Research ReportDisturbed line bisection is associated with posterior brain lesions
Introduction
Patients suffering from spatial neglect show profound deficits in everyday tasks, spontaneously gazing predominantly to the ipsilesional side, only eating food from the ipsilesional side of their plate and ignoring people located on their contralesional side. This syndrome is of great theoretical interest, as it can help us describe how the intact brain completes spatial tasks. In addition, spatial neglect is of clinical interest, as the presence of neglect is a predictor of impaired long-term functional recovery (for review Karnath and Zihl, 2003). Thus, it is unsurprising that neglect is a heavily researched topic. However, the literature investigating this syndrome has led to a large number of disagreements regarding its nature and anatomical basis. Indeed, some have even despaired that neglect may be a meaningless entity (Halligan and Marshall, 1992).
This skepticism is exemplified by the observation that the two most popular tasks traditionally used to diagnose the syndrome, the line bisection and the cancellation tasks, appear to dissociate from each other. In the line bisection task (Heilman and Valenstein, 1979), the participant is asked to mark the midpoint of a horizontal line. The traditional view is that patients with neglect bisect lines ipsilesionally to the true midpoint (though this pattern reverses with short lines (Halligan and Marshall, 1988)). A second popular task for identifying and quantifying neglect is the cancellation task, where patients are asked to mark a large number of target stimuli in an array of distractors (Gaunthier et al., 1989, Weintraub and Mesulam, 1985). Patients with spatial neglect tend to ignore contralesionally located targets. As both the line bisection and the cancellation tasks are typically used to assess spatial neglect, one would expect performance on these two tests would show a strong correlation with each other. However, there are now many studies that reported apparent dissociations between these two tasks (Binder et al., 1992, Ferber and Karnath, 2001a, Halligan and Marshall, 1992, Halligan et al., 1990, Marshall and Halligan, 1995, McGlinchey-Berroth et al., 1996). Of course, each behavioral test is a noisy measure of underlying deficits, so one might expect occasional dissociations for single subjects. Nevertheless, a series of group studies (Binder et al., 1992, Ferber and Karnath, 2001a, McGlinchey-Berroth et al., 1996) clearly illustrate that these two tasks dissociate. Of particular interest, a recent group study by Ferber and Karnath (2001a) compared how well each of these tasks correlated with the typical clinical behavior associated with spatial neglect, that is, a tendency to spontaneously gaze toward the ipsilesional side, orienting ipsilesionally when addressed from the contralesional side, ignoring people or objects located on the contralesional side, impaired clock drawing, and impaired picture copying. Ferber and Karnath found that performance in the cancellation tasks (especially in the letter cancellation task (Weintraub and Mesulam, 1985) and the bells test (Gaunthier et al., 1989)) corresponded well with the clinical behavior of the patients. On the other hand, their scores in the line bisection task showed very poor correlation with neglect. In fact, 40% of the patients who showed the typical clinical signs for spatial neglect (see above) were unimpaired in the line bisection task. In other words, the performance in line bisecting does not reflect the characteristic behavioral disorders observed in patients with spatial neglect following a right hemisphere stroke.
This observation should not be interpreted as suggesting that poor performance on the line bisection task is not clinically important in its own right: it suggests a profound spatial deficit. One interpretation of this finding is that line bisection is simply a less sensitive task for spatial neglect. On the other hand, line bisection and cancellation make different cognitive demands. It is thus possible that they measure different aspects of performance. In any case, it is worth bearing in mind that performance in the popular line bisection task does not seem to be particularly sensitive to spatial neglect, and may indeed be measuring a very different system than the circuits tapped by the cancellation task.
In this study, we try to determine whether the dissociation between clinical neglect behavior and the behavior in line bisecting might be explained by a different pattern of brain damage. If each of these tasks were linked with a distinct neural substrate, we would gain insight into the different functional roles of the implicated regions.
Indeed, there is a precedent to expect that different types of brain damage might impair performance on these two tasks. Binder et al. (1992) examined 34 patients with right hemisphere stroke. Importantly, they found no significant correlation between performance on the line bisection and cancellation tasks—while some patients showed deficits on each task, performance on one task did not predict performance on the other task. Interestingly, when Binder et al. performed overlay plots of the patients' brain lesions, they observed that neglect patients who exhibited abnormal line bisection tended to have posterior lesions. In contrast, patients who were only impaired on the cancellation task suffered more anterior damage. Further support for this dissociation comes from studies that show that size perception errors for visual objects (lines, rectangles) are correlated with neglect patients who exhibit posterior damage and visual field defects (Barton and Black, 1998, Daini et al., 2002, Doricchi and Angelelli, 1999, Ferber and Karnath, 2001b).
This finding has important implications for the current dispute regarding the neuroanatomical basis for spatial neglect. Traditionally, damage to the right inferior parietal lobule (IPL) and temporo-parietal junction (TPJ) has been considered the best predictor for spatial neglect (Heilman et al., 1983, Mort et al., 2003, Vallar and Perani, 1986). However, four recent anatomical studies have found the center of lesion overlap in the right superior temporal cortex and insula, suggesting that these structures rather than the inferior parietal lobule are the most frequent cortical substrates for spatial neglect in humans (Karnath et al., 2001, Karnath et al., 2003, Karnath et al., 2004a, Karnath et al., 2004b).
Two of these studies intentionally excluded patients with separate neurological disorders that can co-occur with neglect, specifically (i) patients with visual field cuts (Karnath et al., 2001) and (ii) patients with visual field cuts or extinction (Karnath et al., 2003), aiming to isolate the anatomical regions involved with the core deficit of spatial neglect (Fig. 1). However, the exclusion of patients may be a problematic experimental strategy that may lead to inadvertent selection biases (Husain and Rorden, 2003). Two further studies thus were based on unselected samples comparing consecutively admitted neglect and control patients without excluding subjects for any additional symptoms (Fig. 2). One of these latter studies employed a technique where the location of the lesion was drawn directly on the patient's own magnetic resonance imaging (MRI) scan using SPM normalization and cost-function masking for subsequent transformation into stereotaxic space (Karnath et al., 2004b). The other study used a very large sample size of 140 brain-damaged subjects and compared neglect and control patients by voxelwise statistical testing (Karnath et al., 2004a). The idea of voxelwise testing is to compute an independent statistical test between two groups of subjects for each and every voxel of the brain. Moreover, logistic regression was used to account for the differences in lesion volume between the two patient populations.
In all four studies that implicate the critical brain regions near the superior temporal cortex and insula (Karnath et al., 2001, Karnath et al., 2003, Karnath et al., 2004a, Karnath et al., 2004b), patients were selected based on two cancellation tasks and a number of clinical measures, but were explicitly not selected based on their performance on the line bisection task. This selection method differs from a number of anatomical studies of neglect (for example, the most recent one of Mort et al. (2003)) as well as numerous single case studies, where poor performance on line bisection tasks was used to identify ‘neglect’ patients. Based on the work of Binder et al. (1992), we would expect that studies that identify neglect based on line bisection errors would tend to report more posterior lesions compared to studies where performance on this task is not considered. This conclusion appears to reconcile at least some of the different findings regarding the anatomy of spatial neglect. However, there is some evidence that suggests that Binder et al.'s findings should be treated with some caution. In particular, a study by McGlinchey-Berroth et al. (1996) replicated Binder et al.'s behavioral result (showing that line bisection and cancellation do not correlate well with each other) but failed to find anatomical differences between these groups. While the anatomical analysis in this study was not as sophisticated as the procedure used by Binder's group, it does suggest that these findings should be viewed with caution. In any case, given the important theoretical importance of Binder's et al.'s findings, a replication would certainly add weight to the claims of Binder et al.
In the present study, we conducted an anatomical analysis of the patients reported by Ferber and Karnath (2001a). We were interested to see if we could replicate the finding of Binder et al. (1992) who showed that poor performance on line bisection and cancellation tasks are associated with distinct regions of damage. In addition, we were interested to observe whether a larger group might give us a clearer picture of the anatomy involved. This is of interest as in Binder et al.'s study the areas most strongly associated with poor performance in the line bisection task appeared more closely associated with those posterior brain regions that have traditionally been associated with spatial neglect (Heilman et al., 1983, Mort et al., 2003).
Section snippets
Subjects
Ferber and Karnath (2001a) investigated 35 patients with spatial neglect following right-sided brain damage. Spatial neglect had been diagnosed when the patients showed the following clinical signs: (i) a spontaneous deviation of the head and eyes toward the ipsilesional side, (ii) orienting towards the ipsilesional side when addressed from the front or the left, and (iii) ignoring of contralesionally located people or objects. In addition, each patient had to achieve a minimum omission score
Results
Fig. 3a illustrates conventional lesion density plots for those neglect patients who showed a pathological bias in line bisection and for those who did not show this deficit. The number of overlapping lesions is color coded with increasing frequencies from violet (n = 1) to red (n = max. number in the respective group). To identify the cortical structures that are commonly damaged in neglect patients who show additional deviation in line bisection but are typically spared in neglect patients
Discussion
Our comparison of neglect patients with and without additional bisection bias clearly indicates that poor performance on the line bisection task is associated with relatively posterior brain damage. This finding may help reconcile the discrepant findings regarding the cortical anatomy of spatial neglect. While we have previously suggested that spatial neglect most frequently is associated with damage to the right superior temporal cortex and insula (Karnath, 2001, Karnath et al., 2001, Karnath
Acknowledgments
This work was supported by the Deutsche Forschungsgemeinschaft (SFB 550-A4) and the National Institutes of Health (R01-042047). We would also like to thank Masud Husain for his insightful comments.
References (42)
Performing allocentric visuospatial judgements with induced distortion of the egocentric reference frame: an fMRI study with clinical implications
NeuroImage
(2003)- et al.
Right hemisphere control of visuospatial attention: ‘Line-bisection’ judgments evaluated with high-density electrical mapping and source analysis
NeuroImage
(2003) - et al.
How long is a piece of string? A study of line bisection in a case of visual neglect
Cortex
(1988) - et al.
Left visuo-spatial neglect: a meaningless entity
Cortex
(1992) - et al.
Spatial cognition: evidence from visual neglect
Trends Cogn. Sci.
(2003) - et al.
Pseudoneglect: a review and meta-analysis of performance factors in line bisection tasks
Neuropsychologia
(2000) - et al.
Disorders of spatial orientation
- et al.
Within- and between-task dissociations in visuo-spatial neglect: a case study
Cortex
(1995) Extrapersonal visual unilateral spatial neglect and its neuroanatomy
NeuroImage
(2001)- et al.
The anatomy of unilateral neglect after right-hemisphere stroke lesions—A clinical CT-scan correlation study in man
Neuropsychologia
(1986)
The generality of parietal involvement in visual attention
Neuron
Line bisection in hemianopia
J. Neurol., Neurosurg. Psychiatry
Distinct syndromes of hemineglect
Arch. Neurol.
Control of goal-directed and stimulus-driven attention in the brain
Nat. Rev., Neurosci.
Exploring the syndrome of spatial unilateral neglect through an illusion of length
Exp. Brain Res.
Misrepresentation of horizontal space in left unilateral neglect: Role of hemianopia
Neurology
How to assess spatial neglect-line bisection or cancellation tasks
J. Clin. Exp. Neuropsychol.
Size perception in hemianopia and neglect
Brain
Timing of right parietal and frontal cortex activity in visuo-spatial perception: a TMS study in normal individuals
NeuroReport
Line bisection judgments implicate right parietal cortex and cerebellum as assessed by fMRI
Neurology
The bells test: a quantitative and qualitative test for visual neglect
Int. J. Clin. Neuropsychol.
Cited by (126)
Middle Cerebral Artery Disease
2021, Stroke: Pathophysiology, Diagnosis, and ManagementDisconnection somewhere down the line: Multivariate lesion-symptom mapping of the line bisection error
2020, CortexCitation Excerpt :The pioneering work of Binder et al. (1992) showed that patients with line bisection errors were likely to have posterior lesions, whereas patients who were impaired on the cancellation task had more anterior damage. This finding has been confirmed by Rorden et al. (2006), and also further studies found dissociations between both tasks (Thiebaut de Schotten et al., 2014; Vaessen et al., 2016; Verdon et al., 2010). Line bisection was associated with lesions to the inferior parietal lobe, posterior parts of the SLF, the arcuate fasciculus and nearby callosal fibres.
Keeping an eye on visual search patterns in visuospatial neglect: A systematic review
2020, NeuropsychologiaMapping the anatomy of perceptual pseudoneglect. A multivariate approach
2020, NeuroImageCitation Excerpt :Better experimental control might also explain why our results do agree, yet only in part, with existing lesion studies. Damage due to stroke in the middle occipital gyrus (Binder et al., 1992; Rorden et al., 2006; Zago et al., 2017) as well as anterior insula and inferior frontal cortex (Zago et al., 2017) has been associated with pathological rightward biases in line bisection or landmark tasks, consistent with functional activity observed along the IOS and in VFC, respectively. However, lesions to the supramarginal gyrus associated with a principal component of perceptive neglect (Verdon et al., 2010) and TOJ lesions causing rightward biases (Zago et al., 2017) are more ventral than activation associated with pseudoneglect as found here.