Introduction—Horizontal and Radial Spatial Neglect

In this issue of the Journal of the International Neuropsychology Society (JINS), Wilkinson and Sakel (Reference Wilkinson and Sakel2011) report a left-handed man with bilateral hemispheric lesions, larger in the left than right hemisphere and involving the frontal lobes, who demonstrates neglect when performing both horizontal (parallel to the chest) and radial (perpendicular to the chest) line bisections. Whereas prior studies have reported that spatial neglect can take place in multiple ego- or body-centered Cartesian coordinates, including horizontal and radial as well as vertical (Rapcsak, Cimino, & Heilman, Reference Rapcsak, Cimino and Heilman1988; Shelton, Bowers, & Heilman, Reference Shelton, Bowers and Heilman1990), these authors report that the direction of horizontal spatial bias can be influenced by the type of line stimuli used. This case compares the influence of lines composed of characters (including alphanumeric characters and geometric shapes) versus solid lines, as well as the presence or the absence of distracting stimuli. The distracters are short lines that were positioned above and below, as well as to the right and left of the line that was to be bisected. With lines oriented in the horizontal coordinate, this patient deviated to the right when attempting to bisect solid lines and to the left when attempting to bisect character lines as well as lines presented with distracting stimuli.

Focal Versus Global Attention

When performing the character or symbol line bisection task, subjects have to focus their attention on these symbols (focal attention). They also have to focus their attention in the condition with distracters, but in the solid line bisection the subject takes a more global view of the line. The important observation made by these investigators is that, just as different attentional demands (e.g., focal vs. global) can alter neglect in the horizontal coordinate (leftward deviation vs. rightward deviation), these same factors can also produce changes in the radial coordinate. In fact, these changes appear to be parallel to those in the horizontal coordinate, such that when performing the solid line bisection task in the radial coordinate, the patient demonstrated a distal (away from the body) bias. In contrast, when performing the character lines and distracter radial bisection tasks, this patient had a proximal (toward the body) bias.

Influence of selective hemispheric activation on the allocation of spatial attention

The authors stated that they could not “determine which particular stimulus property (e.g., number of perceptual items, physical dimensions) reversed the direction of bisection error.” Kosslyn and his coworkers (Reference Kosslyn, Koenig, Barrett, Cave, Tang and Gabrieli1989) reported that normally in right-handed people, the right hemisphere is likely to mediate continuous relationships and the left hemisphere appears to be dominant for mediating categorical relationships. Robinson et al. (1988) demonstrated that, whereas the left hemisphere appears to be dominant for mediating focal attention, the right appears to mediate global attention. Whereas in many right-handed people the right hemisphere has a greater capacity than the left hemisphere to attend bilaterally (Heilman & Van Den Abell, Reference Heilman and Van Den Abell1980; Pardo, Fox, & Raichle, Reference Pardo, Fox and Raichle1991), in general, each hemisphere attends primarily to contralateral body-head-eye hemispace. In addition, it has been demonstrated that, when patients with hemispatial neglect perform a task that is primarily mediated by one hemisphere, the task will activate that hemisphere and with this activation the subject may increase their sensory attention or action-intention toward the hemispace that is contralateral to this activated hemisphere. For example, Heilman and Watson (Reference Heilman and Watson1978) demonstrated that patients with left hemispatial neglect had more severe neglect with verbal (mediated by the left hemisphere) than a spatial cancellation task (mediated by the right hemisphere). Because in right-handed people, the right hemisphere is more likely to mediate global attention and perceive continuous stimuli and the left hemisphere being dominant for mediating focal attention and perceiving categorical stimuli, when performing the solid line bisection task, there should be greater activation of the right than left hemisphere, and when performing the character line bisection, there should be more activation of the left hemisphere.

Left handedness and reversed global–local hemispheric dominance

Support for this global-continuous versus focal-categorical attention spatial bias postulate comes from the work of Lee et al. (Reference Lee, Kang, Park, Son, Lee, Adair and Na2004) who found that patients were more likely to deviate to the right when performing the character line bisection task than the solid line bisection task, and this rightward deviation was attributed to activation of these patients’ left hemisphere. The results of Lee et al. are, however, opposite to that of this Neurobehavioral Grand Rounds report (Wilkinson and Sakel, Reference Wilkinson and Sakel2011). When this patient attempted to bisect solid lines, he should have been more likely to deviate to the left and to the right with character lines and distracting stimuli. This right–left reversal of deviation demonstrated by this patient (vs. prior reports) may be related to differences in handedness and hemispheric specialization. Whereas the participants in the study by Lee et al. were right-handed, the patient reported by Wilkinson and Sakel (Reference Wilkinson and Sakel2011) was left-handed. Thus, unlike right-handed people, it is possible that this patient's right hemisphere was important in mediating focal attention and his left hemisphere global attention. Therefore, when performing a horizontal solid line bisection task that requires more global than focal attention, he might have activated his left hemisphere more than his right and, thus, deviated to the right. In contrast with the character or distractor bisections, which require more focused attention, this left-handed man may have activated his right hemisphere more than his left, which would cause a leftward deviation.

A second important observation made by Wilkinson and Sakel (Reference Wilkinson and Sakel2011) is that these different types of line bisections (e.g., solid vs. character) also altered performance on a radial line bisection task. Unlike the horizontal line bisection, these proximal–distal deviations cannot be explained by the postulate that the solid line task activates the left hemisphere and the character line task as well as the distracter task activate the right hemisphere and this activation induces a lateral bias in which attention is shifted to the space that is contralateral to each activated hemisphere. Heilman, Chatterjee, and Doty (Reference Heilman, Chatterjee and Doty1995), however, demonstrated in right-handed normal subjects that, whereas the left hemisphere appears to have a proximal peripersonal spatial bias, the right has a distal bias. Since the left-handed patient described by Wilkinson and Sakel appears to have a reversal of his left–right hemispheric attentional asymmetries, this patient's left hemisphere, unlike the right-handed people tested by Heilman et al. (Reference Heilman, Chatterjee and Doty1995), may have a distal radial bias and his right hemisphere a proximal radial bias. If this idea is correct, as mentioned above, his attempted bisections with solid lines should activate his left hemisphere. This hypothesis would predict deviations to the right for the horizontal line bisections and a distal bias for the radial solid line bisections, which is what these investigators reported. In contrast, when attempting to bisect character lines or lines with distracting stimuli, he should have primarily activated his right hemisphere, and this asymmetrical activation should have caused a proximal deviation in his attempted bisections of radial oriented character lines, which is also what he demonstrated. When performing radial line bisections, however, the participant's head and eyes are often in a downward position; thus, while the line is in the radial coordinate with respect to the subject's trunk, this line in respect to the eyes and head is at least partially vertical. Therefore, it is unclear if the radial biases reported by these authors are purely radial or if there is also a vertical component.

Influence of lesion loci

The patient reported in the Wilkinson and Sakel (Reference Wilkinson and Sakel2011) in the Neurobehavioral Grand Rounds had bilateral lesions. Weintraub, Daffner, Ahern, Price, and Mesulam (Reference Weintraub, Daffner, Ahern, Price and Mesulam1996) reported that visual hemispatial inattention occurred with greater frequency and severity in patients with bilateral lesions than in patients with unilateral left-sided or right-sided lesions. This increase in neglect with bilateral lesions, as observed in Wilkinson and Sakel's patient may be related to a reduced capacity to avoid spatial attentional deviations when performing hemispheric mediated computations.

There is, however, another possible mechanism. This patient's bilateral lesions appear to involve both the right and left hemisphere's frontal–subcortical networks, but the right hemisphere lesion appears to be smaller than the left and restricted to the frontal–subcortical system. The left hemisphere lesion may have also injured the connections between the frontal and temporoparietal lobes or even involved the temporoparietal attentional network. Thus, the asymmetric performance with focal character lines versus global straight lines may be related to the asymmetry of these lesions. As mentioned by these authors, patients with frontal injuries can demonstrate an “attention grasp” (Butter, Rapcsak, Watson, & Heilman, Reference Butter, Rapcsak, Watson and Heilman1988; Kwon and Heilman, Reference Kwon and Heilman1991), and this attention grasp may in part be related to disinhibition of the parietal lobe that mediates approach behaviors. Thus, when the task required focal-categorical attention, this patient may have activated his right hemisphere's parietal lobe, which appears to be intact; but because of the larger left hemisphere lesion, this patient was impaired at activating the left parietal lobe, and this asymmetry induced a leftward attentional grasp with horizontal character lines as well as proximal attentional grasp with radial character lines. This lesion model alone, however, cannot explain why with solid lines he deviated to the right when the line was horizontal and a distal bias when the line was radial.

Future studies

To further learn about attentional spatial biases in right-handed people, future research should examine horizontal, radial, and vertical character and solid line bisections in right-handed patients. Perhaps these patients, unlike the patient described by Wilkinson and Sakel, would deviate to the left when bisecting solid horizontal lines and to the right when bisecting character lines. With radially oriented solid lines, they should deviate distally and with vertical solid lines upward; but with radial character lines, they should deviate proximally and with vertical character lines downward. Future investigation may also want to dissociate the capacity versus the grasp hypotheses and learn if these spatial biases are sensory-attentional, action-intentional, or both. This information may be important when developing rehabilitation strategies and when providing guidance to patients and caregivers about the dangers of these spatial misperceptions and actions.