usually less profound and less long-lasting than leftward neglect following right brain damage. The development oflanguage and praxis in left posterior and frontal regions may be a reason for this hemispheric difference. Right hemisphere regions in the inferior parietal and frontal lobe may have a special role in sustained attention and right parietal regions have also been implicated in spatialworking memory. When combined with a rightward directional bias following strokes to these regions (see Directiona Attentional Deficits below), this might lead to more profound neglect after right compared to left hemisphere lesions.
Many different cognitive deficits have been identified in patients with neglect (Driver & Mattingley, 1998; Heilman & Watson, 2001; Halligan et al., 2003; Husain & Rorden, 2003; Mesulam, 1999). These experimental findings have led to a range of hypotheses about the mechanisms underlying the condition. A large number ofdissociations have been reported with some patients showing deficit A but not B, and others showing B but not A (Vallar, 1998). Many patients show combinations of deficits, but the exact combination differs across patients (Buxbaum et al., 2004).
One way to consider the mechanisms underlying the syndrome is to divide them intospatial or directional deficits versusnon-spatial or non-directional ones.
Many researchers have proposed that neglect may be due to a deficit in directing spatial attention, specifically indisengaging attention from ipsilesional objects and shifting it contralesionally towards the neglected side of space (for reviews see Losier & Klein, 2001 and Bartolomeo & Chokron, 2002). Such a mechanism was originally implicated in patients with visual extinction following unilateral brain damage (Posner et al., 1984). Cueing attention towards the neglected side of space can help to reduce spatial biases, for example in line bisection (Riddoch & Humphreys, 1983).
Other investigators have emphasized a spatiallylateralised bias or gradient of attention in neglect, due to disruption of the normal balance between the hemispheres in directing attention (Kinsbourne, 1993). Thus, after right hemisphere damage, left hemisphere mechanisms which normally orient attention rightwards may be left relatively unopposed (Corbetta et al., 2005). Hence the ipsilesional bias in attention observed in patients with neglect.
Some researchers have considered the spatial attention deficit in terms of the biased competition theory of attention, with ipsilesional stimuli winning in the competition for selection over contralesional ones in neglect (Duncan et al., 1997). According to such accounts, more stimuli on the non-neglected, ipsilesional side would also hinder attention being directed toward contralesional items (see Fig. 3).
Evidence from a study in which the eye position of right hemisphere patients was monitored shows that their rightward bias can be completely corrected if thevisual salience of stimuli to the left is increased relative to those to the right (Bays et al., 2010). Indeed, if leftward stimuli are made highly salient then these patients' gaze can be shifted into previously neglected space. These findings would be consistent with an ipsilesional directional bias modulated by competition between leftward and rightward items in neglect patients.
Several studies have shown that the degree of neglect may be modulated by the position of stimuli relative to the trunk, head, eye position and even gravitational field (e.g., Ladavas, 1987; Karnath et al., 1998). Thus different egocentric spatial reference frames appear to exert an influence on the sector of space that is neglected. Some investigators have also reported neglect for stimuli in near space or far space in different patients. All these reports have drawn inspiration from computational considerations of the transformations involved in sensorimotor control, as well as neurophysiological studies of the role of monkey posterior parietal cortex in sensorimotor transformations and motor control (Andersen, 1997). In general, these findings support the hypothesis that there may be a deficit of spatial representation in neglect (Bisiach & Luzzatti, 1978). However, an impaired spatial representation in neglect might also be secondary to reduced attention or exploration of contralesional space.
Evidence also exists for a deficit in directing eye or limb movements contralesionally or to targets in contralesional space in some individuals with neglect (Heilman et al., 1985; Mattingley et al., 1992). This may be a disorder of initiating movements (sometimes referred to asdirectional hypokinesia) or in slowness of movement execution (termeddirectionalbradykinesia). Directional motor deficits may be modulated by locations of visual targets. One study has demonstrated slowness in initiatingleftward movements to targets in left hemispace, but not those in right hemispace, in right parietal patients with neglect (Mattingley et al., 1998).
Recent investigations have revealed that some neglect patients also have difficulty in keeping track of spatial locations across saccadic eye movements (Husain et al., 2001; Mannan et al., 2005). Such a deficit in spatial working memory appears to exacerbate any lateralised biases in these patients. The findings suggest limitations invisual short term memory, particularly for the locations of objects (Pisella et al., 2004; Malhotra et al., 2005; Ferber & Danckert 2006).
A range of techniques has been used to probe non-spatial or non-directional deficits. Theattentional blink paradigm has revealed a profound and long-lasting deficit in the temporaldynamics of visual processing for stimuli presented at fixation in right-hemisphere neglect patients (Husain et al., 1997). When attention is engaged on one item, such patients have difficulty in attending to subsequent items for >1 second, even when items are presented centrally, a finding reminiscent of non-spatial extinction in parietal patients (Humpreys et al. 1994).
Bilateral deficits (i.e. on both sides of space) have also been reported in parietal patients, with reduced capacity to encode visual stimuli presented transiently in either visual field (Duncan et al., 1999; Batelli et al., 2001).
Several groups have also reported impaired ability tosustain attention or maintain vigilance over protracted periods of time in patients with neglect, even for central auditory stimuli (see Robertson, 2001). Others have shown that there may be a bilateral constriction of the effective field ofvision - the sector of space that can be attended to - which may lead to 'local bias' and failure to attend 'globally' to the periphery (Rafal, 1994; Russell et al., 2004). Deficits in 'global' visual processing have also been reported in non-neglect patients with right temporoparietal lesions (Robertson et al., 1988).
In the past, there has been a great deal of interest in object-based attention deficits in neglect (Driver et al., 1994; for a review, see Walker, 1995), although a pure-object centred deficit is probably extremely rare (Driver & Pouget, 2000).
Combinations of spatial/directional and non-spatial deficits appear to be present in different patients with neglect (Buxbaum et al., 2004). Moreover, some of these deficits can exist in isolation of the neglect syndrome. For example, deficits in spatial working memory or sustained attention have been documented in right-hemisphere patients without neglect. However, when combined with spatially lateralised or directional biases, these deficits can serve to exacerbate the severity of neglect (see Husain & Rorden, 2003).
The combination of directional bias plus non-spatial / non-directional deficits may explain why right hemisphere lesions lead to more severe and longer lasting neglect. Both left and right hemisphere lesions may produce directional biases (e.g. extinction) but the right hemisphere may have a special role in sustained attention and spatial working memory. When deficits in these domains are comined with a directional bias, neglect may be far more severe.
Several types of therapy or experimental modulation have been attempted to ameliorate neglect (for reviews see Robertson & Halligan, 1999; Barrett et al., 2006; Luate et al., 2006; Singh-Curry & Husain, 2008). These have consisted either of behavioural interventions (sometimes including the use of specialist devices) or drug treatments. The results of these studies have been variable, most likely due to heterogeneity between patients and the small numbers of individuals tested. It is unlikely that one treatment will be suitable for all patients with neglect because, as we have seen, different patients have different combinations of underlying cognitive deficit.
A related concern is that for a particular treatment to be effective it may be crucial that certain brain regions or functions are intact. For example, learning a new strategy may depend upon the integrity of prefrontal cortical regions. Patients who have large parts of such critical brain regions destroyed may not be so responsive to certain treatment approaches. Thus mapping the lesions of patients involved in trials may be important to understand variations in response to therapy.
It is now generally acknowledged that interventions which simply attempt to improve exploration of contralesional space, for example by directing gaze in that direction on search tasks, are effective in reducing neglect on those tasks but these improvements fail to generalize or 'transfer' to amelioration of neglect in everyday life. When placed back into everyday situations, patients do not scan the neglected side of space significantly better following such training alone. However, when combined with vibration of the contralesional neck muscles, there has been evidence for significant improvement in neglect and functional outcome measures for up to 2 months follow-up (Schindler et al., 2002). Neck muscle vibration is thought to effect a recentering of the egocentric frame of reference, by an illusory modification of the afferent information regarding the orientation of the head in space.
In recent years there has also been a great deal of interest in prism adaptation as a means of improving neglect (Rossetti et al., 1998; Frassinetti et al., 2002). The process of adapting to rightward-deviating prisms appears to lead to long-lasting improvements in left-sided neglect. However, the mechanisms underlying this shift in behaviour remain controversial, and it has become evident that not all patients respond to this form of intervention.
Teaching patients to self-alert or use an alerting device that acts to arouse them has shown benefits in some individuals, as has limb activation therapy which involves directing attention to the contralesional hand (Robertson, 1999).
Some groups have reported improvements in neglect following repetitivetranscranial magnetic stimulation administered over the contralesional hemisphere (e.g. Oliveri et al., 2001). The underlying mechanism is considered to be short-lastinginhibition of the intact side of the brain, thereby redressing any imbalance in attentional orienting.
Two different classes of pharmacological agent have been used, targeting dopaminergic, noradrenergic, or both systems. Dopaminergic drugs such as levodopa, bromocriptine and apomorphine have been reported to have positive effects in some individuals, but the studies are extremely small and reported responses variable (Fleet et al., 1987; Geminiani et al., 1998; Grujic et al., 1998; Mukand et al., 2001). The α2-noradrenergic agonist, guanfacine, which has been used successfully in some children with attention deficit hyperactivity disorder, may also be beneficial in some individuals with neglect (Malhotra et al., 2006). Again, the extent of brain damage and the integrity of critical brain regions may be factors which account for response variability.
There has been no reported attempt to combine behavioural or device-based interventions with neuropharmacological modulation. In the future, such concurrent treatment might potentially be a very powerful approach if there were synergistic effects.
Internal references
Useful general reviews include:
Books containing in-depth reviews and opinions include:
Author's webpages:
Other labs: