Spatial Dimensions of Automatic and Voluntary Orienting Components of Attention
Dipartimento di Psicologia, Universita di Bologna, Bologna, Italy
Selective Attentional Deficits for Different Sectors of Space
Hemispatial neglect (neglect for the sake of brevity) occurs when patients do not report, respond or orient to stimuli presented contralaterally to the lesioned hemisphere, provided the deficit cannot be attributed to sensory or motor impairments (see, e.g. papers in Jeannerod, 1987). Because neglect is almost always observed after right parietal lesions (Vallar & Perani, 1987), in which follows I will consider only neglect for the left side of space.
Neglect is typically discussed in terms of selective difficulty along the horizontal plane, but there are good reasons to expect neglect for other spatial planes. Objects have a three-dimensional representation and it has been shown that attention can be oriented in depth as well as in the horizontal and vertical dimensions (Gawryszewski, Riggio, Rizzolatti, & Umiltà, 1987). Therefore, if attention can be oriented along each of these dimensions and neglect is defined as a deficit in the orienting component of attention, then it is reasonable to expect attentional deficit related to each of these spatial dimensions. In this section, I will confine myself largely to discuss neglect as a deficit in the automatic orienting component of attention, which can manifest itself in the horizontal dimension as well as in the vertical and depth dimensions. The hypothesis that the attentional deficit is not related to the voluntary orienting component of attention will be also discussed. Moreover, an attempt will be made to explain the deficit by taking into account the pre-motor theory of attention (Rizzolatti, Riggio, Dascola, & Umiltà, 1987; Tassinari, Alioti, Chelazzi, Marzi, & Berlucchi, 1987).
Horizontal Dimension
Most of the neuropsychological investigations of human visual neglect have produced an enormous body of data about the perceptual aspect of the deficit and much less attention has been paid to the role of the response coding in determining the deficit. This is surprising because it is well known that normal subjects codify the spatial positions of the stimuli as well as that of the response, and that the employment of one spatial code for the classification of the response determines also the spatial code used for the classification of the stimuli and vice versa. When these spatial codes are applied on the horizontal dimension, the outcome of this mental operation is the specification of the spatial codes “left” and “right” both for stimuli and responses. In the case of space of stimulation, the specification of “left” and “right” codes can be performed for example on the basis of the retinotopic co-ordinates (e.g. left or right visual field) or the environmental co-ordinates (e.g. the relative position of two stimuli). In the case of a manual response, the specification of “left” and “right” codes can be made by taking into account the hand used to respond and its actual position in space (Làdavas & Moscovitch, 1984; Umiltà & Nicoletti, 1990). Since it has been proposed that spatial codes are formed through an attentional process (Umiltà & Liotti, 1987), it is not surprising to find a neglect for the space of stimulation as well as for the space of response. For sake of clarity, I will discuss the two kinds of neglect in two separate sections, although, up to now, there is no empirical evidence for considering them two independent aspects of horizontal neglect.
Neglect for the Space of Stimulation
Definition of Left Space. Neglect typically affects the left side of space, but the definition of “what is left” depends on the frame of reference adopted by the subject. Objects per se do not have intrinsic lefts and rights, and it is the viewer who attributes left and right to objects on the basis of different co-ordinate systems.
In the egocentric frame of reference, the spatial locations of the stimuli are coded according to body co-ordinates (retinal, head and body midline). The relevance of the retinal co-ordinates has been proved, among others, by Làdavas (1987; 1990), who showed that under restricted lateralised presentation of visual stimuli, patients with extinction were less accurate and slower to respond to left than to right visual field stimuli.
In the allocentric frame of reference, the distinction between left and right operates on the basis of stimulus display (display-centred co-ordinates) or object display (object-centred co-ordinates). The relevance of the allocentric co-ordinates is provided by those studies which show that patients with neglect, presented with a visual display, can ignore one side of each individual object in a scene (object-centred co-ordinates) or one side of the scene as a whole (display-centred co-ordinates) (Driver & Halligan, 1991; Gainotti, D’Erme, & De Bonis, 1989). Therefore, it can be said that patients always ignore the left side of a representation, but the co-ordinates according to which left–right spatial codes are formed can vary, probably in accordance with task demands.
The finding that visual neglect manifests itself in object-centred and/or display-centred co-ordinates as well as in body co-ordinates (retinal, head and body midline), is relevant per se because it shows that orienting attention in space is a dynamic process which operates according to different systems of co-ordinates and mainly according to task demands. But it must be pointed out that, although this is an important issue, it does not explain in any way why patients with neglect explicitly ignore one side of the display and/or object and/or egocentric space. This issue will be addressed next.
Neglect as a Result of an Hyperattention Deficit. In order to understand why patients ignore some specific portions of space, it is important to study how the attention of these patients is distributed in the visual field. In a study designed to face this problem (Làdavas, 1990), it was shown that their attention is not equally distributed in the visual field but is focused on the right-most portion of the visual display. Attentional performance of patients with visual extinction was assessed in the experimental condition in which they had to pay attention simultaneously to three spatial positions to the left, right and directly above the fixation stimulus (experiment 1: distributed attention condition) and to only one of the three positions at a time (experiment 2: focused attention condition). The results showed that in experiment 1, the speed (and accuracy) of response to horizontally aligned stimuli increased (decrease in the case of accuracy) gradually from right to left. Reaction times (RTs) to the right stimulus were faster than to the central and left stimuli. Moreover, RTs to the central stimulus were slower than to the right stimulus and faster than to the left stimulus. In contrast, in experiment 2, RTs to the central stimulus were faster than the RTs to the left and right stimuli, and RTs to the right stimulus were faster than the RTs to the left stimulus. More important, speed and accuracy of response to the right stimulus in the distributed attention condition (experiment 1) and in the focused attention condition (experiment 2) were the same. These results showed that, in experiment 1, although the patients were requested to distribute attention simultaneously on the three spatial positions, they focused attention on the right-most stimulus and increased attention to the right was accompanied by decreased attention to the left.
According to several two-stage models of visual attention (for a review, see Umiltà, 1988), attention can be allocated evenly across a large portion of the visual field (i.e. the distributed mode), or can be concentrated on a restricted portion of it (i.e. the focal mode). When attention is in the distributed mode, all possible stimulus positions are processed in parallel at a uniform and relatively slow rate. In the focal mode, processing of the position on which attention is selectively allocated is facilitated, whereas processing of the other positions is inhibited. This is because the area to which attention is allocated can be varied (it can be as small as 1°, or even less), and efficiency of processing decreases as the area covered by attention increases (see, e.g. Castiello & Umiltà, 1990).
On the basis of this model, the results of Làdavas’s study seem to suggest that in patients with neglect, spatial attention cannot be allocated to multiple regions of the visual field because it is narrowly focused on the most ipsilesional position. If this interpretation is true, then the deployment of the attentional focus on the right-most position should in turn produce in patients with neglect an enhanced efficiency of processing for the stimuli presented in that position. Therefore, a patient who is presented in the ipsilesional (i.e. intact) visual field with two stimuli horizontally aligned, should process the stimulus on the right relative position faster, not only in comparison with the other stimulus, but also in comparison with control patients. This is because a neglect patient is assumed to have brought focal attention to the right position, whereas the control patients should have distributed attention between the two possible stimulus positions.
This prediction was confirmed by a recent study by Làdavas, Petronio and Umiltà (1990), who showed that neglect patients were faster than controls to respond to the stimuli in the relative right position. Differences in speed and/or accuracy between the two positions could be due to changes in processing efficiency (d') or to changes in response bias (beta). The results of the previous study showed that the effects are attributable to changes in processing efficiency because the differences in response speed were accompanied by congruent differences in sensitivity, whereas no differences in response bias were found.
The findings of Làdavas et al. (1990) and Làdavas (1990) are compatible only with the hypothesis that in patients with neglect, spatial attention cannot be allocated to multiple regions of the visual field because it is narrowly focused on the most ipsilesional area. As a consequence, the benefits in processing efficiency are confined to this area, whereas in adjacent locations only costs are observable. Considering these findings it is possible to define neglect as the result of hyperattention for certain sectors of space, as opposed to an attentional deficit for specific locations in space, because the latter effect seems to be a consequence of the former.
Interpretation of the Hyperattentional Deficit. The next point that needs to be addressed is why the affected sector of space is not fixed, but is related to the task, and why patients with neglect show hyperattention for the most ipsilateral stimuli. There are three main hypotheses of neglect.
The arousal hypothesis (Heilman, Bowers, Valenstein, & Watson, 1987) maintains that each side of the brain contains its own activating mechanisms and, when one is lesioned, the corresponding hemisphere cannot organise orienting responses towards the contralateral space. This hypothesis cannot explain the first of the points mentioned above. If the deficit is caused by hypoarousal of the right hemisphere, patients should only neglect stimuli shown to the left of fixation, i.e. in the left visual field (LVF). It appears, therefore, that this hypothesis cannot account for the fact that patients may neglect stimuli that are shown in the right visual field (RVF) but are, relatively, on the left (Làdavas, 1987; Rapcsak, Watson, & Heilman, 1987). The authors explained the attentional imbalance within the ipsilesional visual field with reference to the receptive fields of neurons located in cortical structures which appear to be crucially involved in spatial representation (Rapcsak et al., 1987; Rizzolatti, Gentilucci, & Matelli, 1985). They interpreted the right–left attentional asymmetries found in both visual fields in terms of a partial representation of each half of the visual field in attentional areas of both cerebral hemispheres. Rizzolatti et al. (1985), recording from cells in the postarcuate cortex of the monkey, found that of the neurons studied, 29% has exclusively contralateral visual receptive fields, 2% had ipsilateral receptive fields and 69% had bilateral receptive fields. As a consequence, the most lateral part of visual space is represented almost exclusively in one hemisphere, whereas the central part of visual space is represented in both hemispheres. After a lesion in one hemisphere, the capacity to orient to visual stimuli will therefore be affected in the entire visual space, but with a gradient of severity from a maximum in the extreme contralateral hemifield to a minimum in the extreme ipsilateral hemifield.
This hypothesis assumes a fixed relation between the right and left visual fields, as projected through the right and left hemiretinae, and the two hemispheres. It cannot, therefore, account for the attentional bias found in patients with extinction (Làdavas, 1987) and patients with disconnected hemispheres (Làdavas, Del Pesce, Mangun, & Gazzaniga, in press) under the head-tilted condition, because according to the retinal co-ordinates, the two stimuli fall in the upper and lower visual fields. Therefore, we can conclude that this hypothesis (Rapcsak et al., 1987; Rizzolatti et al., 1985) cannot explain the attention imbalance between two separate points on the left–right dimension shown by patients with visual extinction and the two disconnected hemispheres of split-brain patients (see Chapter 3, this volume, for a similar criticism).
According to the representational hypothesis (Bisiach & Vallar, 1988), neglect is caused by a deficit in the ability to form a whole representation of space. This hypothesis can explain why the portion of the space affected by the deficit is not fixed, simply by assuming that the space can be represented in different co-ordinate systems. In contrast, this hypothesis cannot explain why neglect patients outperform controls in detecting stimuli which occupy a relative right position in the intact RVF (Làdavas et al., 1990). It is difficult to conceive of a lesion that produces a better representation in some sectors of space.
In contrast, this finding can easily be interpreted within the framework of the orienting hypothesis by assuming that focal attention is “captured” by stimuli that lie in the right-most location. This observation can be consistent with the directional attentional model proposed by Kinsbourne, although, as we will see, there are some data which are not in line with his model. Kinsbourne (1977; 1987; see also Chapter 3, this volume) has postulated that each hemisphere is responsible for shifting attention in a contraversive direction, either in the ipsilateral or contralateral half of space, e.g. in the whole space. Damage to one hemisphere would unbalance the attentional system in favour of shifts contraversive to the intact side. This orienting bias would direct attention to the most ipsilesional position and, as a consequence, patients with neglect are faster and more accurate than controls to respond to that spatial position.
Kinsbourne’s model implies that the right–left gradient is continuous across the whole field so that costs should increase as a function of the distance of the target from the attended position. This prediction is not supported in Làdavas’s (1990) study, because the results obtained in patients with visual extinction showed a discontinuity of this right–left attentional gradient at the junction of the left and right hemisphere. A larger cost was paid when the stimuli were presented at non-attended locations in the hemifield contralateral to the lesion. This phenomenon is observable mainly under restricted lateralised presentation. In free ocular scanning conditions, most patients show an attentional boundary beyond which sustained attention cannot be further directed leftwards. The spatial position of this boundary, however, is not intrinsically linked to the mid-sagittal plane as in the case of restricted laterialised presentation, but to the overall severity of the deficit (Marshall & Halligan, 1989). It must be noted, however, that in this study, the free scanning paradigm renders difficult to determine the exact position of the vertical retinal meridian at any given moment, which of course shifts in accordance with the eyes.
Considering the discontinuity of this right–left attentional gradient at the junction of the left and right hemispace shown by patients with neglect, it must be admitted that the attentional bias towards the right side cannot be the sole explanation of the neglect phenomenon. Therefore, in the case of neglect patients, an additional deficit is likely to play a role. Perhaps not only is attention captured by the right-most location, but, in addition, contralesional stimuli cannot activate the right hemisphere, as suggested by Heilman et al. (1987). The arousal hypothesis maintains that each side of the brain contains its own activation system and, when one system is lesioned, the corresponding hemisphere cannot process sensory information and organise motor responses. As a consequence, there would be a selective loss of the orienting response to the space contralateral to the lesion.
The Pre-motor Theory of Attention. Another explanation of the discontinuity of this right–left attentional gradient at the junction of the left and right hemispaces is provided by a pre-motor theory of spatial attention (Rizzolatti et al., 1987; Tassinari et al., 1987; see also Chapter 4, this volume). In order to understand how this theory can explain the effect, it is worthwhile introducing an important distinction between overt orienting, which is accompanied by head and eye movements, and purely covert orienting, which can be achieved in the absence of body movements (see, e.g. Posner, 1978; 1980).
As pointed out by Shepherd, Findley and Hockey (1986; also see Umiltà, 1988), the relationship between eye movements and spatial attention can logically manifest itself in three forms. According to the identity hypothesis, the mechanisms involved in the generation of eye movements are identical with those that produce attention shifts. The independence hypothesis maintains that there are two mechanisms, one for eye movements and one for attention shifts, which are not functionally related. A third view, namely the interdependency hypothesis, is that the two mechanisms are neither identical nor completely independent, so that the functioning of one of the two can be facilitated or inhibited by the other.
The identity hypothesis has been disproved by a number of experiments that clearly demonstrated that attention can be directed to different points in space, regardless of eye position (see e.g. Jonides, 1983; Posner, 1978, 1980; Umiltà, 1988). The evidence concerning the independence hypothesis and the interdependence hypothesis is less decisive, even though several attempts have been made to compare them (Klein, 1980; Posner, 1980; Shepherd et al., 1986; Shepherd & Muller, 1989; Remington, 1980; Rizzolatti et al., 1987). All considered, it seems that the evidence available to date favours the interdependence hypothesis. A strong version of this hypothesis was recently proposed by Rizzolatti et al. (1987) and Tassinari et al. (1987), who postulated a strict link between covert orienting of attention and the programming of ocular movements (see also Chapter 5, this volume). The basic idea is that overt orienting and covert orienting are both controlled by the neural mechanisms that are also in charge of saccade programming. Upon the presentation of a stimulus, a motor programme for the saccade is prepared, which specifies the direction and the amplitude of the eye movement. This occurs regardless of whether the saccade is actually executed (i.e. overt orienting) or is not executed (i.e. covert orienting).
It is possible to hypothesise that in patients with a right parietal lobe lesion, the orienting response is under the control of the intact left hemisphere and, therefore, the essential element which characterises the preparation to respond is the specification of the direction “right” as opposed to “left”. When the patient has to respond to RVF stimuli, the task does not require any correction of the directional programme, because the system is already prepared to respond to that direction. Conversely, if the subject has to respond to LVF stimuli, the task requires the correction of directional bias, and as a result the emission of response is retarded or, in the case of patients with severe neglect, omitted. This interpretation implies that reorienting of attention is necessary before a manual response can be emitted to a stimulus that appears in a left unattended position. Posner (1980) has suggested that an arbitrary response, that is one not automatically triggered by a stimulus, only occurs through the commitment of conscious attention. In other words, a stimulus must enter the focus of attention and become conscious before an arbitrary response to it can be emitted. If the stimulus does not enter the focus of attention, the response is omitted, even though the stimulus can influence the patient’s behaviour without the patient being aware of the stimulus (Bisiach & Rusconi, 1990; Marshall & Halligan, 1988; Volpe, Le Doux, & Gazzaniga, 1979).
Therefore, because the pre-motor theory of attention postulates a strict link between covert orienting of attention and the programming of ocular movements, it can explain the discontinuity of the right–left attentional gradient at the junction of the left and right hemispace found in patients with neglect, as well as the finding that in these patients right attentional shifts are very often associated with a gaze displacement to the right-most position (De Renzi, Gentilini, Faglioni, & Barbieri, 1989). De Renzi, Colombo, Faglioni and Gibertoni (1982) pointed out that gaze displacement to the right-most position, triggered by visual stimulation, can be viewed as a milder manifestation of the phenomenon of eye and head deviation towards the side of lesion. According to these authors, the imbalance between the hemispheric turning apparatuses caused by a unilateral lesion is initially independent of stimulation as far as it is present even in soporose patients having their eyes closed and during sleep. Doricchi, Guariglia, Paolucci and Pizzamiglio (1991) showed that in patients with neglect, the leftward rapid eye movements (REMs) were virtually absent, whereas rightward REMs were present. In contrast, when the deficit is of a milder degree of severity, the patient can control and inhibit the compulsory shift of gaze to the right-most end of the structured space and, as a consequence, the eyes are not any longer deviated towards the right. None the less, it is still possible to observe a rightward attentional bias after an RVF stimulation because the oculomotor system is still programmed to respond in that direction.
On the other hand, this uncontrollable gaze deviation towards the rightmost extremity of space is associated with increased time for leftward eye movements and for left shifts of attention. Chedru, Leblanc and Lhermitte (1973) found that marked left unilateral neglect was associated with increased time for leftward eye movements and the degree of asymmetry in eye movement exploration was positively correlated with the degree of unilateral inattention. Moreover, it seems that the deficit can be reduced by asking the patient to turn his or her eyes to the left. Meador et al. (1989) have shown that, when patients with left neglect are requested to imagine a very familiar scene, recall of the items located on the left side of the image is more accurate when they turn their eyes/head to the left.
Within this theoretical framework, other recent data on caloric vestibular stimulation (Cappa, Sterzi, Vallar, & Bisiach, 1987; Vallar et al., 1990) and on the induction of optokinetic nystagmus (Pizzamiglio et al., 1990) fit very well. Either caloric vestibular stimulation or optokinetic nystagmus causes a striking, though temporary, reduction in neglect. Considering that this procedure produces eye deviation towards the neglected side, and that according to the pre-motor theory of attention a motor programme for an eye movement is always accompanied by a corresponding attentional shift, one can well imagine how the eye deviation also renders possible the orienting of attention towards the neglected side.
In conclusion, neglect for the space of stimulation can be regarded as a directional deficit, due to the fact that the component of the motor system which specifies the direction for eye movements or shifts of attention is programmed, by default, to respond only to one direction, e.g. the one ipsilateral to the lesion.
Neglect for Space of Response
In a recent study, Duhamel and Brouchon (1990) stressed the importance of a subtle interplay between afferent (coding of the target) and efferent (coding of the response) processes in the neglect phenomenon. They studied a patient with neglect for the left half of space in a task requiring manual pointing to visual targets located to the left and right visual fields, and found that the difference in speed and accuracy between left and right targets was dependent upon the response condition. They found that the hand contralateral to the lesion was slower than the ipsilateral hand in responding to LVF stimuli and not to RVF stimuli and that the initial spatial position of the responding hand influenced the patient’s responses, e.g. the right hand located on the left side of the space was always associated to longer RTs compared to the condition in which it was located on the right side of space. Similar results were found by Halligan, Manning and Marshall (1991), who showed that the position of the responding hand modulates the expression of left neglect in a line bisection task. These findings indicate that hand position in space is taken into account in response preparation and that a lesion on the parietal lobe causes an impairment for the responses emitted in the space contralateral to the lesion.
Considering these results, it seems that in patients with neglect there are at least two components of response coding which can be related and interfere with the stimulus coding: the responding hand (e.g. the left and right hand) and the absolute position of the hand in space (e.g. the left and right space defined according to the body midline). It is known from studies on spatial compatibility effects that there is at least another component of response coding which is taken into account for the emission of response, e.g. the coding of the relative position of the effectors (Nicoletti, Umiltà, & Làdavas, 1984). Considering the relevance of this component, it is not surprising to find neglect also for those responses that require a left–right coding of the relative position of the effectors.
An experimental task in which this third component of response coding is taken into account for the emission of the response requires patients to respond to visual stimuli with the hand ipsilateral to the lesion (the right hand) located on the ipsilesional side (the right side according to the body midline) and left–right finger discrimination. Due to task demands, it is possible to find neglect attributable to left–right finger discriminations and not to the hand used or to the hemispace where the response is emitted. We tested this hypothesis in a study where patients with neglect were presented with two numbers, 1 and 2, shown in the LVF and RVF, and they were required to respond with the index finger to number 1 and the middle finger to number 2, independent of the visual field (Làdavas & Cimatti, in prep.). The results showed that patients with neglect never responded to LVF stimuli, and that responses to RVF stimuli associated with the index finger, which occupied the left position, were slower and less accurate than responses associated with the middle finger, which occupied the right position.
This finding cannot be explained by motor neglect, because the patients respond with the hand ipsilateral to the lesion, or with directional akinesia, because the response does not imply any overt directional response. The results can be explained by taking into account that in these patients there is a motor predisposition to react to the right. When the patient has to respond with the middle finger, which occupies a right position, the task does not require any correction of the directional programme, because the motor system is already prepared to respond towards the right. Conversely, when the patient has to respond with the index finger, which occupies the left position, the task requires the correction of a directional bias, and as a result the emission of the response is retarded or omitted.
If we accept this interpretation, we can conclude that the motor programme responsible for a directional specification of hand movement along the horizontal dimension is biased towards the right and that this bias becomes manifest in any task which requires left–right discriminations. Taking the results of this study into account, as well as those mentioned above, we can conclude that neglect for the space of response manifests itself in tasks which require specification of the hand to be used to respond, for each hand the specification of the hemispace where the hand operates and, finally, in the case of left–right finger discriminations, the specification of the spatial code describing the positions of the fingers.
The observation that the oculomoter system and the motor programme which guides the hand along the horizontal dimension in patients with neglect are biased towards the right can be explained by findings that showed that ocular and manual responses are integrated within a single spatial frame of reference (Fisk & Goodale, 1985; Mather & Fisk, 1985). Additional evidence has been provided by Fries, Swihart and Danek (1989). They reported a patient with a mild left hemianopia and visuo-spatial neglect who could localise LVF stimuli with great difficulty by a series of small (“stair-case”) saccades. However, localising eye movements were normal if the visual stimulus had previously been pointed to by the right hand.
Vertical Dimension
The horizontal dimension is not the only one affected by neglect, because left neglect patients can also show an attentional deficit along the vertical dimension. Other authors have already pointed out the existence of altitudinal neglect (Bender & Teuber, 1948; Rapcsak, Cimino, & Heilman, 1988) and that this deficit can be more pronounced in the lower quadrant of the left hemispace (Halligan & Marshall, 1989). Moreover, Halligan and Marshall (1989) have recently shown that this phenomenon is more evident in the contralateral than in the ipsilateral visual field.
In a recent study, Làdavas and Carletti (in prep.) have tested the hypothesis that neglect for the lower part of the space can appear independently of the neglect field, i.e. also along the vertical meridian. The experimental condition was the same as that used in the experiment designed to test neglect for the space of response, with the exception that the two stimuli (1 or 2) were presented above and below fixation. The results showed that patients were much slower and less accurate to respond to the stimuli presented below than above the fixation mark. The deficit for the lower part of the visual field was much more severe when the bottom stimulus required a left response than when it required a right response, indicating an interaction between neglect for the lower part of the visual field and neglect for the space of response. The observed deficit cannot be sensory in nature, because all of the patients had the lower quadrant of the visual field intact, as assessed by campimetry. Moreover, the magnitude of the deficit was reduced when the patient’s attention was cued to the lower visual field. In other words, when Posner’s paradigm was used to cue attention to the lower visual field, the difference between up and down was reduced in the same way as the left-side deficit is reduced when the LVF is cued by a central arrow (Làdavas, Menghini, & Umiltà, 1991; Posner, Walker, Friedrich, & Rafal, 1987).
Therefore, the deficit observed along the vertical dimension seems to be attentional in nature, although, at this stage, we cannot say whether the pre-motor theory of attention can accommodate these results. The patients, in fact, did not show, as the theory predicts, any partial or total incapacity to move the eyes or the head vertically. Rizzolatti et al. (1987) proposed the pre-motor theory of attention on the basis of the findings that motor deficits are associated with attentional deficits and the latter are congruent with the former. If the closed relationship between eye movement deficits and attentional shift deficits found for the horizontal dimension in patients with left neglect can be explained by the pre-motor theory of attention, this theory cannot yet account for the neglect found for the lower part of the visual field.
In conclusion, considering the fact that objects have a three-dimensional representation and that attention can be oriented in depth as well as in the horizontal and vertical dimensions (Gawryszewski et al., 1987), it is not surprising to find neglect for each of these spatial dimensions. As far as the depth dimension is concerned, it is worth mentioning some studies (Bisiach, Perani, Vallar, & Berti, 1986; Halligan & Marshall, 1991) that described patients showing personal neglect without having extrapersonal neglect and vice versa, which can be interpreted as a dissociation between the capacity to shift attention in the far and near space (or personal and peripersonal space).
Is the Deficit Related to Automatic or Voluntary Orienting of Attention?
An important distinction that emerges from the study of normal subjects is that between voluntary and automatic orienting (Muller & Findley, 1988; Muller & Rabbitt, 1989; Spencer, Lambert, & Hockey, 1988; Yantis & Jonides, 1984). It is known that attention can be directed to the position of an impending target by the use of centrally located cues or peripherally located cues. In the first case, a conventional signal shown at or around fixation indicates the position to which attention must be directed. In the second case, a peripheral marker, in the form of a salient discontinuity in a non-foveal area, is shown near the location to which the stimulus will be presented.
Based on criteria such as capacity demands, resistance to suppression and sensitivity to expectancy, peripheral cues are shown to cause automatic shifts of attention, whereas central cues start voluntary shifts. This has led to the notion that automatic and voluntary orienting can be achieved by two separate mechanisms (Muller & Findley, 1987; Muller & Rabbitt, 1989). In contrast, Posner (1980) and Jonides (1983) assume that there is one mechanism only, which can be triggered automatically by peripheral cues or initiated voluntarily by central cues. In their view, the two modes of orienting are thought to differ only in the ways attention shifts are initiated, rather than in the mechanisms that guide them.
If we consider neglect as an attentional deficit, it can be reasoned that if a single-mechanism subserves automatic and voluntary orienting of attention, neglect patients should be impaired in responding to LVF stimuli both in the peripheral and central cue conditions. In contrast, if the dual-mechanism hypothesis holds true, we should expect the attentional deficit to be associated more with one type of orienting than with the other.
Automatic and voluntary shifts of attention in patients with neglect were tested in two experiments (Làdavas, 1992). In the automatic orienting experiment, an arrow head in the periphery was chosen as the cue that automatically draws attention. In the voluntary orienting experiment, an arrow head at fixation was chosen as the cue that could cause a voluntary shift of attention. In the first experiment, the arrow did not predict the location of the stimulus, whereas in the second experiment the cue was highly predictive of the location of the stimulus. This is because it has been shown that cue validity, i.e. the probability with which the cue predicts the target location, affects the voluntary orienting of attention. There were four possible stimulus locations which were arranged either horizontally or vertically, above, below, to the right or left of a fixation point. The instructions were to respond manually as fast as possible to a visual stimulus, regardless of whether it occurred in a cued or in a non-cued location.
The results showed that when the cue appeared in the same location of the stimulus, patients with neglect never responded to LVF stimuli under the automatic orienting condition, whereas they were much more accurate to respond under the voluntary attention condition. Such results indicate that any deficits in the automatic orienting component of attention can be modulated by voluntary attentional processes (see Chapters 5 and 7, this volume, for a similar view). Therefore, it seems that in patients with neglect, voluntary shifts of attention are preserved and this is probably due to the fact that this function is subserved by the frontal lobe, which in these patients was intact.
If it is true that the frontal lobe mediates voluntary shifts of attention, then opposite results to those obtained in patients with a parietal lesion would be expected in patients with frontal lobe lesions. A recent study by Làdavas, Delia Sala, Cimatti and Trivelli (1991) showed that patients with frontal dorsolateral lesions were impaired in the voluntary orienting of attention, whereas their automatic orienting was fairly intact. These patients very rarely responded to stimuli presented in the peripheral field when the stimulus was preceded by a central cue, whereas they always detected the stimulus presented in the peripheral field when it was preceded by a peripheral cue. They also made more false alarms than patients with parietal damage, and this pattern of results appeared only in the peripheral cue condition. The inability shown by patients with a frontal syndrome to voluntary attend a specific spatial position and to suppress the tendency to respond to the peripheral cues, allowed the authors to hypothesise that the frontal lobe mediates the voluntary orienting component of attention. The same anatomical dissociation has been proposed for reflex-like orienting eye movements and volitional eye movements (Guitton, Buchtel, & Douglas, 1985; Doricchi et al., 1991).
These findings seem to support the dual-mechanism hypothesis, which maintains that automatic and voluntary orienting are subserved by separate mechanisms located in different parts of the brain. More specifically, it would seem that the frontal lobe mediates voluntary orienting of attention, whereas the parietal lobe modulates automatic orienting of attention. In particular, it appears that the right parietal lobe mediates automatic orienting towards the left and the left parietal lobe mediates automatic orienting towards the right. Damage to the right hemisphere would unbalance the attentional system in favour of automatic rightward shifts, which can be overcome by a rehabilitation procedure aimed at training the patients to voluntary direct attention to stimuli shown in the neglected contralesional field (Làdavas, Menghini, & Umiltà, in press a & b).
References
Bender, M.B. & Teuber, H.L. (1948). Spatial organization of visual perception following injury to the brain. Archives of Neurology and Psychiatry, 59. 39–62.
Bisiach, E., Perani, D., Vallar, G., & Berti, A. (1986). Unilateral neglect: Personal and extra-personal. Neuropsychologia, 24, 759–767.
Bisiach, E. & Rusconi, M.L. (1990). Breakdown of perceptual awareness in unilateral neglect. Cortex, 26, 1–7.
Bisiach, E. & Vallar, G. (1988). Hemineglect in humans. In F. Boller & J. Grafman (Eds), Handbook of Neuropsychology, pp. 195–222. Amsterdam: Elsevier.
Cappa, S., Sterzi, R., Vallar, G., & Bisiach, E. (1987). Remission of hemineglect and anosognosia during vestibular stimulation. Neuropsychologia, 25, 775–782.
Castiello, U. & Umiltà, C. (1990). Size of the attentional focus and efficiency of processing. Ada Psychologies 73, 195–209.
Chedru, F., Leblanc, M., & Lhermitte, F. (1973). Visual searching in normal and braindamaged subjects: Contribution to the study of unilateral inattention. Cortex, 9, 94–111.
De Renzi, E., Colombo, A., Faglioni, P., & Gibertoni, M. (1982). Conjugate gaze paresis in stroke patients with unilateral damage. Archives of Neurology, 39, 482–486.
De Renzi, E., Gentilini. M., Faglioni, P., & Barbieri, C. (1989). Attentional shift towards the rightmost stimuli in patients with left visual neglect. Cortex, 25, 231–237.
Doricchi, F., Guariglia, C., Paolucci, S., & Pizzamiglio, L. (1991). Disappearance of leftward rapid eye movements during sleep in the left visual hemi-inattention. Cognitive Neuroscience and Neuropsychology, 2. 285–288.
Driver, J. & Halligan, P.W. (1991). Can visual neglect operate in object-centered coordinates? An affirmative single-case study. Cognitive Neuropsychology, 6, 475–496.
Duhamel, J.R., & Brouchon, M. (1990). Sensory aspects of unilateral neglect: A single case analysis. Cognitive Neuropsychology, 7, 57–74.
Fisk, J.D. & Goodale, M.A. (1985). The organization of eye and limb movements during unrestricted reaching to targets in contralateral and ipsilateral visual space. Experimental Brain Research, 60. 159–178.
Fries. W., Swihart, A.A., & Danek, A. (1989). Somatosensory substitution of spatial information improves oculomotor performance in visual neglect. In Brain damage and rehabilitation: A neuropsychological approach. Munich: Edition Wissenschaft Kyrill und Method.
Gainotti, G., D’Erme, P., & De Bonis, C. (1989). Aspects cliniques et mecanismes de la négligence visuo-spatiale. Revue Neurologique, 145, 626–634.
Gawryszewski, L.D.G., Riggio, L. Rizzolatti, G., & Umiltà. C. (1987). Movements of attention in three spatial dimensions and the meaning of “neutral” cues. Neuropsychologia, 25, 19–29.
Guitton, D., Buchtel, H.A., & Douglas, M. (1985). Frontal lobe lesions in man cause difficulties in suppressing reflexive glances and in generating goal-directed saccades. Experimental Brain Research, 58, 455–472.
Halligan, P.W., Manning, L. & Marshall, J.C. (1991). Hemispheric activation vs spatio-motor cueing in visual neglect: A case study. Neuropsychologia, 29, 165–176.
Halligan, P.W. & Marshall, J.C. (1989). Is neglect (only) lateral? A quadrant analysis of line cancellation. Journal of Clinical and Experimental Neuropsychology, 6, 793–798.
Halligan, P.W. & Marshall, J.C. (1991). Left neglect for near but not far space in man. Nature, 350, 498–500.
Heilman, K.M., Bowers, D., Coslett, H.B., Whelan. H.. & Watson. R.T. (1985). Directional hypokinesia. Neurology, 35, 855–859.
Heilman, K.M., Bowers, D., Valenstein, E., & Watson, R.T. (1987). Hemispace and hemispatial neglect. In M. Jeannerod (Ed.), Neurophysiological and neuropsychological aspects of spatial neglect, pp. 115–150. Amsterdam, North-Holland.
Jeannerod, M. (Ed.) (1987). Neurophysiological and neuropsychological aspects of spatial neglect. Amsterdam: North-Holland.
Jonides, J. (1983). Further toward a model of the mind’s eye’s movement. Bulletin of the Psychonomic Society, 21, 247–250.
Kinsbourne, M. (1977). Hemineglect and hemisphere rivalry. In E.A. Weinstein & R.P. Friedland (Eds), Hemi-inattention and hemisphere specialization, pp. 4l^t9. New York: Raven Press.
Kinsbourne, M. (1987). Mechanisms of unilateral neglect. In M. Jeannerod (Ed.), Neurophysiological and neuropsychological aspects of spatial neglect, pp. 69–86. Amsterdam: North-Holland.
Klein, R. (1980). Does oculomotor readiness mediate cognitive control of visual attention? In R.S. Nickerson (Ed.), Attention and performance VIII. pp. 259–276. Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Làdavas. E. (1987). Is the hemispatial deficit produced by right parietal lobe damage associated with retinal or gravitational coordinates? Brain, 110, 167–180.
Làdavas, E. (1990). Selective spatial attention in patients with visual extinction. Brain, 113, 1527–1538.
Làdavas, E. (1992). Automatic and voluntary orienting of attention: A dissociation between patients with frontal ami parietal lesions. Paper presented at TENNET 111, Montreal, 20–22 May.
Làdavas, E. & Carletti, M. (in prep.). Attentional orienting deficit in patients with visual neglect.
Làdavas, E. & Cimatti (in prep.). Directional motor aspects of visual neglect.
Làdavas, E., Delia Sala, S., Simatti, D., & Trivelli, C. (1991). Selective visual attention in patients with frontal lobe lesions. Paper presented at the Third IBRO World Congress of Neuroscience, Montreal, 4–9 August.
Làdavas, E., Del Pesce, M., Mangun, R., & Gazzaniga, M. (in press). Variations in attentional bias in the two disconnected cerebral hemispheres. Cognitive Neuropsychohgy.
Làdavas, E., Menghini, G., & Umiltà, C. (in press a). On the rehabilitation of hemispatial neglect. In M.J. Riddoch & G.W. Humphreys (Eds), Cognitive neuropsychologv and cognitive rehabilitation. Hove: Lawrence Erlbaum Associates Ltd.
Làdavas, E., Menghini, G., & Umiltà, C. (in press b). A rehabilitation study of hemispatial neglect. Cognitive Neuropsychohgy.
Làdavas, E. & Moscovitch, M. (1984). Must egocentric and environmental frames of reference be aligned to produce spatial S-R compatibility effects? Journal of Experimental Psychology-Human Perception and Performance, 10, 205–215.
Làdavas. E., Petronio. A., & Umiltà. C. (1990). The deployment of visual attention in the intact field of hemineglect patients. Cortex, 26, 307–317.
Marshall, J.C. & Halligan, P.W. (1988). Blindsight and insight in visuo-spatial neglect. Nature, 336, 766–777.
Marshall. J.C. & Halligan, P.W. (1989). Does the midsagittal plane play any privileged role in “left” neglect? Cognitive Neuropsychohgy, 6, 403–422.
Mather, J.A. & Fisk, J.D. (1985). Orienting to targets by looking and pointing: Parallels and interaction in ocular and manual performance. Quarterly Journal of Experimental Psychology, 37A, 315–338.
Meador, K.J., Loring, D.W., Lee, G.P., Brooks. B.S., Nichols, F.T., Thompson, E.E., Thompson, W.O., & Heilman. K.M. (1989). Hemisphere asymmetry for eye gaze mechanisms. Brain, 112, 103–111.
Muller, H.J. & Findley, M. (1987). Sensitivity and criterion effects in the spatial cuing of visual attention. Perception and Psychophvsics, 42, 383–399.
Muller, H.J. & Findley, J.M. (1988). The effect of visual attention on peripheral discrimination threshold in single and multiple element displays. Acta Psychologica, 69, 129 155.
Muller, H.J. & Rabbitt, P.M. (1989). Reflexive and voluntary orienting of visual attention: Time course of activation and resistance to interruption. Journal of Experimental Psychology/ Human Perception and Performance, 15, 315–330.
Nicoletti, R., Umiltà, C, & Làdavas, E. (1984). Compatibility due to the coding of the relative position of the effectors. Acta Psychologica. 57, 133–143.
Pizzamiglio, L., Frasca, R., Guariglia, C., Incaccia, R., & Antonucci, G. (1990). Effect of optokinetic stimulation in patients with visual neglect. Cortex, 26, 535–540.
Posner, M.I. (1978). Chronometric exploration of mind. Hillsdale, NJ: Lawrence Erlbaum Associations Inc.
Posner, M.I. (1980). Orienting of attention. Quarterly Journal of Experimental Psychology, 32. 3–25.
Posner, M.L. Walker. J.A., Friedrich. F.A.. & Rafal. R.D. (1987). How do the parietal lobes direct covert attention? Neuropsychologia, 25. 135 145.
Rapcsak, S.Z., Watson, R.T., & Heilman. K.M. (1987). Hemispace- visual field interactions in visual extinction. Journal of Neurology, Neurosurgery anil Psychiatry. 50, 1117–1124.
Rapcsak, S.Z., Cimino, C.R., & Heilman. K.M. (1988). Altitudinal neglect. Neurology. 38. 277–281.
Remington. R.W. (1980). Attention and saccadic eye movements. Journal of Experimental Psychology. 6, 726–744
Rizzolatti, G. & Camarda, R. (1987). Neural circuits for spatial attention and unilateral neglect. In M. Jeannerod (Ed.), Neurophysiological and neuropsychological aspects of spatial neglect. Amsterdam: North-Holland.
Rizzolatti, G., Gentilucci, M., & Matelli. M. (1985). Selective spatial attention: One center, one circuit or many circuits? In M.I. Posner & O.S.M. Marin (Eds), Attention and performance IX. Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Rizzolatti, G., Riggio, L., Dascola, I, & Umiltà, C. (1987). Reorienting attention across the horizontal and vertical meridians: Evidence in favor of a premotor theory of attention. Neuropsychologia. 25. 31 -40.
Shepherd, M. & Muller. H.J. (1989). Movement versus focusing of attention. Perception and Psychophysics. 46. 146–154.
Shepherd, M., Findley, J.M., & Hockey, R. J. (1986). The relationship between eye movements and spatial attention. Quarterly Journal of Experimental Psychology. 38 A. 475 491.
Spencer. M.B.H., Lambert. A.J.. & Hockey. R. (1988). The inhibitory component of orienting, alertness and sustained attention. Acta Psychologica 69. 165–184.
Tassinari, G., Alioti. S., Chelazzi. L., Marzi, C, & Berlucchi. G. (1987). Distribution in the visual field of the costs of voluntary allocated attention and of the inhibitory after-effects of covert orienting. Neuropsychologia. 25. 55 71.
Umiltà, C. (1988). Orienting of attention. In F. Boller & J. Grafman (Eds), Handbook of neuropsychology. Vol. 1, pp. 175–193. Amsterdam: Elsevier.
Umiltà, C. & Liotti, M. (1987). Egocentric and relative spatial codes in S-R compatibility. Psychological Research, 49, 81–90.
Umiltà, C. & Nicoletti, R. (1990). Spatial stimulus response compatbility. In R.W. Proctor & T.G. Reeve (Eds), Stimulus response compatibility: An integrated perspective. Amsterdam: North-Holland.
Vallar, G. & Perani, D. (1987). The anatomy of spatial neglect in humans. In M. Jeannerod (Ed.), Neurophysiological and neuropsychological aspects of spatial neglect, pp. 235 258. Amsterdam: North-Holland.
Vallar, G., Sterzi, R., Bottini, G., Cappa. S., & Rusconi. L. (1990). Temporary remission of left hemianesthesia after vestibular stimulation. A sensory neglect phenomenon. Cortex. 26. 123–131.
Volpe, B.T., Le Doux, J.E., & Gazzaniga, M.S. (1979). Information processing of visual stimuli in an “extinguished” field. Nature. 282. 722–724.
Yantis. S. & Jonides. J. (1984). Abrupt visual onsets and selective attention: Evidence from visual search. Journal of Experimental Psychology. 10, 601–620.