Multimodal and multispatial deficits of verticality perception in hemispatial neglect

Neuroscience 188 (2011) 68 –79

MULTIMODAL AND MULTISPATIAL DEFICITS OF VERTICALITY PERCEPTION IN HEMISPATIAL NEGLECT K. S. UTZ,a* I. KELLER,b F. ARTINGER,c O. STUMPF,c J. FUNKd AND G. KERKHOFFa,e

Key words: neglect, visual, haptic, subjective vertical, multisensory, 3D space.

a International Research Training Group 1457 “Adaptive Minds,” Saarland University, Building A1.3, 66123 Saarbruecken, Germany

BRAIN DAMAGE, SPATIAL ORIENTATION AND HEMINEGLECT

b

Schoen Clinic Bad Aibling, Department of Neuropsychology, Kolbermoorer Stra␤e 72, 83043 Bad Aibling, Germany

Reading the clock is an essential part of our everyday activities. However, imagine not being able to judge the exact position of the clock hands and therefore never being on time. Or imagine, all the paintings on your wall are crooked, but you don’t recognize it. Such impairments frequently occur in patients with right-hemispheric brain lesions after stroke. Their perception of vertical, horizontal or oblique lines (such as clock hands or picture frames) is often impaired. A method commonly used to measure this deficit is the assessment of the subjective visual vertical (SVV) or horizontal (SVH). Here, a movable rod is rotated in the frontal (roll) plane, until the patient indicates that it is aligned with the physical vertical or horizontal. Bender and Jung (1948) were among the first who systematically studied the SVV and SVH in healthy and brain-damaged individuals. In the latter group, they found contraversive deviations larger than 2° from the veridical vertical in case of a frontal or parietal, but not occipital lobe lesion. Several subsequent studies confirmed these findings showing that patients with right-hemispheric brain lesions show a contraversive tilt of their SVV in the frontal plane (Kerkhoff and Zoelch, 1998; Kerkhoff, 1999; Saj et al., 2005a), and a similar or even larger tilt of the subjective haptic vertical (SHV; Kerkhoff, 1999; Funk et al., 2010a). Furthermore, these deficits seem to be multimodal as the tilt in the visual modality is significantly correlated with that in the haptic modality (Kerkhoff, 1999; Perennou et al., 2008), and with postural deficits in patients with right parietal lesions (Perennou, 2006). Lesion analyses in patients with a contraversive SVV tilt in the frontal plane identified the posterior insula, the human homologue of the monkey parieto-insular vestibular cortex (Guldin and Grusser, 1998), as one crucial lesion site (Brandt et al., 1994). However, recent research revealed that also the right parietal cortex, which is considered as a key area for multimodal space integration and representation (for review, see Andersen et al., 1997) is critical for multimodal axis tilts (Perennou et al., 2008). Notably, these lesion sites are adjacent to and are partly overlapping with those anatomical regions associated with visual hemineglect, namely the insula (Karnath et al., 2004), the temporo-parietal junction (Vallar and Perani, 1986), the superior temporal gyrus (Kar-

c

University of Applied Sciences, Faculty of Mechanical Engineering and Mechatronic, Moltkestra␤e 30, 76133 Karlsruhe, Germany d Ludwig Maximilian University, Department of General and Experimental Psychology/Neuro-cognitive Psychology, Leopoldstrasse, 13, 80802 Munich, Germany e Saarland University, Clinical Neuropsychology Unit, Building A1.3, 66123 Saarbruecken, Germany

Abstract—Recent evidence suggests that patients with leftsided visuospatial neglect often show deviations in their visual and haptic perception of verticality in the frontal and sagittal plane. However, little is known about the multimodality of these impairments and the relationship between deviations in the frontal and the sagittal plane. Moreover, no previous study has combined investigations of verticality judgments in both modalities and both spatial planes within the same sample of subjects using the same apparatus. Thus, the aim of the present study was to investigate both subjective visual vertical (SVV) and subjective haptic vertical (SHV) judgments in the frontal and the sagittal plane in rightbrain-damaged patients with visuospatial neglect (nⴝ16), right-brain-damaged patients without neglect (nⴝ18) and age-matched healthy individuals (nⴝ16) using the same testing device for all tasks. This allowed for direct comparisons of visual vs. haptic and frontal vs. sagittal verticality judgments. Neglect patients showed significant counterclockwise tilts in their SVV and SHV judgments in the frontal plane as well as marked backward (upper end of the rod towards the observer) tilts in the sagittal plane. In contrast, right-braindamaged patients without neglect and healthy individuals showed no marked deviations in the frontal plane, but small forward (upper end of the rod away from the observer) tilts in the sagittal plane. Moreover, neglect patients showed significantly higher unsigned errors in all tasks. These results demonstrate multimodal and multispatial deficits in the judgment of verticality in patients with visuospatial neglect which are most likely due to an altered representation of verticality caused by lesions of brain areas related to multisensory integration and space representation in the right temporoparietal cortex. © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. *Corresponding author. Tel: ⫹49-681-302-57385; fax: ⫹49-681302-57382. E-mail address: [email protected] (K. S. Utz). Abbreviations: BJLOT, Benton Judgment of Line Orientation test; M, mean; SD, standard deviation; SEM, standard error of the mean; SHV, subjective haptic vertical; SV, subjective vertical; SVH, subjective visual horizontal; SVV, subjective visual vertical.

0306-4522/11 $ - see front matter © 2011 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2011.04.068

68

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

nath, 2001; Karnath et al., 2004), posterior parietal (Mesulam, 1999) and intraparietal cortices (Mort et al., 2003; Verdon et al., 2010), the thalamus and basal ganglia (Vallar and Perani, 1986; Karnath et al., 2004). Given the anatomical proximity, it is not surprising that neglect patients show visuospatial deficits in the perception of axis orientation judgments (Kerkhoff and Zoelch, 1998; Yelnik et al., 2002; but see Johannsen et al., 2006 for slightly diverging results in neglect patients with pusher symptoms), in addition to their well-known contralesional deficits in visual search, cancellation and drawing tasks. Moreover, a close association between visuospatial neglect and postural disorders has been repeatedly found (Saj et al., 2005b; Perennou, 2006), and an association between disturbances of the subjective vertical (SV) and poor postural recovery was shown (Bonan et al., 2007). Current theories of gravity processing hypothesize that the representation of verticality is based on the integration of visual, vestibular and somatosensory input (Brandt and Dieterich, 1999), which is assumed to depend crucially on the vestibular cortex (Bronstein, 1999; Mittelstaedt, 1999). Consequently, impaired integration of sensory signals due to brain lesions in the above-mentioned brain areas, leads to asymmetrical processing of sensory input resulting in perturbed spatial representations (Funk et al., 2010a). Modulation of the SV due to asymmetrical sensory input was shown to be inducible in healthy individuals by changes of posture (Luyat and Gentaz, 2002) with reduced precision of vertical judgments during body tilt or via galvanic vestibular stimulation with SV deviations towards the anode (Mars et al., 2001). In neglect patients lateral head tilts modulated SV judgments in direction of the head inclination (Funk et al., 2010b) and changes of posture resulted in larger SV deviations in supine as compared to an upright body position (Funk et al., 2010a). Furthermore galvanic vestibular stimulation reduced the counterclockwise SV tilts of neglect patients during left-cathodal stimulation (Saj et al., 2006). Together, these studies show that manipulations of sensory input (via postural changes or galvanic vestibular stimulation) affect SV judgments in general. However, patients with visuospatial neglect seem to be much more susceptible to such manipulations than healthy subjects or non-neglecting patients. These results have been interpreted in favour of an unstable and tilted representation of verticality in neglect patients (Funk et al., 2010b). However, verticality perception does not only include the frontal (roll) plane, but also the sagittal (pitch) plane (see Fig. 1A). Up to now, only few studies have investigated the perception of the SV in the sagittal (pitch) plane in patients with brain lesions. Saj et al. (2005a) found that neglect patients showed a backward tilt of the SVV (upper end of the rod towards the observer), in addition to a counterclockwise SVV tilt in the roll plane. Moreover, a greater individual variability in SVV judgments in the pitch plane compared to the roll plane was shown, indicating that the task is more difficult in pitch than in roll. Furthermore, the SVV judgments in pitch

69

and roll were not intercorrelated, suggesting independent processing of verticality in these two spatial planes. Funk et al. (2010a) extended this finding to the haptic modality showing backward tilts in SHV judgments of neglect patients in the pitch plane. Moreover, a larger tilt was observed in patients with severe as compared to moderate visuospatial neglect, suggesting a close relationship to neglect severity. In contrast to Saj et al. (2005a), SHV values in roll and pitch were highly intercorrelated in the study of Funk et al. (2010a), indicating comparable results in both spatial planes. One limitation of the study of Saj et al. (2005a) was that patients touched the rod for visual SVV adjustments to the subjective vertical. Thus, this task was not purely visual but rather a combined visual-haptic SV task which does not allow for a direct comparison of the two modalities (haptic, visual). Therefore, one aim of the present study was to investigate SVV judgments in rightbrain damaged patients by excluding haptic input cues for verticality. Furthermore we sought to analyze the relationship between visual and haptic SV judgments in pitch and roll. Thus, the experiment presented here, investigated for the first time, both SVV and SHV judgments in pitch and roll in right-brain-damaged patients with neglect, right-brain-damaged patients without neglect and age-matched healthy individuals using a novel testing device for all these tasks in order to allow direct comparisons between modalities and spatial planes. Based on previous findings (Saj et al., 2005a; Funk et al., 2010a) the following hypotheses were formulated: (1) Neglect vs. non-neglect: Right-brain damaged patients with left visuospatial neglect show larger directional errors as well as unsigned errors of the SVV and SHV in pitch and roll than right-brain-damaged patients without neglect or healthy individuals. (2) Direction of tilts: SVV and SHV tilts of neglect patients, but not of patients without neglect or healthy controls, are directed counterclockwise in the roll plane and backward in the pitch plane. (3) Modalities: With respect to the two modalities, we expected comparable deviations in the visual and haptic modality and significant intercorrelations based on previous findings (Kerkhoff, 1999). (4) Roll vs. pitch plane: Concerning the relationship between SVV in roll and SVV in pitch as well as between SHV in roll and SHV in pitch we had no specific predictions, since there were both observations of correlated deviations between the SHV in pitch and roll (Funk et al., 2010a), as well as dissociated impairments of the SVV in pitch and roll (Saj et al., 2005a).

EXPERIMENTAL PROCEDURES Participants Sixteen patients (nine males, seven females) with right-hemispheric vascular brain lesions and left visual neglect as indicated by six clinical neglect tests (see “Assessment of visual field, visual neglect and visuospatial perception,” below) with a mean age of 71 years (range: 52– 86) participated in the study. Furthermore, 18

70

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

Fig. 1. (A) Schematic illustration of the frontal (or roll) and the sagittal (or pitch) plane; (B) Illustration of the visual/haptic vertical testing device. For details see “Experimental verticality tasks”.

patients (14 males, four females) with right-hemispheric vascular brain lesions, but without visual neglect with a mean age of 70 years (range: 47– 84) were recruited. Additionally, 16 agematched healthy control subjects (10 males, six females) with a mean age of 66 years (range: 37– 85) were included in the study. Table 1 displays the demographic and clinical data of the patients. There was no significant difference between the groups concerning age (F(2)⫽.724, P⫽.49), nor did the gender distribution differ significantly between groups (␹2(2)⫽1.88, n⫽50, P⫽.39). The mean time since lesion was 78 days (standard deviation (SD)⫽53.02) for the neglect patients and 61 days (SD⫽79.91) for the patients without neglect, which was not significantly different (t(32)⫽⫺0.77, P⫽.45). All participants were right-handed as indicated by the German Version of the Edinburgh Handedness Inventory (Oldfield, 1971) and had normal or corrected to normal vision with a decimal visual acuity of at least 0.63 (20/30 Snellen equivalent) for the near viewing distance (0.4 m) as determined by a letter acuity test (Oculus Nahleseprobe, Oculus, Wetzlar, Germany). Visual acuity did not differ significantly between the three groups (F(2)⫽2.98, P⫽.06). All participants were informed of the

experimental protocol which was conducted in accordance with the Declaration of Helsinki II and gave their written informed consent prior to their participation in the study.

Assessment of visual field, visual neglect and visuospatial perception The binocular visual field was mapped in all patients with a software for static binocular campimetry (EyeMove, Kerkhoff and Marquardt, 2009). Visual neglect was tested with six conventional neglect tests. They included letter cancellation, star cancellation, figure copy (star, rhomb, flower) and paragraph reading of a 140-word reading test (subtests of the Neglect-Test [NET; Fels and Geissner, 1996], German version of the Behavioural Inattention Test [BIT; Wilson et al., 1987]). Additionally, number cancellation (20 targets in 200 distractors, 10 targets per hemispace) and horizontal line bisection (Schenkenberg et al., 1980) were conducted. All tests were presented on a 29.7⫻21 cm white sheet of paper with its centre aligned with the patient’s trunk midline. All patients carried out the tasks using their ipsilesional, right hand.

Table 1. Clinical and demographic data of right-brain-damaged patients with neglect (RBD⫹) and without neglect (RBD⫺, see “Participants” for further details) Age, sex

Etiology

Lesion location

Days since lesion

Motor deficit

Visual field defect

Visual acuity, near (0.4 m)

RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ RBD⫹ Mean

74, f 65, f 67, m 70, f 68, m 78, f 62, m 83, m 86, m 52, f 70, m 77, f 70, f 63, m 73, m 72, m 71 y; 7 f/9 m

MCI ICB MCI MCI MCI MCI MCI ICB MCI MCI MCI ICB ICB MCI MCI MCI, PCI

Capsula interna Basal ganglia Frontal, parietal Frontal, temporal Frontal, parietal Frontal, parietal Frontal Basal ganglia Frontal Parietal, temporal Frontal, parietal Thalamus, basal ganglia Parietal Temporal, parietal Parietal, temporal Parietal, basal ganglia —

26 88 97 44 201 88 155 31 117 27 129 93 63 16 44 30 78

HP HP HP HP HP HP N HPL HP N HPL HP HP HPL HP HP —

N L-HH — N N — — L-HH N L-HH L-HH L-HH L-HH L-HH L-HH L-HH —

100 63 63 100 80 80 63 80 125 63 63 63 80 63 125 125 83.5%

RBD⫺ RBD⫺ RBD⫺

70, m 47, m 84, f

MCI MCI MCI

29 27 18

HPL N HP

N N N

RBD⫺

84, m

MCI

13

HP

N

RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ RBD⫺ Mean

80, m 76, m 63, m 76, f 65, m 61, m 55, m 75, f 62, m 72, f 69, m 71, f 83, m 65, m 66 y, 6 f/10 m

MCI /PCI MCI MCI Pons infarction MCI ACI ICB MCI MCI MCI ICB MCI MCI ChAI —

Parietal, basal ganglia Temporal, parietal Claustrum, capsula externa Frontal, parietal temporal, capsula interna Temporal Parietal, insula Temporal Pons Frontal Capsula interna Basal ganglia, insula Frontal Frontal Frontal Frontal Thalamus, white matter Capsula interna Temporal —

110 23 54 30 10 312 23 20 214 95 22 31 42 25 61

N HP HP HP HP HP HP HP HP HP HPL HP HP HP —

L-HH N L-UQ N N N N N N N N N N N —

Star cancellation, omissions L/R, max (27/27)

Letter cancellation, omissions L/R, max (20/20)

Number cancellation, omissions L/R, max (10/10)

Reading, omissions L/M/R, max (47/46/47)

Figure copy, L/R

Line bisection ⫺/⫹ mm

5 57 0.7 ⬍0.7 22 40 5 11 11 ⬍0.7 ⬍0.7 ⬍0.7 ⬍0.7 ⬍0.7 ⬍0.7 ⬍0.7 9.48

0/0 3/0 27/11 2/0 4/4 27/6 27/7 13/0 6/3 9/7 27/17 27/11 27/17 27/20 27/16 27/2 19/8

5/4 0/2 15/3 9/13 12/12 19/5 20/4 12/0 2/6 2/2 20/4 16/10 20/15 20/17 20/13 2/1 12/6

3/1 2/1 10/2 1/0 2/0 10/2 1/0 2/0 10/2 2/2 10/8 10/6 10/7 10/6 10/5 7/0 6/3

3/2/5 1/0/0 47/46/19 1/0/4 21/5/5 47/34/2 47/44/42 0/1/0 0/0/1 — 47/46/31 47/2/0 47/46/42 47/46/47 47/46/41 47/46/0 30/24/16

⫹11 ⫹18 ⫹18 ⫹5 ⫹7 ⫹43 ⫹56 ⫹13 ⫹5 ⫹20 ⫹56 ⫹24 ⫹68 ⫹51 ⫹57 ⫹30 ⫹30

125 125 125

74 74 ⬍0.7

2/0 0/0 2/1

0/1 1/0 1/1

0/0 0/0 0/0

0/0/0 0/0/0 1/0/0

⫹/⫹ ⫹/⫹ ⫹/⫹ ⫺/⫹ ⫺/⫹ ⫺/⫹ ⫺/⫹ ⫺/⫹ ⫹/⫹ ⫺/⫺ ⫺/⫹ ⫺/⫹ ⫺/⫺ ⫺/⫺ ⫺/⫹ ⫺/⫺ 12 impaired ⫹/⫹ ⫹/⫹ ⫹/⫹

80

5

1/2

1/1

0/0

0/0/0

⫹/⫹

22 ⬎74 ⬎74 ⬎74 ⬎74 ⬎74 74 40 0.7 57 ⬎74 ⬎74 ⬎74 ⬎74 66.48

2/3 0/1 0/0 1/0 0/0 0/0 0/0 0/0 2/1 0/0 0/5 0/0 1/1 0/0 1/1

1/0 0/0 1/0 1/3 0/0 1/1 0/1 0/0 2/3 1/3 4/0 1/0 1/1 0/0 1/1

0/0 0/1 0/0 0/0 0/0 0/0 0/0 1/0 0/0 5/0 1/1 0/0 1/0 0/0 0/0

0/0/0 0/0/0 0/1/1 0/0/1 0/0/0 0/1/0 1/0/0 — 0/0/0 0/0/0 0/0/0 1/1/0 0/0/0 0/0/0 0/0/0

125 100 63 125 100 125 125 63 100 100 80 80 63 125 101%

BJLOT, percentile

⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ ⫹/⫹ 0 impaired

⫹9 ⫹1 ⫹2 ⫺3

0 ⫺1 ⫹1 ⫺4 ⫹1 ⫺2 ⫹2 ⫹1 ⫺1 ⫹6 ⫹3 ⫹10 ⫹1 0 ⫹1

71

Abbreviations: m, male; f, female; ICB, intracerebral bleeding; MCI, middle cerebral artery infarction; PCI, posterior cerebral artery infarction; ChAI, choroid artery infarction; HP, hemiparesis; HPL, hemiplegia; N, normal function; L-HH, left homonymous hemianopia; L-UQ, left upper homonymous quadrantanopia; BJLOT, Benton Judgment of Line Orientation Test. Neglect tests: L, left; M, middle; R, right. Figure copy: ⫺, omissions or distortions; ⫹, normal performance. Line Bisection: ⫹, rightward deviation; ⫺, leftward deviation.

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

Group

72

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

To determine pathological performance in the neglect tests, the following cut-off scores were applied: more than one omission in the left hemispace in the cancellation and reading tasks and omissions or significant distortions of the left half of the copied figures. For line bisection, the confidence interval of the performance of 10 healthy, age-matched individuals (mean age: 72 years; range: 48 – 84), all using their right dominant hand for bisection, was used as cut-off criterion. The limits were ⫺0.34 mm deviation to the left and ⫹2.88 mm deviation to the right from the true centre. Patients were considered as having left-sided visuospatial neglect when they were impaired in at least three out of the six neglect tests. In order to assess visuospatial perception beyond the typical neglect tests, we also administered the short form (Qualls et al., 2000) of the Benton Judgment of Line Orientation Test (BJLOT; Benton et al., 1978). This frequently used measure of visuospatial perception consists of matching line segments of different spatial orientation with 11 longer lines on a response card. Right-braindamaged patients with visuospatial neglect had significantly lower percentiles (M (mean)⫽9.48, SD⫽16.71) indicating impaired visuospatial perception as compared to right-brain-damaged patients without neglect (t(32)⫽5.64, P⬍.001), who showed normal performance on average (M⫽66.48, SD⫽37.16).

performed both tasks in the roll and pitch plane in this way, except that the device was repositioned for pitch plane assessment to allow for rotation of the rod back and forth. The initial starting positions of the rod were ⫹20° or ⫺20° rotated from the true vertical (90°) in each spatial plane. Positive values indicated a counterclockwise tilt of the rod in the roll plane or a backward tilt in the pitch plane. Negative values indicated a clockwise tilt of the rod in the roll plane or a forward tilt in the pitch plane. Before each task (SVV, SHV) in each spatial plane (roll, pitch), two corresponding practice trials were performed (not scored) to ensure that all subjects were familiar with the task before the experimental trials began. Afterwards, participants had to perform six trials per task and spatial plane, three from each starting position. The order of the modalities and the spatial planes were counter-balanced across participants to control for sequence effects. During the presetting of the rod, participants kept their eyes closed. Constant errors as a measure of the central tendency of the SV deviation from the veridical vertical were calculated: the true vertical (i.e. 90°) was subtracted from the individual SV judgment for each of the six measurements (degrees of angle) in each modality and each plane and the mean of the six difference scores of each condition was computed. Likewise, unsigned, averaged errors were computed via the absolute values indicating the mean error of deviation, regardless of its direction.

Experimental verticality tasks The visual- and haptic-spatial performance of the participants was measured with a specifically designed apparatus for the measurement of visual and haptic axis orientation in different spatial planes (“Haptic & Vision Meter,” see Fig. 1B). The electricity-driven device consists of an aluminium box, 75 cm high, 59.5 cm wide and 31 cm deep. On the front side a plastic rod (0.7⫻21.5 cm2) was mounted on a disc (6 cm in diameter), which was fixed 12 cm away from the lower edge of the plane. The rod together with the disc could be rotated along the frontal plane. The rod could be illuminated from the inside in red for the visual verticality tasks. Inside the box, a multiphase motor including an encoder for angle measurement was mounted in the rotation axis. The device was connected to a touch-panel by which the starting position of the rod could be controlled and which also displayed the final verticality settings of the subject with a precision of 0.16°. A black blanket with a slot for the disc and rod covered the outer edges of the box during testing to remove any visible vertical or horizontal reference cue when the rod was illuminated for testing of the SVV. For the haptic condition, participants were blindfolded and the rod was not illuminated. Participants were seated directly in front of the apparatus at a distance of 40 cm, easily allowing manipulation of the disc carrying the rod with their right, ipsilesional arm. Their head was supported by a head- and chinrest with lateral stabilizers to ensure an earth vertical orientation of head and trunk during testing.

Statistical analyses All data analyses were computed with SPSS, version 17. Mixed 3⫻2⫻2 analyses of variance (ANOVAs) with the between factor “group” (neglect, no neglect, healthy controls) and the within factors “plane” (roll, pitch) and “modality” (visual, haptic) were conducted, separately for unsigned errors and constant errors, followed by Bonferroni-corrected post hoc tests. To determine, whether constant errors of the three groups differ significantly from zero, one-sample t-tests were calculated for each group. Two-tailed Pearson correlation coefficients between SVV and SHV performance (separately for constant errors and unsigned errors) in roll and pitch were computed in order to determine the relationship between SVV and SHV performance in both planes and between measurements within the same plane across the two modalities. Furthermore the same analyses were used to determine relationships between SVV in roll and SVV in pitch as well as between SHV in roll and SHV in pitch.

RESULTS Fig. 2 graphically summarizes the main results of the verticality tasks in the visual and haptic modality as well as in roll and pitch separately for the three subject groups. Unsigned errors

Procedure Each participant was tested individually in a quiet room which could be completely darkened. First, patients performed the screening tests (neglect tests, BJLOT and perimetry; not performed by healthy controls). The SVV was assessed in complete darkness to remove any visual reference cues apart from the luminous red rod. Before each experimental session, the apparatus and rod were calibrated according to the earth vertical. All participants were instructed to hold the disc connecting the illuminated rod with the device with their right hand and to adjust the rod to the vertical by rotating the disc. Participants were not allowed to touch the rod during their adjustments, thus ruling out the influence of additional haptic information of verticality on their settings. For the assessment of the SHV, participants were blindfolded and had to adjust the very same rod to the vertical, this time by touching and rotating the rod (instead of the disc). Participants

There was a significant main effect of group on the unsigned error (F(2,47)⫽18.58, P⬍.001, ␩2⫽.44). Pairwise comparisons revealed that unsigned errors of right-braindamaged patients with neglect (M⫽7.13, SEM (standard error of the mean)⫽0.51) were generally larger (P⬍.001) than those of right-brain-damaged patients without neglect (M⫽4.17, SEM⫽0.48) and larger (P⬍.001) than those of healthy participants (M⫽2.85, SEM⫽0.51). The latter two groups did not differ significantly in their unsigned errors (P⫽.20). There was also a significant main effect of “plane” on the unsigned error (F(1,47)⫽14.57, P⬍.001, ␩2⫽.24), suggesting that unsigned errors were larger in pitch (M⫽5.59, SEM⫽0.42) than in roll (M⫽3.84, SEM⫽0.31). No signifi-

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

73

Fig. 2. Performance of right-brain-damaged patients with neglect (RBD⫹), healthy controls and right-brain-damaged patients without neglect (RBD⫺) in the SVV and SHV tasks in roll and pitch. Each line represents the mean constant error of one participant. Note the counterclockwise and backward (upper end of the rod towards the observer) tilts of the SVV and SHV and the larger interindividual variability of judgments of RBD⫹ patients (with neglect) as compared with the other two subject groups.

cant main effect was found for “modality” (F(1,47)⫽2.24, P⫽.14, ␩2⫽.05), indicating that unsigned errors in the visual and haptic SV tasks were in general comparable. Furthermore, the factor “plane” interacted significantly with the factor “modality” (F(1,47)⫽10.92, P⫽.002, ␩2⫽.19), reflecting different effects of “plane” on the unsigned errors in the two modalities. Post hoc tests, com-

paring the different levels of the two factors showed that unsigned errors differed significantly in roll and pitch in the visual SV task (P⬍.001), but not in the haptic SV task (P⬎.05). Fig. 3 shows the interaction graph, suggesting that unsigned errors in roll were smaller in the SVV than in the SHV task, but vice versa in the pitch plane. Although unsigned errors for both SVV and SHV were increased in

74

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

SVV in pitch (P⫽.13), when Bonferroni-corrected. No significant differences in the unsigned errors were found between healthy individuals and control patients (all Ps⬎.34). Constant errors

Fig. 3. Interaction between plane and modality for mean unsigned errors of the SVV and SHV (⫹/⫺1 SEM). Note that the larger unsigned errors in the visual compared to the haptic modality in roll contrast with slightly smaller unsigned errors in the visual compared to the haptic modality in pitch.

pitch compared to roll, this increase was more pronounced in the SVV, indicating that SVV judgments were more susceptible to the plane in which the task was performed than SHV judgments. No significant interactions were found for “plane”⫻“group” (F(2,47)⫽7.51, P⫽.48, ␩2⫽.03), “modality”⫻“group” (F(2,47)⫽0.02, P⫽.98, ␩2⫽.001), or “plane”⫻“modality”⫻“group” (F(2,47)⫽0.07, P⫽.93, ␩2⫽ .003). Furthermore, pairwise comparisons (see Fig. 4) showed that neglect patients had significantly larger unsigned errors (all Ps⬍.01) than healthy individuals in all task combinations (SVV in roll, SVV in pitch, SHV in pitch, SHV in pitch). Neglect patients’ unsigned errors were also significantly larger than those of control patients in the SVV (P⫽.003) and SHV in roll (P⫽.02), marginally larger for the SHV in pitch (P⫽.08), but not significantly different for the

As reported for unsigned errors, there was a significant main effect of “group” on the constant errors (F(2,47)⫽ 13.83, P⬍.001, ␩2⫽.37). Post hoc tests showed that constant errors of neglect patients (M⫽3.52, SEM⫽0.65) significantly differed (P⬍.001) from those of the patients without neglect (M⫽⫺0.93, SEM⫽0.69) and healthy participants (M⫽⫺0.85, SEM⫽0.69). No significant differences were found between control patients and healthy participants (P⫽1). As for unsigned errors, no main effect of modality on the constant errors was found (F(1,47)⫽0.40, P⫽.53, ␩2⫽.01). However, the main effect of “plane” was significant (F(1,47)⫽6.26, P⫽.02, ␩2⫽.12), and was further specified by a significant “plane”⫻“group” interaction (F(2,47)⫽4.83, P⫽.01, ␩2⫽.17), suggesting different effects of “plane” on the unsigned errors in the three subject groups. Post hoc tests revealed that constant errors differed significantly between neglect patients and healthy individuals in the roll plane (P⫽.03), but not between neglect patients and control patients (P⫽.13) or healthy individuals and control patients (P⫽1). In the pitch plane, neglect patients significantly differed in their constant errors from healthy individuals (P⬍.001) as well as from control patients (P⬍.001). Again, healthy individuals and control patients did not differ significantly in their constant errors (P⫽1). However, the reported differences between those groups were larger in pitch than in roll. When looking at the interaction graph (Fig. 5), it can be seen that the positive constant errors of neglect patients are larger in the pitch compared to the roll plane. In contrast, the negative constant errors in roll of the control patients and healthy participants become more negative in pitch.

Fig. 4. Mean unsigned errors (⫹/⫺1 SEM) of the three subject groups in the different tasks. HC, healthy controls; RBD⫹, right-brain-damaged patients with neglect; RBD⫺, right-brain-damaged patients without neglect. Note that RBD⫹ patients showed larger unsigned errors in all tasks as compared to the other groups.

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

75

Relation between SV in pitch and roll and in the visual and haptic modality Table 2 displays the two-tailed Pearson correlations between the constant errors and unsigned errors in the different spatial tasks for the whole sample. Both directional (constant) as well as unsigned errors were significantly (marginally in one case) correlated in both modalities and both planes. It is interesting to note that the two nonsignificant correlations concern correlations across the two modalities and two spatial planes. This may indicate, that the commonalities between verticality judgments in different modalities are less pronounced when different reference planes are considered.

DISCUSSION

Fig. 5. Interaction between plane and group for mean constant errors of the SVV and SHV (⫹/⫺1 SEM). RBD⫹, right-brain-damaged patients with neglect; RBD⫺, right-brain-damaged patients without neglect; HC, healthy controls. Positive values indicate a counterclockwise tilt, negative values a clockwise tilt in roll; positive values indicate a backward (upper end of the rod towards the observer) tilt, negative values a clockwise (upper end of the rod away from the observer) tilt in pitch. Note the positive constant errors of RBD⫹ patients in roll, which get larger in pitch. In contrast, the minimal positive constant errors of the other two groups in roll get larger but negative in pitch.

The interactions “modality”⫻“group” (F(2,47)⫽1.14, P⫽.33, ␩2⫽.05), “plane”⫻“modality” (F(1,47)⫽0.20, P⫽.66, ␩2⫽.004) and “plane”⫻“modality”⫻“group” (F(2,47)⫽0.53, P⫽.59, ␩2⫽.02) were not significant. Additional Bonferroni-corrected pairwise comparisons (see Fig. 6) showed that right-brain-damaged patients with neglect had significantly larger counterclockwise/backward deviations than healthy controls and right-brain-damaged patients without neglect in all task combinations (SVV in roll, SVV in pitch, SHV in pitch; all Ps⬍.04) except for the SHV in roll (P⫽.25). Healthy individuals and right-braindamaged patients without neglect had smaller counterclockwise/backward tilts in the roll or clockwise/forward tilts in pitch and those groups did not differ significantly from each other in their constant errors (all Ps⬎.73). Additionally, one-sample t-tests against zero were conducted for the constant errors in each task of each group. This was done in order to evaluate whether differences in constant errors between neglect patients and the control groups were caused by significant counterclockwise or backward tilts in the SV of neglect patients. Constant errors of right-brain-damaged patients with neglect differed significantly from zero in all tasks (all Ps⬍.04), except in the SHV in roll, where constant errors differed only marginally from zero (P⫽.08). In contrast, constant errors of right-brain-damaged patients without neglect did not differ significantly from zero (all Ps⬎.24), except for the SVV in pitch (P⫽.002). Similarly, there were no significant differences from zero of constant errors in the healthy participants (all P⬎.20), except for the SHV in pitch (P⫽.03).

The purpose of the present study was to investigate SVV and SHV judgments and their relationship in roll and pitch in right-brain-damaged patients with visuospatial neglect, right-brain-damaged patients without neglect and agematched healthy individuals using the very same testing device for all tasks. This device was developed to rule out potential cross-modal influences during task handling in order to allow direct and unbiased comparisons of performance in the two modalities and the two spatial planes. With respect to our four hypotheses (see above) our results were clear-cut. As expected, right-brain-damaged patients with visuospatial neglect showed larger and systematic counterclockwise (in roll) and backward deviations (in pitch) in their SVV and SHV judgments as compared to right-brain-damaged patients without neglect and healthy individuals. These findings are in line with previous studies on visual-haptic (Saj et al., 2005a) and haptic SV (Funk et al., 2010a) judgments in neglect patients, and extend these findings to combined SV measurements in two modalities and two spatial planes in two large patient samples. Multimodal impairments of verticality perception in spatial neglect The SV deviations in both planes across both modalities were most likely due to an impaired representation of the vertical in consequence of lesions in brain areas that elaborate these vertical representations via the integration of visual, vestibular and somatosensory input. Previous studies on the SVV in the roll plane suggest impaired processing of axis orientation due to damage in areas such as the posterior insula (Brandt et al., 1994), the supramarginal and postcentral gyri (Von Cramon and Kerkhoff, 1993), the thalamus (Dieterich and Brandt, 1993) and the brainstem (Friedmann, 1970; Frisen, 2010). These structures are believed to be part of a graviceptive pathway projecting from the brainstem via the posterior thalamus to the vestibular cortex (Brandt et al., 1994). However, little evidence exists on specific lesion locations in patients with impaired SHV perception. Based on their findings on the visual, haptic and postural vertical in brain-damaged patients, Perennou et al. (2008) suggest an integrated cross-modal verticality perception which is processed in the right pari-

76

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

Fig. 6. Mean constant errors (⫹/⫺1 SEM) of the three groups in the different tasks. RBD⫹, right-brain-damaged patients with neglect; RBD⫺, right-brain-damaged patients without neglect. Positive values indicate a counterclockwise tilt, negative values a clockwise tilt in roll; positive values indicate a backward (upper end of the rod towards the observer) tilt, negative values a forward (upper end of the rod away from the observer) tilt in pitch. Note that RBD⫹ patients had large counterclockwise tilts in roll and backward tilts in pitch. Constant errors of RBD⫺ and healthy controls were generally smaller. Their SV was tilted slightly counterclockwise in roll, but forward in pitch.

etal cortex. The latter finding neatly fits the notion of a multimodal spatial orientation deficit in neglect (Kerkhoff, 1999; Funk et al., 2010a) which was replicated and extended to the pitch plane in the present study. Significant tilts in the SVV and SHV were apparent in neglect patients in contrast to healthy individuals and control patients. The idea of a multimodal representation of verticality in the parietal cortex is further corroborated by animal studies identifying “axis-orientation-selective” (Sakata et al., 1997) and multimodal (e.g. Graziano and Gross, 1995; Duhamel et al., 1998) neurons in the monkey parietal cortex. Moreover, parietal and temporal areas belong to the multisensory cortical vestibular system with the posterior insula— the homologue of the monkey parieto-insular-vestibular cortex—as the core region which is often lesioned in patients with SVV deficits in roll (Brandt et al., 1994). Taken together, these findings suggest a multisensory represenTable 2. Pearson correlations between constant errors/unsigned errors of the whole sample in the different spatial tasks

SVV roll SVV pitch SHV roll SHV pitch

SVV roll

SVV pitch

SHV roll

SHV pitch

—

.21†/.29* —

.34**/.40* .20/.32* —

.40**/.15 .36**/.32* .35**/.53*** —

† P⬍.10 (2-tailed); * P⬍.05 (2-tailed); ** P⬍.01 (2-tailed); *** P⬍.001 (2-tailed).

tation of verticality in the temporo-parietal and vestibular cortex. In addition, the significant correlations between SVV and SHV, which were also shown previously (Kerkhoff, 1999; Perennou et al., 2008), suggest at least partly shared mechanisms of verticality perception in the visual and haptic modality. With respect to the multimodal nature of the SV deficits in our patients, who were recruited on the basis of visual neglect screening tests, it is interesting to note that all patients showed also a nonvisual, haptic deficit in form of a significant tilt of their SHV to the contralesional side and to the back. This finding suggests that patients with visual neglect are very likely to show signs of impaired spatial orientation beyond the visual modality, that is, a multimodal deficit. Multispatial impairments of verticality perception in spatial neglect Apart from the multisensory characteristics of verticality perception, the present study found a multispatial deficit in neglect patients affecting different spatial reference planes (roll, pitch). Their constant errors were directed counterclockwise in the roll plane and backwards in the pitch plane in contrast to healthy individuals and right-brain-damaged patients without neglect. Furthermore, constant errors in roll and pitch were intercorrelated, which is in line with a previous study concerning the SHV (Funk et al., 2010a), suggesting a certain degree of multi-spatiality. To date, there exist no lesion studies on SV deficits in the pitch

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

plane in cortical lesions. In patients with brainstem lesions Frisen (2010) observed deviations of the SVV in pitch and roll. A neurophysiological investigation in awake monkeys (Sakata et al., 1997) showed that so-called “axis-orientation-selective” neurons in the monkey parietal cortex respond to tilted visual objects in different planes, including the pitch plane. Moreover, the vestibular system also acts in a multispatial way, as it is able to assess the position and movement of our body in three-dimensional space (Brandt and Dieterich, 1999). In sum, the above-mentioned studies and the present results imply a multimodal representation of verticality in multiple spatial planes, probably in the same or adjacent areas in the temporo-parietal cortex and the ascending graviceptive pathway. The neglect patients in the present study showed lesions in exactly those areas, centering most often on the right temporal and parietal cortex. As expected, deviations were generally larger in pitch than in roll. Concerning the unsigned errors, this was however only the case for the visual task, not for the SHV as shown by the significant “plane”⫻“modality” interaction. An explanation for this finding could be the requirement for depth perception in the visual but not in the haptic SV task. While the visual rod can be judged from its retinal projection in the roll plane, additional depth cues are needed in the pitch plane (Saj et al., 2005b). On the other hand, constant errors were generally larger in pitch as compared to roll, irrespective of modality, which seems to be incompatible with this interpretation. Nevertheless, SV judgments generally seem to be more difficult in pitch than in roll, since both healthy individuals (in the SHV task) and control patients (in the SVV task) differed significantly from zero in their SV adjustments in pitch. Another evidence for at least partly shared mechanisms underlying verticality perception in different modalities and different spatial planes is the direction of tilts in neglect patients. In neglect patients, the SV deviations of both the SVV and SHV were systematically tilted counterclockwise in the roll plane and analogously backwards in the pitch plane. These counterclockwise or backward tilts are most likely due to right-parieto-temporal lesions. The directions of the tilts seem to depend on the lesion location as patients with left-hemispheric lesions show clockwise tilts in the SVV in roll (Kerkhoff and Zoelch, 1998). Furthermore counterclockwise tilts in the SVV in roll have been observed after paramedian thalamic infarction and counterclockwise as well as clockwise tilts after posterolateral thalamic infaction (Dieterich and Brandt, 1993). In patients with brainstem lesions backward deviations of the SVV in pitch were associated with lesions in the pineal region, cerebellum and craniocervical junction. Forward tilts were seen in patients with posterior thalamic and medullary lesions, and tilts in both directions were equally prevalent with focal intra-axial lesions (Frisen, 2010). Again, all these lesion areas are located in the parietal cortex and within the graviceptive pathway (brainstem, thalamus, vestibular cortex). However, as some of the RBD-patients in the present study also show lesions in temporo-parietal areas, it

77

seems that the multimodal and multispatial SV deficits depend on the presence of neglect per se, rather than on the mere lesion location. In favor of this interpretation, Kerkhoff and Zoelch (1998) found a relationship between the severity of spatial neglect and SVV deficits in the roll plane, that is, the more severe the neglect the larger the SVV tilts. Similarly, Yelnik et al. (2002) observed a correlation between SVV deficits in the frontal plane with visuospatial neglect, but no significant relation between SVV deficits and lesion location. In summary, the present study highlights the relevance of perceptual tilts of the SV in the visual and haptic modality, and points to the considerable deficits of patients with visuospatial neglect in judging the visual and haptic vertical not only in the roll plane, but also in the pitch plane. To our knowledge, such a demonstration of SV deficits in two modalities and two spatial planes using the same testing device for all investigations, is novel and has not been reported before. This technique allows direct and unbiased comparisons between SV judgments in different modalities and spatial planes and the analysis of relations between them. Consequences of impaired SV perception for daily life and the clinical practice It is likely that multimodal deficits in judging the visual or haptic vertical in 3-D-space severely affect the patients’ daily life. Presumably, neglect patients have great difficulties in estimating verticality in light and dark, when no visual orientation cues are present (as in the SHV). This impairment is likely to cause considerable problems in a variety of postural requirements in the roll and the pitch plane (for review see Perennou, 2006). As mentioned in the introduction, stroke patients with deficits in verticality perception have difficulties in many everyday situations including clock-reading (Kerkhoff, 1998), mobility (Kerkhoff, 1999), body stabilization (Perennou, 2006) and body orientation to gravity (Perennou et al., 2008). They frequently show an abnormal trunk position (Saj et al., 2005b) as well as poor balance recovery (Bonan et al., 2007). It is interesting to note in this context, that the association between left-sided hemiplegia, poor ambulation and impaired verticality perception was already noted five decades ago in neurorehabilitation (Bruell et al., 1957; Bruell and Peszczynski, 1958). These deficits should therefore be actively addressed in the diagnosis and neurorehabilitation of patients with visuospatial neglect after right hemisphere lesions. Limitations of the present study In line with previous findings (e.g. Perennou et al., 2008; Funk et al., 2010a,b, 2011), the present study indicates a close functional relationship between visuospatial neglect and a tilted, less precise perception of verticality in two modalities and two spatial planes. However, the present results do not necessarily imply a direct causal link between the syndrome of visuospatial neglect and the spatial orientation deficits reported in this and other studies. Given the tendency of larger lesions in neglect, the frequent co-occurrence of spatial neglect and orientation deficits

78

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79

may result from combined lesions affecting neighbouring but independent neural regions involved in spatial attention or spatial orientation respectively.

CONCLUSION The present findings support the notion of an impaired verticality perception in two modalities and two spatial planes in patients with left-sided neglect after right-brain damage. Systematic counterclockwise tilts were observed in neglect patients in roll and backward tilts in the pitch plane, both for the visual and haptic modality. These impairments clearly suggest a tilted and less accurate representation of verticality in 3-D-space most likely caused by lesions of multisensory brain areas in the temporal-parietal cortex. Acknowledgments—This work was supported by a Deutsche Forschungsgemeinschaft (DFG) grant to Georg Kerkhoff (IRTG 1457 “Adaptive minds”).

REFERENCES Andersen RA, Snyder LH, Bradley DC, Xing J (1997) Multimodal representation of space in the posterior parietal cortex and its use in planning movements. Annu Rev Neurosci 20:303–330. Bender M, Jung R (1948) Abweichungen der subjektiven optischen Vertikalen und Horizontalen bei Gesunden und Hirnverletzten. Eur Arch Psychiatry Clin Neurosci 181:193–212. Benton AL, Varney NR, Hamsher KD (1978) Visuospatial judgment. A clinical test. Arch Neurol 35:364 –367. Bonan IV, Hubeaux K, Gellez-Leman MC, Guichard JP, Vicaut E, Yelnik AP (2007) Influence of subjective visual vertical misperception on balance recovery after stroke. J Neurol Neurosurg Psychiatry 78:49 –55. Brandt T, Dieterich M (1999) The vestibular cortex. Its locations, functions, and disorders. Ann N Y Acad Sci 871:293–312. Brandt T, Dieterich M, Danek A (1994) Vestibular cortex lesions affect the perception of verticality. Ann Neurol 35:403– 412. Bronstein AM (1999) The interaction of otolith and proprioceptive information in the perception of verticality. The effects of labyrinthine and CNS disease. Ann N Y Acad Sci 871:324 –333. Bruell JH, Peszczynski M (1958) Perception of verticality in hemiplegic patients in relation to rehabilitation. Clin Orthop 12:124 –130. Bruell JH, Peszczynski M, Volk D (1957) Disturbance of perception of verticality in patients with hemiplegia: second report. Arch Phys Med Rehabil 38:776 –780. Dieterich M, Brandt T (1993) Thalamic infarctions: differential effects on vestibular function in the roll plane (35 patients). Neurology 43:1732–1740. Duhamel JR, Colby CL, Goldberg ME (1998) Ventral intraparietal area of the macaque: congruent visual and somatic response properties. J Neurophysiol 79:126 –136. Fels M, Geissner E (1996) Neglect-Test (NET). Ein Verfahren zur Erfassung visueller Neglectphänomene. Göttingen: Hogrefe. Friedmann G (1970) The judgement of the visual vertical and horizontal with peripheral and central vestibular lesions. Brain 93:313– 328. Frisen L (2010) Deviations of the visual upright in three dimensions in disorders of the brainstem: a clinical exploration. Brain 133: 3541–3551. Funk J, Finke K, Muller HJ, Preger R, Kerkhoff G (2010a) Systematic biases in the tactile perception of the subjective vertical in patients with unilateral neglect and the influence of upright vs. supine posture. Neuropsychologia 48:298 –308. Funk J, Finke K, Muller HJ, Utz KS, Kerkhoff G (2010b) Effects of lateral head inclination on multimodal spatial orientation judgments

in neglect: evidence for impaired spatial orientation constancy. Neuropsychologia 48:1616 –1627. Funk J, Finke K, Muller HJ, Utz KS, Kerkhoff G (2011) Visual context modulates the subjective vertical in neglect: evidence for an increased rod-and-frame-effect. Neuroscience 173:124 –134. Graziano MSA, Gross CG (1995) The representation of extrapersonal space: a possible role for bimodal, visual-tactile neurons. In: The cognitive neurosciences (Gazzaniga MS, ed), pp 1021–1034. Cambridge, MA: The MIT Press. Guldin WO, Grusser OJ (1998) Is there a vestibular cortex? Trends Neurosci 21:254 –259. Johannsen L, Fruhmann Berger M, Karnath HO (2006) Subjective visual vertical (SVV) determined in a representative sample of 15 patients with pusher syndrome. J Neurol 253:1367–1369. Karnath HO (2001) New insights into the functions of the superior temporal cortex. Nat Rev Neurosci 2:568 –576. Karnath HO, Fruhmann Berger M, Kuker W, Rorden C (2004) The anatomy of spatial neglect based on voxelwise statistical analysis: a study of 140 patients. Cereb Cortex 14:1164 –1172. Kerkhoff G (1998) Rehabilitation of visuospatial cognition and visual exploration in neglect: a cross-over study. Restor Neurol Neurosci 12:27– 40. Kerkhoff G (1999) Multimodal spatial orientation deficits in left-sided visual neglect. Neuropsychologia 37:1387–1405. Kerkhoff G, Marquardt C (2009) EYEMOVE. Standardized assessment and treatment of visual search disorders. Nervenarzt 80:1190 –1204. Kerkhoff G, Zoelch C (1998) Disorders of visuospatial orientation in the frontal plane in patients with visual neglect following right or left parietal lesions. Exp Brain Res 122:108 –120. Luyat M, Gentaz E (2002) Body tilt effect on the reproduction of orientations: studies on the visual oblique effect and subjective orientations. J Exp Psychol Hum Percept Perform 28:1002–1011. Mars F, Popov K, Vercher JL (2001) Supramodal effects of galvanic vestibular stimulation on the subjective vertical. Neuroreport 12:2991–2994. Mesulam MM (1999) Spatial attention and neglect: parietal, frontal and cingulate contributions to the mental representation and attentional targeting of salient extrapersonal events. Philos Trans R Soc Lond B Biol Sci 354:1325–1346. Mittelstaedt H (1999) The role of the otoliths in perception of the vertical and in path integration. Ann N Y Acad Sci 871:334 –344. Mort DJ, Malhotra P, Mannan SK, Rorden C, Pambakian A, Kennard C, Husain M (2003) The anatomy of visual neglect. Brain 126:1986 –1997. Oldfield RC (1971) The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9:97–113. Perennou D (2006) Postural disorders and spatial neglect in stroke patients: a strong association. Restor Neurol Neurosci 24:319 –334. Perennou DA, Mazibrada G, Chauvineau V, Greenwood R, Rothwell J, Gresty MA, Bronstein AM (2008) Lateropulsion, pushing and verticality perception in hemisphere stroke: a causal relationship. Brain 131:2401–2413. Qualls CE, Bliwise NG, Stringer AY (2000) Short forms of the Benton Judgment of Line Orientation Test: development and psychometric properties. Arch Clin Neuropsychol 15:159 –163. Saj A, Honore J, Bernati T, Coello Y, Rousseaux M (2005a) Subjective visual vertical in pitch and roll in right hemispheric stroke. Stroke 36:588 –591. Saj A, Honore J, Davroux J, Coello Y, Rousseaux M (2005b) Effect of posture on the perception of verticality in neglect patients. Stroke 36:2203–2205. Saj A, Honore J, Rousseaux M (2006) Perception of the vertical in patients with right hemispheric lesion: effect of galvanic vestibular stimulation. Neuropsychologia 44:1509 –1512. Sakata H, Taira M, Kusunoki M, Murata A, Tanaka Y (1997) The TINS lecture—the parietal association cortex in depth perception and visual control of hand action. Trends Neurosci 20:350 –357.

K. S. Utz et al. / Neuroscience 188 (2011) 68 –79 Schenkenberg T, Bradford DC, Ajax ET (1980) Line bisection and unilateral visual neglect in patients with neurologic impairment. Neurology 30:509 –517. Vallar G, Perani D (1986) The anatomy of unilateral neglect after right-hemisphere stroke lesions. A clinical/CT-scan correlation study in man. Neuropsychologia 24:609 – 622. Verdon V, Schwartz S, Lovblad KO, Hauert CA, Vuilleumier P (2010) Neuroanatomy of hemispatial neglect and its functional components: a study using voxel-based lesion-symptom mapping. Brain 133:880 – 894.

79

Von Cramon DY, Kerkhoff G (1993) On the cerebral organization of elementary visuo-spatial perception. In: Functional organization of the human visual cortex (Gulyás B, Ottoson D, Roland PE, eds), pp 211–231. Oxford: Pergamon Press. Wilson B, Cockburn J, Halligan P (1987) Development of a behavioral test of visuospatial neglect. Arch Phys Med Rehabil 68:98 –102. Yelnik AP, Lebreton FO, Bonan IV, Colle FM, Meurin FA, Guichard JP, Vicaut E (2002) Perception of verticality after recent cerebral hemispheric stroke. Stroke 33:2247–2253.

(Accepted 30 April 2011) (Available online 12 May 2011)